PA Hiss
mlcguy
GeezerSonics
"mlcguy" <vthor...@comcast.net> wrote:
> What causes PA hiss? Is there anyway to control it or reduce it?
>
It's typically caused by not paying attention to gain structure. If can
also be related to a noisy device(s) in your signal chain.
The first remedy, attenuate the input of your power amplifier(s) use the
gain available from your console and processing equipment.
Start by reading this document:
http://www.prosoundweb.com/studyhall/studyjump.php?pdf=basic_gain
Zigakly
9 times out of 10 it's the EQ. The DBX 2231 actually has a built-in noise
gate to de-emphasize its self-noise, a rather dubious feature. Just the
other day I was recommending digital EQ's for budget systems for the sake of
lower noise, like the Alesis DEQ230. With cheap analog EQ's it helps to run
a hotter signal into the EQ and turn the EQ output level down.
John Halliburton
"Zigakly" <zig...@nospam.cx> wrote in message
news:Is_5e.609$K24...@read1.cgocable.net...
>> What causes PA hiss? Is there anyway to control it or reduce it?
>
> 9 times out of 10 it's the EQ.
No, 90% of the time it's a gain structure issue. If the gear is in good
repair(to factory specs) most will be quiet enough for live sound work when
the gain structure is good.
JHH
Phil Allison
"mlcguy"
>
> What causes PA hiss? Is there anyway to control it or reduce it?
** The snake is hungry.
Feed him.
............ Phil
Kevin Aylward
> In article <1113084880.0...@g14g2000cwa.googlegroups.com>,
> "mlcguy" <vthor...@comcast.net> wrote:
>
>> What causes PA hiss?
Thermal noise and current shot noise due to the random motion of
elementary particles.
> Is there anyway to control it or reduce it?
>>
>
> It's typically caused by not paying attention to gain structure.
Yes. Although the equipment may just be crap.
>If
> can also be related to a noisy device(s) in your signal chain.
Yes.
>
> The first remedy, attenuate the input of your power amplifier(s) use
> the gain available from your console and processing equipment.
Yes, although I doubt if many amps nowadays generate relevant hiss. This
can be more useful for mitigating the effect of ground induced hum.
>
> Start by reading this document:
>
> http://www.prosoundweb.com/studyhall/studyjump.php?pdf=basic_gain
I had a quick look, and I don't see this as very useful at all for
average musicians. It is requesting voltmeters and oscilloscopes!!!
For a typical *PA*, the main offender for noise will be the mic channels
of the main mixer. The procedure is simple for pretty much any non
totally basic desk.
1 Set the tone controls to how one wants them, noting that too much
treble will increase hiss.
2) Shout into the mic and let the peak level LED just briefly flicker by
adjusting the mic *gain* control, not the channel fader.
3) Turn off all unused mic channels or dont route them to the main
output, or turn their gain fully down.
Kevin Aylward
informati...@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
GeezerSonics
"Kevin Aylward" <see_w...@anasoft.co.uk> wrote:
> > The first remedy, attenuate the input of your power amplifier(s) use
> > the gain available from your console and processing equipment.
>
> Yes, although I doubt if many amps nowadays generate relevant hiss. This
> can be more useful for mitigating the effect of ground induced hum.
I recommended attenuating the input stage of the amplifier, because most
amplifiers have sensitivities that are out of wack, given the signal
levels available from most pro audio devices. The purpose of attenuation
at this point, is to increase signal to noise by enabling up chain
devices to work optimally, amplifier hiss is not the issue.
> > Start by reading this document:
> >
> > http://www.prosoundweb.com/studyhall/studyjump.php?pdf=basic_gain
>
> I had a quick look, and I don't see this as very useful at all for
> average musicians. It is requesting voltmeters and oscilloscopes!!!
I should have provided this link:
http://www.prosoundweb.com/studyhall/studyjump.php?pdf=gain
It provides a good representation of gain structure, graphically and
mathematically.
> For a typical *PA*, the main offender for noise will be the mic channels
> of the main mixer. The procedure is simple for pretty much any non
> totally basic desk.
> 1 Set the tone controls to how one wants them, noting that too much
> treble will increase hiss.
> 2) Shout into the mic and let the peak level LED just briefly flicker by
> adjusting the mic *gain* control, not the channel fader.
>
> 3) Turn off all unused mic channels or dont route them to the main
> output, or turn their gain fully down.
This is not bad advise, in general, execpt that I wouldn't adjust
equalization ahead of the preamp gain. But it does not provide
direction for the OP's original complaint.
Kevin Aylward
> In article <B656e.11388$mV1....@fe3.news.blueyonder.co.uk>,
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote:
>
>>> The first remedy, attenuate the input of your power amplifier(s) use
>>> the gain available from your console and processing equipment.
>>
>> Yes, although I doubt if many amps nowadays generate relevant hiss.
>> This can be more useful for mitigating the effect of ground induced
>> hum.
>
> I recommended attenuating the input stage of the amplifier, because
> most amplifiers have sensitivities that are out of wack, given the
> signal levels available from most pro audio devices.
I can't agree that this is really an accurate description of typical
pro-audio systems.
Power amps and mixers are usually all *calibrated* to +4db, or Odb, and
this is the preferred default. There is even often a link inside gear to
switch between them.
>The purpose of
> attenuation at this point, is to increase signal to noise by enabling
> up chain devices to work optimally, amplifier hiss is not the issue.
>
>>> Start by reading this document:
>>>
>>> http://www.prosoundweb.com/studyhall/studyjump.php?pdf=basic_gain
>>
>> I had a quick look, and I don't see this as very useful at all for
>> average musicians. It is requesting voltmeters and oscilloscopes!!!
>
> I should have provided this link:
>
> http://www.prosoundweb.com/studyhall/studyjump.php?pdf=gain
>
> It provides a good representation of gain structure, graphically and
> mathematically.
It still uses the scope and pink noise generator, so is essential
useless advice for any typical musician.
I only skimmed, but some of the arguments are wrong. For example:
"A reasonable choice is an amplifier whose rating that is 2 times (+3
dB) the RMS rating of the loudspeaker. The RMS sine wave used to rate
amplifiers has an inherent peak power component of 3 dB. So this all
works out to a 6 dB allowance for power peaks over the loudspeaker's RMS
rating."
This is mixing apples with oranges. The "peak" power associated with
sine wave average power(typically referred to, incorrectly, as rms
power) is *not* the *same* "peak" that is referred to in loudspeaker
ratings. The peak of a sine wave is the *instantaneous* value. This
value is repeated twice per cycle. e.g. at 1khz, every 500us. The "peak"
rating of a loudspeaker is the *average* (i.e. the so called rms value)
that can be sustained over *many* cycles. This can be larger than its
steady state rating. The idea is that it is thermal heating that will
burn out the voice coil. If say, a 1Kkz signal is pulsed on for say,
10ms (several cycles) then off for 10ms, its "average average" (rms)
power will only be 1/2 of its "peak average" power. That is the "peak
power" is the average power during the on 10 cycles. It is not the
instantaneous peak of the power waveform.
Typically, the "peak average power" to "average average power" is more
than 2 for typical audio, hence why, it may be ok to use a higher rated
amplifier. However, one must note that some sine type signals might blow
the cone right off its end stops!
So, a basic misunderstanding such as this in a paper, indicates that the
paper is suspicious. I just don't have the time to evaluate the rest at
it at the moment. What I will say is that, overall, it appears way too
involved to be of much use to typical musicians.
>
>> For a typical *PA*, the main offender for noise will be the mic
>> channels of the main mixer. The procedure is simple for pretty much
>> any non totally basic desk.
>
>> 1 Set the tone controls to how one wants them, noting that too much
>> treble will increase hiss.
>> 2) Shout into the mic and let the peak level LED just briefly
>> flicker by adjusting the mic *gain* control, not the channel fader.
>>
>> 3) Turn off all unused mic channels or dont route them to the main
>> output, or turn their gain fully down.
>
>
> This is not bad advise, in general, execpt that I wouldn't adjust
> equalization ahead of the preamp gain.
Oh?
The peak detector is pretty much always *after* the EQ. If you EQ after
setting the gain, HF might still clip, as EQ might well add up to 20db
of gain. So, yes, you want to set the EQ first, then the input gain.
> But it does not provide
> direction for the OP's original complaint.
Oh?
The poster is complaining of PA hiss. I identified what the main
offender usually is, and how to correct it.
Kevin Aylward - Ex-Studiomaster design engineer - Designer of the MOSFET
1000:-)
Phil Allison
"Kevin Aylward"
> GeezerSonics wrote:
>>
>> I recommended attenuating the input stage of the amplifier, because
>> most amplifiers have sensitivities that are out of wack, given the
>> signal levels available from most pro audio devices.
>
> I can't agree that this is really an accurate description of typical
> pro-audio systems.
** Possibly because you entirely missed the point.
> Power amps and mixers are usually all *calibrated* to +4db, or Odb, and
> this is the preferred default.
** That remark simply has no meaning - it is totally circular.
An unattenuated power amp will clip just beyond 0 dB or +4 dB - while
mixers and ancillaries have a lot of headroom left to go when producing
those same audio levels. The system may well be better for it AND the
sound operator much happier if the amps do not clip the signal when it
reaches the mixer's *nominal *operating level.
> The poster is complaining of PA hiss. I identified what the main offender
> usually is, and how to correct it.
** That is a highly doubtful claim.
1. If mic pre-amp hiss is predominates, then the PA system is probably well
set up gain wise.
2. If, at the same time as #1 is the case, mic pre-amp hiss is audible and
*annoying* - then just why the system is not *howling* with acoustic
feedback needs explaining.
> Kevin Aylward - Ex-Studiomaster design engineer - Designer of the MOSFET
> 1000:-)
** An off-key trumpet solo if I ever heard one.
............ Phil
noisepolice
> (snipped)
> ** An off-key trumpet solo if I ever heard one.
> ............ Phil
OMIGOD !!! ... NO !!! ... NO !!! ... AAAARRRRGGGGHHHH !!!!!! ...
Phil, please don't upset these crazy people. Look what happened last
time ...
http://groups-beta.google.com/group/alt.audio.pro.live-sound/msg/d5dae0ffbdeee0b7
"we are all to blame"
*^_^*
Pooh Bear
I have to agree with Phil on this one Kevin.
If the amps are going to clip with a 0dBu or +4dBu input, what's the point of
having +22 or +28dBu capable outputs on the desk ?
Running high level to the outboard also reduces any pickup issues there too.
Would you *really* want to run a mix where the system clips when the output
meters are half way up the scale ? I like to see the meters way into the
yellow. :-)
Graham Stevenon - current Studiomaster design engineer - indeed for some 15
yrs or so ( and also predating Kevin too - lol ) . Designer of so much I
couldn't give you the whole list. ;-)
Kevin Aylward
> "Kevin Aylward"
>> GeezerSonics wrote:
>>>
>>> I recommended attenuating the input stage of the amplifier, because
>>> most amplifiers have sensitivities that are out of wack, given the
>>> signal levels available from most pro audio devices.
>>
>> I can't agree that this is really an accurate description of typical
>> pro-audio systems.
>
>
> ** Possibly because you entirely missed the point.
You mean the one about:
"What causes PA hiss? Is there anyway to control it or reduce it?"
>
>> Power amps and mixers are usually all *calibrated* to +4db, or Odb,
>> and this is the preferred default.
>
>
> ** That remark simply has no meaning - it is totally circular.
Pardon? I made a statement of *facts*. The notion of "circular" I would
note, as in circular argument, where one is attempting to derive a
result form its own result. No such argument was made.
>
> An unattenuated power amp will clip just beyond 0 dB or +4 dB -
Yes.
> while mixers and ancillaries have a lot of headroom left to go when
> producing those same audio levels.
They may do.
>The system may well be better for
> it AND the sound operator much happier if the amps do not clip the
> signal when it reaches the mixer's *nominal *operating level.
Sure, but this is extremely unlikely to have any relevant effect on PA
hiss, which is what we are discussing.
What part of:
"What causes PA hiss? Is there anyway to control it or reduce it?" did
you have trouble understanding?
PA amps essentially, make no hiss. Its hard to even know if they are
switched on. So, sure, it can make some sense to back the gain off on
the power amp, but this is simple irrelevant to PA *hiss*.
I even stated where it may help, e.g. minimising the effect of hum,
which of course, was not asked for by the poster.
>
>
>> The poster is complaining of PA hiss. I identified what the main
>> offender usually is, and how to correct it.
>
>
> ** That is a highly doubtful claim.
Oh?
>
> 1. If mic pre-amp hiss is predominates, then the PA system is
> probably well set up gain wise.
The usually case for fixable hiss is simply the case that the operators
doesn't know how to set things up. Modern equipment is usually
reasonably noise free, of course there are exceptions.
Either the gain has been set up correctly or it has not. If the gain is
not setup correctly, it may well hiss. If the gain is setup correctly,
it will not hiss. Exception, the mixer is of poor design.
>
> 2. If, at the same time as #1 is the case, mic pre-amp hiss is
> audible and *annoying* - then just why the system is not *howling*
> with acoustic feedback needs explaining.
Low gain at front end, large gain at output end = High gain at front
end, low gain at output end.
One is more nosier than the other, but have the same gain. Dah...
Mic far from the speakers
Low output mic
Mic pad on
Poor mixer design
"Annoying" is relative.
>
>
>
>> Kevin Aylward - Ex-Studiomaster design engineer - Designer of the
>> MOSFET 1000:-)
>
>
> ** An off-key trumpet solo if I ever heard one.
But clearly more in tune then your own.
Kevin Aylward
Kevin Aylward
But you obviously let the thread *question* slip by you:-)
>
> If the amps are going to clip with a 0dBu or +4dBu input, what's the
> point of having +22 or +28dBu capable outputs on the desk ?
>
> Running high level to the outboard also reduces any pickup issues
> there too.
>
> Would you *really* want to run a mix where the system clips when the
> output meters are half way up the scale ? I like to see the meters
> way into the yellow. :-)
None of this is in dispute requiring general noise, as in interference
and hum. I stated as much. This is all, obviously, all obvious. I often
even back of the gain on the PA amps myself at gigs. None of it has any
*practical* relevence to *hiss* though. I was addressing the posters
question:
"What causes PA hiss? Is there anyway to control it or reduce it?"
The first part of my answer was:
"Thermal noise and current shot noise due to the random motion of
elementary particles."
The second part of my answer address this white noise "hiss".
>
>
> Graham Stevenon - current Studiomaster design engineer - indeed for
> some 15 yrs or so ( and also predating Kevin too - lol ) . Designer
> of so much I couldn't give you the whole list. ;-)
Hi Graham, nice to hear from you.
Have you tried my software yet?
Kevin Aylward
GeezerSonics
"Kevin Aylward" <see_w...@anasoft.co.uk> wrote:
> GeezerSonics wrote:
> > In article <B656e.11388$mV1....@fe3.news.blueyonder.co.uk>,
> > "Kevin Aylward" <see_w...@anasoft.co.uk> wrote:
> >
> >>> The first remedy, attenuate the input of your power amplifier(s) use
> >>> the gain available from your console and processing equipment.
> >>
> >> Yes, although I doubt if many amps nowadays generate relevant hiss.
> >> This can be more useful for mitigating the effect of ground induced
> >> hum.
> >
> > I recommended attenuating the input stage of the amplifier, because
> > most amplifiers have sensitivities that are out of wack, given the
> > signal levels available from most pro audio devices.
>
> I can't agree that this is really an accurate description of typical
> pro-audio systems.
>
> Power amps and mixers are usually all *calibrated* to +4db, or Odb, and
> this is the preferred default. There is even often a link inside gear to
> switch between them.
Yes, most pro audio gear provides a nominal output level of +4dB, and
and maximums of +18 to +24 before clipping.
Most amplifiers provide MAXIMUM output with +4dB input levels. That's
20dB> sensitivity then needed. That's why many people recommend thaty
attenuate the amplifier's input. This allows the entire signal chain to
run at optimum, and allows the console and amplifier to have the same
clip point.
> So, a basic misunderstanding such as this in a paper, indicates that the
> paper is suspicious. I just don't have the time to evaluate the rest at
> it at the moment. What I will say is that, overall, it appears way too
> involved to be of much use to typical musicians.
>
> >
> >> For a typical *PA*, the main offender for noise will be the mic
> >> channels of the main mixer. The procedure is simple for pretty much
> >> any non totally basic desk.
> >
> >> 1 Set the tone controls to how one wants them, noting that too much
> >> treble will increase hiss.
> >> 2) Shout into the mic and let the peak level LED just briefly
> >> flicker by adjusting the mic *gain* control, not the channel fader.
> >>
> >> 3) Turn off all unused mic channels or dont route them to the main
> >> output, or turn their gain fully down.
> >
> >
> > This is not bad advise, in general, execpt that I wouldn't adjust
> > equalization ahead of the preamp gain.
>
> Oh?
>
> The peak detector is pretty much always *after* the EQ. If you EQ after
> setting the gain, HF might still clip, as EQ might well add up to 20db
> of gain. So, yes, you want to set the EQ first, then the input gain.
Yes, that's kind of friendly reminder to judiciously use eq. I still
recommend setting gain with eq's flat, then readjusting if necessary.
> > But it does not provide
> > direction for the OP's original complaint.
>
> Oh?
>
> The poster is complaining of PA hiss. I identified what the main
> offender usually is, and how to correct it.
No, you didn't. You ignored the 20dB encroachment of amplfier
sensitivity on S/N ratio.
Kevin Aylward
Sure, but irrelevent to hiss.
>>
>> Oh?
>>
>> The peak detector is pretty much always *after* the EQ. If you EQ
>> after setting the gain, HF might still clip, as EQ might well add up
>> to 20db of gain. So, yes, you want to set the EQ first, then the
>> input gain.
>
>
> Yes, that's kind of friendly reminder to judiciously use eq. I still
> recommend setting gain with eq's flat, then readjusting if necessary.
Why?
Since you agree that doing so may well mean that the input will clip.
What advantage do you claim for setting EQ after the gain adjustment?
>
>
>
>
>>> But it does not provide
>>> direction for the OP's original complaint.
>>
>> Oh?
>>
>> The poster is complaining of PA hiss. I identified what the main
>> offender usually is, and how to correct it.
>
>
> No, you didn't.
Yes I did.
> You ignored the
Yes I did.
>20dB encroachment of amplfier
> sensitivity on S/N ratio.
It doesn't effect the s/n ratio, as far as hiss goes of the system at
all. The fact that the gain at the amp may be reduced by 20db, does not
mean that the s/n is so effected.
The noise level of most PA amps is so low as to be completely irrelevant
in computing final white noise s/n ratios. Hint: go and stick your ear
up to the speakers when the power amp is on with no input. You will be
hard pressed to hear any hiss at all. Since this is the *maximum* noise
it can make, if you cant hear the hiss with no signal, you wil certainly
not hear it with a signal.
Kevin Aylward
Phil Allison
> Phil Allison wrote:
>>> GeezerSonics
>>
>>>> I recommended attenuating the input stage of the amplifier, because
>>>> most amplifiers have sensitivities that are out of wack, given the
>>>> signal levels available from most pro audio devices.
>>>
>>> I can't agree that this is really an accurate description of typical
>>> pro-audio systems.
>>
>>
>> ** Possibly because you entirely missed the point.
>
> You mean the one about:
>
> "What causes PA hiss? Is there anyway to control it or reduce it?"
** The point made right above your asinine comment is the one - shithead.
>>> Power amps and mixers are usually all *calibrated* to +4db, or Odb,
>>> and this is the preferred default.
>>
>>
>> ** That remark simply has no meaning - it is totally circular.
>
> Pardon? I made a statement of *facts*.
** Only "facts* in your mad fantasy world.
> The notion of "circular" I would note, as in circular argument, where one
> is attempting to derive a result form its own result. No such argument was
> made.
** The arguement is circular all right - it begs the question.
>> An unattenuated power amp will clip just beyond 0 dB or +4 dB -
>
> Yes.
>
>> while mixers and ancillaries have a lot of headroom left to go when
>> producing those same audio levels.
>
> They may do.
** Like the sun may rise in the morning..........
>>The system may well be better for
>> it AND the sound operator much happier if the amps do not clip the
>> signal when it reaches the mixer's *nominal *operating level.
>
> Sure, but this is extremely unlikely to have any relevant effect on PA
> hiss, which is what we are discussing.
** On the contrary - it is very likely to reduce it in the same ratio.
> What part of:
>
> "What causes PA hiss? Is there anyway to control it or reduce it?" did you
> have trouble understanding?
** None. Which part of "Kevin Ayward is a Smug, Posturing Pommy Cunt" do
you have any trouble with ???
> PA amps essentially, make no hiss. Its hard to even know if they are
> switched on. So, sure, it can make some sense to back the gain off on the
> power amp, but this is simple irrelevant to PA *hiss*.
** Backing off the input attenuators on the power amps will allow the mixer
and ancillary gear to operate at a higher nominal level. The s/n will
therefore improve.
>>> The poster is complaining of PA hiss. I identified what the main
>>> offender usually is, and how to correct it.
>>
>>
>> ** That is a highly doubtful claim.
>
> Oh?
** Which part of "Kevin Ayward is a Smug, Posturing Pommy Cunt" do you have
any trouble with ???
>> 1. If mic pre-amp hiss is predominates, then the PA system is
>> probably well set up gain wise.
>>
>> 2. If, at the same time as #1 is the case, mic pre-amp hiss is
>> audible and *annoying* - then just why the system is not *howling*
>> with acoustic feedback needs explaining.
>
> Low gain at front end, large gain at output end = High gain at front end,
> low gain at output end.
>
> One is more nosier than the other, but have the same gain. Dah...
>
** Read point #1 again - you raving, fucking nut case.
> Mic far from the speakers
> Low output mic
> Mic pad on
> Poor mixer design
** Are these lines in some new song ???
Can you toot the tune on that horn of yours for us ??
> "Annoying" is relative.
** Relatives are often annoying.
OTOH - Manic Pommy Cunts like Kevin Aylward are always so.
>>> Kevin Aylward - Ex-Studiomaster design engineer - Designer of the
>>> MOSFET 1000:-)
>>
>>
>> ** An off-key trumpet solo if I ever heard one.
>
> But clearly more in tune then your own.
** Only seems that way to the maniac blowing his a tin trumpet.
Get help Kev.
............. Phil
Sean Conolly
news:h1u6e.21023$mV1....@fe3.news.blueyonder.co.uk...
> The noise level of most PA amps is so low as to be completely irrelevant
> in computing final white noise s/n ratios. Hint: go and stick your ear
> up to the speakers when the power amp is on with no input. You will be
> hard pressed to hear any hiss at all. Since this is the *maximum* noise
> it can make, if you cant hear the hiss with no signal, you wil certainly
> not hear it with a signal.
I'm looking forward to this when I get around to replacing my old QSC 1700.
The hiss is annoying, but when funds are tight it's hard to justify
replacing an amp that has never given me 1 second of problems in 14 years.
Sean
Kevin Aylward
> "Kevin Aylward"
>> Phil Allison wrote:
>>>> GeezerSonics
>>>
>>>>> I recommended attenuating the input stage of the amplifier,
>>>>> because most amplifiers have sensitivities that are out of wack,
>>>>> given the signal levels available from most pro audio devices.
>>>>
>>>> I can't agree that this is really an accurate description of
>>>> typical pro-audio systems.
>>>
>>>
>>> ** Possibly because you entirely missed the point.
>>
>> You mean the one about:
>>
>> "What causes PA hiss? Is there anyway to control it or reduce it?"
>
>
> ** The point made right above your asinine comment is the one -
> shithead.
Here we go, if all else fails, start the personal insults.
No one every told you that approach never works?
>
>
>
>>>> Power amps and mixers are usually all *calibrated* to +4db, or Odb,
>>>> and this is the preferred default.
>>>
>>>
>>> ** That remark simply has no meaning - it is totally circular.
>>
>> Pardon? I made a statement of *facts*.
>
>
> ** Only "facts* in your mad fantasy world.
You mean you disagree that pro mixers and amp are usually calibrated to
+4db, or Odb?
Care to support your allegation?
>
>
>> The notion of "circular" I would note, as in circular argument,
>> where one is attempting to derive a result form its own result. No
>> such argument was made.
>
>
> ** The arguement is circular all right - it begs the question.
And just what argument was that?
>
>
>>> The system may well be better for
>>> it AND the sound operator much happier if the amps do not clip the
>>> signal when it reaches the mixer's *nominal *operating level.
>>
>> Sure, but this is extremely unlikely to have any relevant effect on
>> PA hiss, which is what we are discussing.
>
>
> ** On the contrary
Oh?
> - it is very likely to reduce it in the same
> ratio.
There is no way what you claim here is even remotly correct. The
effective input noise is determined by the statistical sum of each stage
referred to the input. To refer the noise to the input, you divide by
the gain. A typical power amp may have, say 10nV/rthz noise, probably
much less. With a front end gain of 40db, it would amount to a mere
0.1nv/rthz at the mic input. A mic input may be say, 1nv/rthz. This will
result in a s/n reduction of 0.04db, as the noise adds on a square
basis.
>
>
>> What part of:
>>
>> "What causes PA hiss? Is there anyway to control it or reduce it?"
>> did you have trouble understanding?
>
>
> ** None. Which part of "Kevin Ayward is a Smug, Posturing Pommy
> Cunt" do you have any trouble with ???
Oh dear, more arguments in support of your views?
>
>
>> PA amps essentially, make no hiss. Its hard to even know if they are
>> switched on. So, sure, it can make some sense to back the gain off
>> on the power amp, but this is simple irrelevant to PA *hiss*.
>
>
> ** Backing off the input attenuators on the power amps will allow
> the mixer and ancillary gear to operate at a higher nominal level.
Yep.
> The s/n will therefore improve.
Not necessarily. It depends on the noise of the following stages and how
much gain is ahead of them.
White noise hiss is usually completely dominated by the input stage. The
input noise of the PA amp is divided by the gain. Since you might have
trouble with the math I presented above, I will paste what I wrote in
another post in this thread:
Hint: go and stick your ear up to the speakers when the power amp is on
with no input. You will be
hard pressed to hear any hiss at all. Since this is the *maximum* noise
it can make, if you cant hear the hiss with no signal, you will
certainly not hear it with a signal.
{snip rest of insults superfluous to the arguments}
Phil Allison
"Kevin Aylward"
** Kev - you have totally lost it.
............ Phil
Kevin Aylward
> "Kevin Aylward"
>
>
> ** Kev - you have totally lost it.
>
I take this as code for "I cannot refute any of Kevin's arguments"
Gene Sweeny
Kevin Aylward wrote:
>
> It doesn't effect the s/n ratio, as far as hiss goes of the system at
> all. The fact that the gain at the amp may be reduced by 20db, does
not
> mean that the s/n is so effected.
>
> The noise level of most PA amps is so low as to be completely
irrelevant
> in computing final white noise s/n ratios. Hint: go and stick your
ear
> up to the speakers when the power amp is on with no input. You will
be
> hard pressed to hear any hiss at all. Since this is the *maximum*
noise
> it can make, if you cant hear the hiss with no signal, you wil
certainly
> not hear it with a signal.
>
Kevin,
You are so utterly wrong in most of your statements, I don't know
really where to start. You should to read the the information on
prosoundweb.com in the Study Hall section on gain staging and all that
jazz.
Basically what it boils down to is if one peice of gear provides more
sensitvity than another, and is further down the chain, it will amplify
the noise floor of the previous stages. Because amplifiers are fixed
gain devices, no matter what the input level is set to, the amplifier
will clip at the same output voltage level. By attenuating the
amplifiers, you allow the gear that is earlier in signal chain to use
more of each devives signal to noise ratio.
A simple example. Say I have a pc of gear (call it "the source") that
clips at +28dbu and has 90db of signal to noise ratio. Then I connect
it to an amplifier that clips at +4dbu. I don't attenuate the amps,
and that effectively limits the signal to noise ratio to 66 db.
Basically, when the when the source hits +4dbu on it's meters it will
clip the amp. A signal that is +4dbu at the source is only 66db above
the noise floor. If I instead adjusted the amp to -20dbu (24 db of att
total) input, the amp would clip at the same point that the source
clipped at +28dbu.
These numbers may seem acceptable (66db of s/n ratio isn't awful for
live applications) however, the problem becomes more complex as more
peices of gear are added in the signal chain.
The articles on PSW.com can do a much better job of describing all
this. But I'll agree with others in saying that gain structure is the
number one source of PA hiss IMO.
Zigakly
"John Halliburton" <j_chall...@ameritech.net> wrote in message
news:VH%5e.22413$Y76....@newssvr31.news.prodigy.com...
I guess I should have said 90% of the time I encounter hiss it's from the
EQ, and the fix you sniped from my post is in essence a misuse of gain
structure, so I still disagree, but YMMV.
Phil Allison
"Kevin Aylward"
> Phil Allison wrote:
>> "Kevin Aylward"
>>
>>
>> ** Kev - you have totally lost it.
>>
>
> I take this as code for "I cannot refute any of Kevin's arguments"
** The alleged arguments are not reality based.
Nor is Kev.
............... Phil
Mike Dobony
"mlcguy" <vthor...@comcast.net> wrote in message
news:1113084880.0...@g14g2000cwa.googlegroups.com...
> What causes PA hiss? Is there anyway to control it or reduce it?
>
Quite a long list of things can cause hiss. Here is a minute list.
Bad mic cables
bad connectors
running cables next to power lines
running cables along steel beams
poor gain structure
defective mics
defective hardware
Reversed phase outlets (hot and neutral reversed)
using unbalanced cables to connect equipment
equipment plugged in to mixed phases, especially with cheap equipment or
blown filters
bad speakers, specifically the crossover
neon or florescent lights, especially with a bad ballast
any faulty equipment on the same circuit, not necessarily your sound gear
and the list goes on and on and on
Best way to isolate the problem is to turn EVERYTHING off except the sound
gear to eliminate outside interference. Then disconnect one piece of gear
at a time until you find the problem gear. Then fix or replace.
--
Mike D.
Remove .spamnot to respond by email
Phil Allison
"Mike Dobony"
> "mlcguy"
>>
>> What causes PA hiss? Is there anyway to control it or reduce it?
>>
>
> Quite a long list of things can cause hiss. Here is a minute list.
>
> Bad mic cables
> bad connectors
> running cables next to power lines
> running cables along steel beams
> poor gain structure
> defective mics
> defective hardware
> Reversed phase outlets (hot and neutral reversed)
> using unbalanced cables to connect equipment
> equipment plugged in to mixed phases, especially with cheap equipment or
> blown filters
> bad speakers, specifically the crossover
> neon or florescent lights, especially with a bad ballast
> any faulty equipment on the same circuit, not necessarily your sound gear
>
> and the list goes on and on and on
>
> Best way to isolate the problem is to turn EVERYTHING off except the sound
> gear to eliminate outside interference. Then disconnect one piece of gear
> at a time until you find the problem gear. Then fix or replace.
>
> --
> Mike D.
** Hisssssssssssssssssssssssssssssssssssssssssssssssss.......
is not hummmmmmmmmmmmmmmmmmmmmmm ........
ROTFL !!
.......... Phil
Pooh Bear
Mike Dobony wrote:
> "mlcguy" <vthor...@comcast.net> wrote in message
> news:1113084880.0...@g14g2000cwa.googlegroups.com...
> > What causes PA hiss? Is there anyway to control it or reduce it?
> >
>
> Quite a long list of things can cause hiss. Here is a minute list.
I prefer an hourly list - but I won't quibble. :-)
> Bad mic cables
Surely they're the ones that actually *reduce* hiss ( and 'noise' - best
description I can think of for *metal* 'singers' ) when they go open circuit
?
Graham ;-)
Pooh Bear
Or even BbbbZZZzzzzzzzzzzzzzzzzzz ? CLICK !!!!! Poppp - pip - pop. MMMmmzzzz <
crackle >
Graham ;-)
Kevin Aylward
> Kevin Aylward wrote:
>>
>> It doesn't effect the s/n ratio, as far as hiss goes of the system at
>
>> all. The fact that the gain at the amp may be reduced by 20db, does
>> not mean that the s/n is so effected.
>>
>> The noise level of most PA amps is so low as to be completely
>> irrelevant in computing final white noise s/n ratios. Hint: go and
>> stick your
> ear
>> up to the speakers when the power amp is on with no input. You will
> be
>> hard pressed to hear any hiss at all. Since this is the *maximum*
>> noise it can make, if you cant hear the hiss with no signal, you wil
>> certainly not hear it with a signal.
>>
>
> Kevin,
>
> You are so utterly wrong in most of your statements,
Oh...
>I don't know
> really where to start.
I know.
>You should to read the the information on
> prosoundweb.com in the Study Hall section on gain staging and all that
> jazz.
Oh...
>
> Basically what it boils down to is if one peice of gear provides more
> sensitvity than another,
Do you mean more gain?
>and is further down the chain, it will
> amplify the noise floor of the previous stages.
Yep, it will amplifier the noise. It will also amplifier the signal.
>Because amplifiers
> are fixed gain devices, no matter what the input level is set to, the
> amplifier will clip at the same output voltage level. By attenuating
> the amplifiers, you allow the gear that is earlier in signal chain to
> use more of each devives signal to noise ratio.
Yes, in *principle*, but as I explained in onother post with a numerical
example, this has no *practical* relevance to the power amp part of the
chain as far as white noise is concerned.
One has to do the actual numbers to see if this principle makes sense in
a particular situation.
*****
The
effective input noise is determined by the statistical sum of each stage
referred to the input. To refer the noise to the input, you divide by
the gain. A typical power amp may have, say 10nV/rthz noise, probably
much less. With a front end gain of 40db, it would amount to a mere
0.1nv/rthz at the mic input. A mic input may be say, 1nv/rthz. This will
result in a s/n reduction of 0.04db, as the noise adds on a square
basis.
********
The *reason* for backing of the gain of later stages, and increasing
earlier stage gain, is so that the noise of later stages don't add to
the total noise. If the noise of the later stages is so low anyway, then
it makes no difference.
>
> A simple example. Say I have a pc of gear (call it "the source") that
> clips at +28dbu and has 90db of signal to noise ratio. Then I connect
> it to an amplifier that clips at +4dbu. I don't attenuate the amps,
> and that effectively limits the signal to noise ratio to 66 db.
Not necessarily. One can simply reduce the gain/output of the source. A
reduction in output level of the source does not imply the that signal
to noise will get worse by the same level. It depends where the dominant
noise sourse is.
What is assumed in your argument is that one can't simply back of the
gain of the source without unduly effecting its S/N. This is not usually
the case.
> Basically, when the when the source hits +4dbu on it's meters it will
> clip the amp. A signal that is +4dbu at the source is only 66db above
> the noise floor. If I instead adjusted the amp to -20dbu (24 db of
> att total) input, the amp would clip at the same point that the source
> clipped at +28dbu.
Now suppose that the pot on the amp is physically located in the source.
What changes?
I suspect that you have misread what I am actually saying.
>
> These numbers may seem acceptable (66db of s/n ratio isn't awful for
> live applications) however, the problem becomes more complex as more
> peices of gear are added in the signal chain.
This isn't a good example. A source that outputs +28db at 90db s/n and
can not be adjusted to a lower level without destroying its s/n is a
poor source. Can you give me an example of such a poor design?
For example, the Mixer. This often has an output amp with its gain
controlled by a pot in its feedback loop. The inherent noise of this
stage is usually insignificant compared with the noise generated from
earlier stages. This means that the output level can be controlled
without effecting the s/n.
>
> The articles on PSW.com can do a much better job of describing all
> this. But I'll agree with others in saying that gain structure is the
> number one source of PA hiss IMO.
With all due respect Gene, it seems you have been reading to many
semi-technical accounts. You need to look at the details. One doesn't
just get sources with a fixed +28db at 90db s/n. They have pots, which
you seem to be ignoring. Adjusting these usually gets one, essentially,
the same s/n with a lower output. Whether the pot is physically at the
amp or at the source is irrelevant.
Kevin Aylward
>> What causes PA hiss? Is there anyway to control it or reduce it?
>
Very unlikely. The EQ is usually after the initial gain stage. In which
case the input referred noise of the EQ will be divided by the gain of
that stage, say 40db.
If the gain structure is correct, noise should always be dominated by
the first mic stage, which typically has an input noise of around
0.8nv-2nv/rthz. Since many opamps are in the 10nv-20nv and less range,
any front gain of more than about 20 should make the op amp noise of EQ
stages minimal.
Its pretty obvious really, EQ and later stages can be run at 1V levels,
while the mic amp is at the mv levels.
Of course, EQing will make the noise of the front end larger if it has a
lot of boost.
Pooh Bear
Kevin Aylward wrote:
> Zigakly wrote:
> >> What causes PA hiss? Is there anyway to control it or reduce it?
> >
> > 9 times out of 10 it's the EQ.
>
> Very unlikely. The EQ is usually after the initial gain stage. In which
> case the input referred noise of the EQ will be divided by the gain of
> that stage, say 40db.
That's dB actually ! Go on - call me a pedant !
You've never heard a noisy EQ section ? I assure you that they exist.
> If the gain structure is correct,
Correct ? Is that the due to the design of the mixer or a result of its
(mis)use ?
> noise should always be dominated by
> the first mic stage, which typically has an input noise of around
> 0.8nv-2nv/rthz. Since many opamps are in the 10nv-20nv and less range,
> any front gain of more than about 20 should make the op amp noise of EQ
> stages minimal.
Hmmmmm. And I *don't* mean 'hum' - lol.
In an ideal world maybe ?
> Its pretty obvious really, EQ and later stages can be run at 1V levels,
If the user drives them so ! Actually I like to drive them at higher volt
levels ! Drive them at a hundred mV or so and the EQ noise contribution will
be significant.
> while the mic amp is at the mv levels.
>
> Of course, EQing will make the noise of the front end larger if it has a
> lot of boost.
And then there's the bus noise to consider ! Never mind the stuff after.
Graham
Arny Krueger
> "mlcguy" <vthor...@comcast.net> wrote in message
> news:1113084880.0...@g14g2000cwa.googlegroups.com...
>> What causes PA hiss? Is there anyway to control it or reduce it?
>>
>
> Quite a long list of things can cause hiss. Here is a minute list.
> Bad mic cables
No way.
> bad connectors
crackle yes, hiss no.
> running cables next to power lines
hum yes, hiss no
> running cables along steel beams
Say what?
> poor gain structure
Yes! a credible answer!
> defective mics
I guess so, but I find many more that crackle.
> defective hardware
IOW everything. Not really an answer.
> Reversed phase outlets (hot and neutral reversed)
hum yes, hiss no
> using unbalanced cables to connect equipment
hum yes, hiss unlikely. Balanced I/O gives only a 3 dB dynamic range
advantage. 3 dB is not a sufficient improvement to make people say
that the hiss went away.
> equipment plugged in to mixed phases, especially with cheap
equipment
hum yes, hiss no.
> or blown filters
Only if they are hiss filters...
> bad speakers, specifically the crossover
Actually, that's a pretty sure way to make hiss go away - disconnect
or short the tweeter.
> neon or florescent lights, especially with a bad ballast
hum is a far more likely consequence.
> any faulty equipment on the same circuit, not necessarily your sound
gear
Not any equipment, just some really bad equipment. The power line and
good equipment is supposed to be very robust and not cross-pollinate
noise.
> and the list goes on and on and on
Especially when about half of it is irrelevant to the question of
hiss.
> Best way to isolate the problem is to turn EVERYTHING off except the
> sound gear to eliminate outside interference. Then disconnect one
> piece of gear at a time until you find the problem gear. Then fix
or
> replace.
This works, but can be hard to implement. For example, can you turn
out all the lights and troubleshoot very effectively?
Then you have to troubleshoot among the equipment.
Pooh Bear
Arny Krueger wrote:
> Mike Dobony wrote:
>
> > or blown filters
>
> Only if they are hiss filters...
Surely *Monster* make a hiss filter ? ;-)
Graham
Kevin Aylward
> Kevin Aylward wrote:
>
>> Zigakly wrote:
>>>> What causes PA hiss? Is there anyway to control it or reduce it?
>>>
>>> 9 times out of 10 it's the EQ.
>>
>> Very unlikely. The EQ is usually after the initial gain stage. In
>> which case the input referred noise of the EQ will be divided by the
>> gain of that stage, say 40db.
>
> That's dB actually ! Go on - call me a pedant !
>
> You've never heard a noisy EQ section ?
Yes. 20 years ago:-)
>I assure you that they exist.
I am talking about competently designed ones, I don't use Realisic
mixers :-)
Note, I am still focusing on the thread question. For me, "PA hiss",
implicitly implies mic to mixer to power amp.
If the input is not a mic, such that signals are at line levels, then
the noise of devices like EQ's might be comparable to other bits and
pieces. But is this case, I would be surprised that hiss is an actual
problem that would be complained about. Its really only when one has the
mic gain up that hiss really becomes objectionable in a PA, well, again,
assuming the gear is not a total piece of shit.
>
>
>> If the gain structure is correct,
>
> Correct ? Is that the due to the design of the mixer or a result of
> its (mis)use ?
I haven't checked that many new mixers, the ones I use (Mackie and
Studiomaster:-)) are quite quiet. Certainly, unless I have the mic gain
up, noise is minimal.
>
>
>> noise should always be dominated by
>> the first mic stage, which typically has an input noise of around
>> 0.8nv-2nv/rthz. Since many opamps are in the 10nv-20nv and less
>> range, any front gain of more than about 20 should make the op amp
>> noise of EQ stages minimal.
>
> Hmmmmm. And I *don't* mean 'hum' - lol.
>
> In an ideal world maybe ?
Well, sure many mic amps have a higher noise when set away from max
gain. There is the cost cutting stuff as well. However, the general
principle is that all things being equal, most things after the first
stage is going to be of the order of 100 times less sensitive to noise
issues.
>
>
>> Its pretty obvious really, EQ and later stages can be run at 1V
>> levels,
>
> If the user drives them so ! Actually I like to drive them at higher
> volt levels ! Drive them at a hundred mV or so and the EQ noise
> contribution will be significant.
Possibly, if the design is not too hot. If set up/designed correctly,
there is no getting around the equivalent input noise concept. Any later
stages get referred to the input by division of the gain. So, if the
input stage has a gain of 40dB (:-)) the later stages have to be pretty
noisy to have a significant net effect.
>
>
>> while the mic amp is at the mv levels.
>>
>> Of course, EQing will make the noise of the front end larger if it
>> has a lot of boost.
>
> And then there's the bus noise to consider ! Never mind the stuff
> after.
Yes. This is a main offender after the mic stages. Its noise effect is
magnified of all the other connected inputs. The noise into it from each
channel will be rms summed, but the noise gain is a simple ratio of the
connected channels.
Gene Sweeny
Kevin Aylward wrote:
>
> Do you mean more gain?
Sensitivity and gain are not the same thing. And I meant sensitivity.
Most devices in a signal chain don't provide gain. They are unity with
repsect to themselves. By adjusting the input sensitivy, you can match
it's operating levels with a preceeding peice of gear. In most
systems, the only devices that are *designed* to provide gain are the
mixer and the amplifiers.
You could obviously use the input and output gain knobs on processing
gear to add gain however, this will just raise the noise floor, and if
you are amplifying a particularly quiet source, the raised noise floor
will be noticeable in the form of hiss.
> Yep, it will amplifier the noise. It will also amplifier the signal.
If your signal is close to the noise floor, this could suck.
> The *reason* for backing of the gain of later stages, and increasing
> earlier stage gain, is so that the noise of later stages don't add to
> the total noise. If the noise of the later stages is so low anyway,
then
> it makes no difference.
You are making assumptions to the quality of the gear etc.
>
> >
> > A simple example. Say I have a pc of gear (call it "the source")
that
> > clips at +28dbu and has 90db of signal to noise ratio. Then I
connect
> > it to an amplifier that clips at +4dbu. I don't attenuate the
amps,
> > and that effectively limits the signal to noise ratio to 66 db.
>
> Not necessarily. One can simply reduce the gain/output of the source.
A
> reduction in output level of the source does not imply the that
signal
> to noise will get worse by the same level. It depends where the
dominant
> noise sourse is.
I guess you *could* do this, but it would screw up your metering. For
instance, on a mixer, the output gain is controlled by the master
faders. If these are attenuated, the VU meters don't show the actual
signal level that the amps are seeing. Well... actually they do, but
this tells the FOH engineer nothing. However, if everything is
properly calibrated, when the mixer is showing signs of clipping, the
amps should be just starting to clip also.
>
> What is assumed in your argument is that one can't simply back of the
> gain of the source without unduly effecting its S/N. This is not
usually
> the case.
>
> > Basically, when the when the source hits +4dbu on it's meters it
will
> > clip the amp. A signal that is +4dbu at the source is only 66db
above
> > the noise floor. If I instead adjusted the amp to -20dbu (24 db of
> > att total) input, the amp would clip at the same point that the
source
> > clipped at +28dbu.
>
> Now suppose that the pot on the amp is physically located in the
source.
> What changes?
As long as the pot is the last device inside the given peice of gear
and doesn't squash dynamic range you are ok.
> >
> > These numbers may seem acceptable (66db of s/n ratio isn't awful
for
> > live applications) however, the problem becomes more complex as
more
> > peices of gear are added in the signal chain.
>
> This isn't a good example. A source that outputs +28db at 90db s/n
and
> can not be adjusted to a lower level without destroying its s/n is a
> poor source. Can you give me an example of such a poor design?
>
> For example, the Mixer. This often has an output amp with its gain
> controlled by a pot in its feedback loop. The inherent noise of this
> stage is usually insignificant compared with the noise generated from
> earlier stages. This means that the output level can be controlled
> without effecting the s/n.
Again, not true of all gear. If one adjusts the input sensitivity of
each peice of gear then you don't have to worry about it, it just
works. Your way, you have to know if the output gain truely does what
it says it does. From what I've read and had experience with, this is
not always the case.
Arny Krueger
Monster makes me think of large reptiles, while snakes are all
relatively small reptiles. Snakes hiss, so *Monster* does not make me
think of a hiss filter. ;-)
Phildo
"Pooh Bear" <rabbitsfriend...@hotmail.com> wrote in message
news:425A5F08...@hotmail.com...
> Would you *really* want to run a mix where the system clips when the
> output
> meters are half way up the scale ? I like to see the meters way into the
> yellow. :-)
I've had mackie VLZ _desks_ that clipped when you went into the yellow.
Nasty hard clipping as well.
Phildo
Kevin Aylward
> Kevin Aylward wrote:
>>
>> Do you mean more gain?
>
> Sensitivity and gain are not the same thing.
Ho hum...
>And I meant sensitivity.
> Most devices in a signal chain don't provide gain. They are unity
> with repsect to themselves. By adjusting the input sensitivy,
And what do you propose the sensitivity control actually does?
What I will say here is that our backgrounds are obviously different, as
we clearly use the same term with different meanings. My use of words
are typically more general that what you imply here. For example, "Gain"
automatically includes attenuation. It just has a negative sign in dB
terms.
>you
> can match it's operating levels with a preceeding peice of gear. In
> most systems, the only devices that are *designed* to provide gain
> are the mixer and the amplifiers.
Well, this is not really accurate. I used to a own a compressor that had
20db of gain. But anyway, this point is not relevant.
>
> You could obviously use the input and output gain knobs on processing
> gear to add gain however, this will just raise the noise floor, and if
> you are amplifying a particularly quiet source, the raised noise floor
> will be noticeable in the form of hiss.
I think you have lost the thread. My argument is that having the amp
pots full up wont degrade the s/n in any practical situation. If we back
off the PA amp gain pots then we have to increase the gain prior to the
amp. I am arguing that we should keep the prior gain attenuated, not
increased, if we have the pa amp pots full up.
This must be due to the confusion on what gain actually means.
Furthermore, in all arguments the inherent assumption is that the total
gain is kept *constant*. Raising the gain earlier on means backing it
off later, thereby keeping the core noise ideally the same, so you
argument above is not really relevant. The noise will be reduced later
on. You cant just have the gain up generating more noise as this means
the signal is going to clip.
What the issue that lead to this debate is, is whether or not it made
any sense to back off the pa amp gain for white noise reasons. My view
is that it doesn't.
>
>
>> Yep, it will amplifier the noise. It will also amplifier the signal.
>
> If your signal is close to the noise floor, this could suck.
If the signal is close to the noise floor, the equipment is useless for
audio.
>
>> The *reason* for backing of the gain of later stages, and increasing
>> earlier stage gain, is so that the noise of later stages don't add to
>
>> the total noise. If the noise of the later stages is so low anyway,
>> then it makes no difference.
>
> You are making assumptions to the quality of the gear etc.
Sure, I am making an assumption on the post question:
"What causes PA hiss? Is there anyway to control it or reduce it?"
I read this as a mic->mixer->amp chain.
A mic input gain of 40db should result in the mic noise being the
dominate noise source.
I have stated a few times. Typical outboard gear is built with
10-20nv/rthz noise op-amps. This referred to the mic input with a gain
of 40db, is only 0.5nv.
>
>
>>
>>>
>>> A simple example. Say I have a pc of gear (call it "the source")
>>> that clips at +28dbu and has 90db of signal to noise ratio. Then I
>>> connect it to an amplifier that clips at +4dbu. I don't attenuate
>>> the amps, and that effectively limits the signal to noise ratio to
>>> 66 db.
>>
>> Not necessarily. One can simply reduce the gain/output of the
>> source. A reduction in output level of the source does not imply the
>> that
> signal
>> to noise will get worse by the same level. It depends where the
>> dominant noise sourse is.
>
> I guess you *could* do this, but it would screw up your metering.
Its done all the time. Power amp gain pots are often full up.
> For
> instance, on a mixer, the output gain is controlled by the master
> faders.
Yep it sure is.
>If these are attenuated, the VU meters don't show the actual
> signal level that the amps are seeing.
You've lost me there dude. The meters are *after* the faders. Its
*these* faders that I am discussing. Whatever the faders send to the
meters, is what is sent to the amp.
> Well... actually they do, but
> this tells the FOH engineer nothing. However, if everything is
> properly calibrated, when the mixer is showing signs of clipping, the
> amps should be just starting to clip also.
As it is if one has the PA amp pots full up, and the mixer calibrated as
standard. Backing off the power amp pots makes the system un-calibrated.
>
>>
>> What is assumed in your argument is that one can't simply back of the
>
>> gain of the source without unduly effecting its S/N. This is not
>> usually the case.
>>
>>> Basically, when the when the source hits +4dbu on it's meters it
>>> will clip the amp. A signal that is +4dbu at the source is only
>>> 66db above the noise floor. If I instead adjusted the amp to
>>> -20dbu (24 db of att total) input, the amp would clip at the same
>>> point that the source clipped at +28dbu.
>>
>> Now suppose that the pot on the amp is physically located in the
>> source. What changes?
>
> As long as the pot is the last device inside the given peice of gear
> and doesn't squash dynamic range you are ok.
By and large it is. Its the master faders of the mixer. It is usually in
the feedback of the last gain stage. Although sometimes they may be a
fixed low gain buffer following it.
I think you missed my point though. It is a topological one.
>
>>>
>>> These numbers may seem acceptable (66db of s/n ratio isn't awful
> for
>>> live applications) however, the problem becomes more complex as
> more
>>> peices of gear are added in the signal chain.
>>
>> This isn't a good example. A source that outputs +28db at 90db s/n
> and
>> can not be adjusted to a lower level without destroying its s/n is a
>> poor source. Can you give me an example of such a poor design?
>>
>> For example, the Mixer. This often has an output amp with its gain
>> controlled by a pot in its feedback loop. The inherent noise of this
>> stage is usually insignificant compared with the noise generated from
>
>> earlier stages. This means that the output level can be controlled
>> without effecting the s/n.
>
> Again, not true of all gear.
But not relevant to the question of this thread. Sure, there is some
bad outboard kit out there, but by and large, its all line level stuff
using opamps with noise levels way larger than that of the mic stage.
For the far majority of cases, once the mic signal is up to level, it
dominates the noise of the whole system. If it don't, you would do
better replace that Radio Shack EQ, its a positive embarrassment.
Gene Sweeny
Kevin Aylward wrote:
>
> > You could obviously use the input and output gain knobs on
processing
> > gear to add gain however, this will just raise the noise floor, and
if
> > you are amplifying a particularly quiet source, the raised noise
floor
> > will be noticeable in the form of hiss.
>
> I think you have lost the thread. My argument is that having the amp
> pots full up wont degrade the s/n in any practical situation. If we
back
> off the PA amp gain pots then we have to increase the gain prior to
the
> amp. I am arguing that we should keep the prior gain attenuated, not
> increased, if we have the pa amp pots full up.
I'll agree, but this could lead to bad things depending on the output
control of preceeding devices. Also, many EQs that I've seen only
provide an input control, not an output control. This may make it hard
to match the operating levels of gear. In this situation, how would
you attenuate the EQ to match the amplifiers operating levels?
>
> You've lost me there dude. The meters are *after* the faders. Its
> *these* faders that I am discussing. Whatever the faders send to the
> meters, is what is sent to the amp.
True, but if your meters are reading -10dBu and your amps are clipping,
what does that tell you? This situation could very easliy occur with
your output attenuating methods.
>
> > Well... actually they do, but
> > this tells the FOH engineer nothing. However, if everything is
> > properly calibrated, when the mixer is showing signs of clipping,
the
> > amps should be just starting to clip also.
>
> As it is if one has the PA amp pots full up, and the mixer calibrated
as
> standard. Backing off the power amp pots makes the system
un-calibrated.
I disagree. Especially in systems that use different amplifier models
in different modes (bridged vs. stereo, etc.). if not probably
adjusted the amps will clip at different points. You don't want your
top amps clipping before your sub amps or vice versa. The purpose of
the gain knobs is to calibrate everything to unity gain. Otherwise...
why are they there?
Basically, my assertion is unity gain can not be acheived with your
method. Your method may reduce PA hiss. However, the method that I
use and have seen numerous other places reduces PA hiss AND achieves
unity gain for the system.
Bob Quintal
> GeezerSonics wrote:
>>
>> Most amplifiers provide MAXIMUM output with +4dB input
>> levels. That's 20dB> sensitivity then needed. That's why many
>> people recommend thaty attenuate the amplifier's input. This
>> allows the entire signal chain to run at optimum, and allows
>> the console and amplifier to have the same clip point.
>
> Sure, but irrelevent to hiss.
>
I don't see how you can say this with a straight face.
Let us take an hypothetical example. say a mixer generates a
constant 0.228 mVRMS of hiss voltage, irrespective of output
signal, in addition to any hiss amplified by the device. The
mixer has a +4dBm nominal output level with 24 dB of headroom to
clipping. It drives an amplifier that is set to produce full
output with a +4 dBm signal.
With those facts, if we reduce the acoustic output of the
speakers by 20 dB, we can reduce the input sensitivity of the
amplifier by that amount, or the mixer output by that amount.
What about S/N ratios in the two options?
since the hiss voltage from the output of the mixer is constant,
reducing the output of the mixer reduces the ratio by the 20 dB
of gain reduction. Turning down the amplifier's gain maintains
the ratio at whatever it was.
Why did I specify the 0.228 mV noise output? 1st it is close to
real post master fader noise levels of several mixers I've
measured, and it's 80dB down from nominal input.
If the maximum SPL from the speakers is 120dBa, that means that
the 20 dB reduction reduces the sound level to 100 dBa, a
comfortable (rock and roll venue) level. Doing it by turning down
the amplifier means that the 80dB Signal to hiss ratio is
maintained giving 20 dBa of hiss. That is barely noticable during
silences. But turning down the amp faders means that there is 40
dba of hiss coming out of the speakers when there is no signal,
and most people can hear that quite well.
Add to that, say, 6 to 10 dB of hiss from preamps and sources
such as keyboards, Effects Units etc, and you have evident hiss
coming out of the speakers.
.
>
>>20dB encroachment of amplfier
>> sensitivity on S/N ratio.
>
> It doesn't effect the s/n ratio, as far as hiss goes of the
> system at all. The fact that the gain at the amp may be
> reduced by 20db, does not mean that the s/n is so effected.
Read above. It doeds too. absoultely, and unequivocably. That's
the law (of physics).
>
> The noise level of most PA amps is so low as to be completely
> irrelevant in computing final white noise s/n ratios. Hint: go
> and stick your ear up to the speakers when the power amp is on
> with no input. You will be hard pressed to hear any hiss at
> all. Since this is the *maximum* noise it can make, if you
> cant hear the hiss with no signal, you wil certainly not hear
> it with a signal.
>
The noise level of the amp itself isn't the problem, but the
amp's available gain over the required gain does contribute dB
for dB to the audible level of hiss coming out of the speaker
connected to that amp.
--
Bob Quintal
PA is y I've altered my email address.
Pooh Bear
Nicely put Bob.
Even though I tend to run amps with their inputs attenuated by up to
10dB it's surprising just how much audible hiss is apparent near the
stacks when the room is quiet ( say after the soundcheck ).
Graham
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in
> news:h1u6e.21023$mV1....@fe3.news.blueyonder.co.uk:
>
>> GeezerSonics wrote:
>>>
>>> Most amplifiers provide MAXIMUM output with +4dB input
>>> levels. That's 20dB> sensitivity then needed. That's why many
>>> people recommend thaty attenuate the amplifier's input. This
>>> allows the entire signal chain to run at optimum, and allows
>>> the console and amplifier to have the same clip point.
>>
>> Sure, but irrelevent to hiss.
>>
> I don't see how you can say this with a straight face.
No prblem at all.
>
> Let us take an hypothetical example. say a mixer generates a
> constant 0.228 mVRMS of hiss voltage,
This seems a bit large for a competently designed mixer. The main output
fader is usually at or very near the output, maybe an extra 10db max
fixed gain after the fader. This fader is in the feedback loop. The
fader will reduce the noise when fully down of all preceding stages and
its own stage. The keep warm output noise of the output opamp, if there,
will be of the order of 20nv*141*gain=15uV. This is 96db down on a 1V
signal.
This is approximate worst case. Many mixers have no final gain, in which
case the output noise will be a few uV.
> irrespective of output
> signal, in addition to any hiss amplified by the device. The
> mixer has a +4dBm nominal output level with 24 dB of headroom to
> clipping. It drives an amplifier that is set to produce full
> output with a +4 dBm signal.
>
> With those facts, if we reduce the acoustic output of the
> speakers by 20 dB, we can reduce the input sensitivity of the
> amplifier by that amount, or the mixer output by that amount.
>
> What about S/N ratios in the two options?
None of this is relevant. What *dominates* the noise is the mic stage.
This is such a simple concept I am at a loss as to what the issue is in
understanding it.
Each noise souce must be appropriately added. If there is one source
that dominates, then we can effectively ignore the rest.
Sure, if a later stage is *dreadfully* noisy it could be significant,
but I don't see how this can occur for the output amp, knowing what the
noise of typical opamp is, and the noise of mic amps.
>
> since the hiss voltage from the output of the mixer is constant,
> reducing the output of the mixer reduces the ratio by the 20 dB
> of gain reduction. Turning down the amplifier's gain maintains
> the ratio at whatever it was.
Sure, there is a constant noise at various parts of the chain, but this
noise is usually way way less than the noise generated by the mic stage
so we can ignore it.
>
> Why did I specify the 0.228 mV noise output? 1st it is close to
> real post master fader noise levels of several mixers I've
> measured, and it's 80dB down from nominal input.
I would suggest that you have measured incorrectly. Probably measuring
hum.
Fact. I just now turned down the faders pots on my mackie mixer (and
studiomaster as a 2nd check), with my mackie pa amp gain full up. The
hiss is completly insignificant when the faders are down. And oh, dear,
and I do this with my 15 year old Tascam 8 track driving MOSFET 1000, a
bit more hiss, but still insignificant. In all case the hiss is way less
then the fan noise!
Are you using Realistic/Tandy gear by chance?
>
> If the maximum SPL from the speakers is 120dBa, that means that
> the 20 dB reduction reduces the sound level to 100 dBa, a
> comfortable (rock and roll venue) level. Doing it by turning down
> the amplifier means that the 80dB Signal to hiss ratio is
> maintained giving 20 dBa of hiss. That is barely noticable during
> silences. But turning down the amp faders means that there is 40
> dba of hiss coming out of the speakers when there is no signal,
> and most people can hear that quite well.
This is nonsense.
With the mics *on* the total noise at the output will be the rms sum of
mic_gain*mic_input noise and the post fader noise. The post fader noise
will be insignificant as the typical mic gain from that point may be say
50db-60db.
I suggest you actually go and do the test. Keep the total gain constant
with and without the pa amp backed of to -20db. If you notice a
difference, you have shit gear.
>
> Add to that, say, 6 to 10 dB of hiss from preamps and sources
> such as keyboards, Effects Units etc, and you have evident hiss
> coming out of the speakers.
> .
>>
>>> 20dB encroachment of amplfier
>>> sensitivity on S/N ratio.
>>
>> It doesn't effect the s/n ratio, as far as hiss goes of the
>> system at all. The fact that the gain at the amp may be
>> reduced by 20db, does not mean that the s/n is so effected.
>
> Read above. It doeds too. absoultely, and unequivocably.
Unequivocally not, unless your output stage has an equivalent input
noise that is truly horrendous.
>That's
> the law (of physics).
Yes its simple physics. RMS Add each noise source divided by the gain up
to that noise source.
>>
>> The noise level of most PA amps is so low as to be completely
>> irrelevant in computing final white noise s/n ratios. Hint: go
>> and stick your ear up to the speakers when the power amp is on
>> with no input. You will be hard pressed to hear any hiss at
>> all. Since this is the *maximum* noise it can make, if you
>> cant hear the hiss with no signal, you wil certainly not hear
>> it with a signal.
>>
> The noise level of the amp itself isn't the problem, but the
> amp's available gain over the required gain does contribute dB
> for dB to the audible level of hiss coming out of the speaker
> connected to that amp.
Not in practice it don't. You should really get rid of that Tandy kit
mate.
Kevin Aylward
> Kevin Aylward wrote:
>
>>
>>> You could obviously use the input and output gain knobs on
>>> processing gear to add gain however, this will just raise the noise
>>> floor, and if you are amplifying a particularly quiet source, the
>>> raised noise floor will be noticeable in the form of hiss.
>>
>> I think you have lost the thread. My argument is that having the amp
>> pots full up wont degrade the s/n in any practical situation. If we
>> back off the PA amp gain pots then we have to increase the gain
>> prior to the amp. I am arguing that we should keep the prior gain
>> attenuated, not increased, if we have the pa amp pots full up.
>
> I'll agree, but this could lead to bad things depending on the output
> control of preceeding devices. Also, many EQs that I've seen only
> provide an input control, not an output control. This may make it
> hard to match the operating levels of gear. In this situation, how
> would you attenuate the EQ to match the amplifiers operating levels?
The precending device is the Mixer.
Hint: "What causes PA hiss? Is there anyway to control it or reduce it?"
I have already pointed many times, its pretty clear that the reasonable
assumption is that the question is about the mic->mixer->amp chain.
There is not much that one will put between the mixer and the amp,
especially in a live situation (again an assumption of the post).
Possible a graphic. However, this is all line level unity gain stuff,
and any one worth its salt will not generate any relevant noise. Again,
its noise is divided by the gain up to that point, which is pretty large
indeed.
>
>>
>> You've lost me there dude. The meters are *after* the faders. Its
>> *these* faders that I am discussing. Whatever the faders send to the
>> meters, is what is sent to the amp.
>
> True, but if your meters are reading -10dBu and your amps are
> clipping, what does that tell you? This situation could very easliy
> occur with your output attenuating methods.
No it couldnt. Your just not listening. The situation I am describing is
leaving everything as is. i.e. mixer is calibrated as per factory, and
the the amp is calibrated as per factory with its gain pot at max. In
this situation the meters are reading correctly. End of story.
I am not suggesting that the mixer output is attenuated at the output
after the jacks. I am simply stating that the mixer faders are
controlling the gain, as per usual.
>
>>
>>> Well... actually they do, but
>>> this tells the FOH engineer nothing. However, if everything is
>>> properly calibrated, when the mixer is showing signs of clipping,
>>> the amps should be just starting to clip also.
>>
>> As it is if one has the PA amp pots full up, and the mixer
>> calibrated as standard. Backing off the power amp pots makes the
>> system un-calibrated.
>
> I disagree.
Oh?
>Especially in systems that use different amplifier models
> in different modes (bridged vs. stereo, etc.). if not probably
> adjusted the amps will clip at different points. You don't want your
> top amps clipping before your sub amps or vice versa. The purpose of
> the gain knobs is to calibrate everything to unity gain. Otherwise...
> why are they there?
>
>
> Basically, my assertion is unity gain can not be acheived with your
> method. Your method may reduce PA hiss.
Again, you have still lost it. You appear to be making an attempt to
contradict me, just for the sake of it. You are paying no attention to
the actual point of this thread or what I am actually claiming. I have
continuously claimed that setting the pa amp pot either to max or
say, -20db does not effect the white noise s/n in practice, so why you
indicted that I am claiming that leaving the amp pot at max, or -20db (I
don't even think you know what the claims is) would effect the noise is
beyond me.
> However, the method that I
> use and have seen numerous other places reduces PA hiss
Simply not possible for the majority of systems. A competent PA system
will have its noise floor dominated by the mic stage. As I have stated
many times. ein at any point in the chain is referred to the input by
dividing by the gain. If the mic gain is of the order of 40db, where the
pa amp pot is set, is pretty much irrelevant as far as white noise is
concerned. Which is what is being discussed. You appear to be discussing
other things not in debate.
With all due respect, do you actually understand technical details of
analogue design? It doesn't seem you do. Everything you say appears to
be all from a technician's point of view.
>AND achieves
> unity gain for the system.
Irrelevent.
Gene Sweeny
Kevin Aylward wrote:
>
> The precending device is the Mixer.
>
> Hint: "What causes PA hiss? Is there anyway to control it or reduce
it?"
>
> I have already pointed many times, its pretty clear that the
reasonable
> assumption is that the question is about the mic->mixer->amp chain.
>
> There is not much that one will put between the mixer and the amp,
> especially in a live situation (again an assumption of the post).
> Possible a graphic. However, this is all line level unity gain stuff,
> and any one worth its salt will not generate any relevant noise.
Again,
> its noise is divided by the gain up to that point, which is pretty
large
> indeed.
I think you need to get out and see more live sound rigs. Also, I
still recommend you go read more of the articles on prosoundweb.com.
They don't put them there cause they are wrong.
> >>
> >> You've lost me there dude. The meters are *after* the faders. Its
> >> *these* faders that I am discussing. Whatever the faders send to
the
> >> meters, is what is sent to the amp.
> >
> > True, but if your meters are reading -10dBu and your amps are
> > clipping, what does that tell you? This situation could very
easliy
> > occur with your output attenuating methods.
>
> No it couldnt. Your just not listening. The situation I am describing
is
> leaving everything as is. i.e. mixer is calibrated as per factory,
and
> the the amp is calibrated as per factory with its gain pot at max. In
> this situation the meters are reading correctly. End of story.
They are reading what is exiting the mixer, yes. But as an FOH
engineer, this tells you nothing if your system is not adjusted
properly. It also forces the signal to operate closer to the noise
floor of the amplifier.
Back to examples. Say I have an amp who's input sensitivity is .775
Vrms. A very common input sensitivity. This equates to 0dBu
(obviously). What this means is that at 0dBu input, the amplifier will
begin to clip beyond that point. It also has a 103 dB s/n ratio. Now
our mixer, it clips at +22dBu. and has a signal to noise ratio of 97
dB. That puts the noise floor for the amp at -103 dB and the noise
floor of the mixer at -75dB. Now, if we don't attenuate the amps, and
I run a mix at 0dB on the console, the gig is going to be exteremely
loud (maybe this is called for, but I doubt it if you can hear hiss
when no music is playing). In order to keep the mix at a resonable
volume, I would be forced to put the master faders at say -20 or -30.
Especailly in situations where you're doing solely vocal reforcement
for a corporate conference or something. These are the situations
where hiss most rears it's ugly head. Now... I'll agree, that having
the faders at -30 will also attenuate the noise -30 coming from the
mixer. So your s/n ratio shouldn't change. However, the VU meters
don't reflect where the amps are at in terms of their ability.
If you instead attenuated the amps about 22 db at their input, the
master VU meters would show red when the amps are about to clip AND the
noise situation would be no worse than your solution.
Back to the real world. MOST rigs DO have a graphic, limiter, delay,
crossovers and possibly some other processing inserted between the
mixer and amps. These all clip at different points, have different s/n
ratios and add noise.
IMO your assumption of a a mix-> mixer -> amp signal chain is a naive
one and is probably true in only the smallest of bar systems. And even
in THAT situation, I would expect the mixer and amps to be of poorer
design and not have the noise performance you are so sure of.
So my assertion is that your method will screw the OP up even more,
though we need more info to be sure.
>
> I am not suggesting that the mixer output is attenuated at the output
> after the jacks. I am simply stating that the mixer faders are
> controlling the gain, as per usual.
Most mixing console manuals I've read suggest and/or recommend the
master faders be left at 0dB for best performance from the mixer.
Maybe it doesn't. Let's forget that for a moment. You said in
original reply to GeezerSonics:
>> The first remedy, attenuate the input of your power amplifier(s) use
>> the gain available from your console and processing equipment.
> Yes, although I doubt if many amps nowadays generate relevant hiss.
This
> can be more useful for mitigating the effect of ground induced hum.
THAT is not the reason to turn down amplifier gain. The reason is with
the pots up full, the amplifier will amplify all noise in signal chain
by it's fixed amount of gain, period. Additionally, if the noise floor
is high, the amps will in turn amplify this to audible levels.
So turning them down, does fix hiss problems, AND it achieves unity
gain. YOu seem to think the issues are separate, but they are not.
You need to achieve both goals (no hiss and unity gain). Leaving amps
on full while it will achieve the first (if the preceeding equipment is
attenuated) is not the way to acheive the second goal.
>
> > However, the method that I
> > use and have seen numerous other places reduces PA hiss
>
> Simply not possible for the majority of systems. A competent PA
system
> will have its noise floor dominated by the mic stage. As I have
stated
> many times. ein at any point in the chain is referred to the input by
> dividing by the gain. If the mic gain is of the order of 40db, where
the
> pa amp pot is set, is pretty much irrelevant as far as white noise is
> concerned. Which is what is being discussed. You appear to be
discussing
> other things not in debate.
What I am discussing is that your solution to the debate does not
achieve the other goal of unity gain.
>
> With all due respect, do you actually understand technical details of
> analogue design? It doesn't seem you do. Everything you say appears
to
> be all from a technician's point of view.
With all do respect, have you ever stepped out of your box to look at
the big picture? Do you even mix on a console on a regular basis?
>From the Bio on your website (http://www.anasoft.co.uk/band/index.htm)
I can see that you are an Engineer who plays guitar. Wonderful. Maybe
you can design a quiet guitar amp, yay. I like how your top three
favorite quotes are from yourself.
>
> >AND achieves
> > unity gain for the system.
>
> Irrelevent.
It is most certianly relavent in the real world of sound reinforcement.
Kevin Aylward
> Kevin Aylward wrote:
>>
>> There is not much that one will put between the mixer and the amp,
>> especially in a live situation (again an assumption of the post).
>> Possible a graphic. However, this is all line level unity gain stuff,
>
>> and any one worth its salt will not generate any relevant noise.
>> Again, its noise is divided by the gain up to that point, which is
>> pretty large indeed.
>
> I think you need to get out and see more live sound rigs.
Oh?
> Also, I
> still recommend you go read more of the articles on prosoundweb.com.
> They don't put them there cause they are wrong.
Oh?
>
>>>>
>>>> You've lost me there dude. The meters are *after* the faders. Its
>>>> *these* faders that I am discussing. Whatever the faders send to
>>>> the meters, is what is sent to the amp.
>>>
>>> True, but if your meters are reading -10dBu and your amps are
>>> clipping, what does that tell you? This situation could very
> easliy
>>> occur with your output attenuating methods.
>>
>> No it couldnt. Your just not listening. The situation I am
>> describing is leaving everything as is. i.e. mixer is calibrated as
>> per factory,
> and
>> the the amp is calibrated as per factory with its gain pot at max. In
>
>> this situation the meters are reading correctly. End of story.
>
> They are reading what is exiting the mixer, yes. But as an FOH
> engineer,
Oh...what's an FOH engineer? Does that trump a B.Sc. degree, 25 years
experience analogue design engineer who actually has designed
professional audio equipment for a rather well known and respected pro
audio company? Hint: its the bit about Studiomaster on my bio.
What's the probability, that such a someone who has been playing guitar
since 11, done loads of gigs, even doing many repairs for a music shop
during university for the odd 20 quid, is out to lunch?
Look, dude you need to listen to what I am actually saying. I have been
at this a while. You know the one about teaching your grandpa to suck
eggs?
> this tells you nothing if your system is not adjusted
> properly. It also forces the signal to operate closer to the noise
> floor of the amplifier.
This sentence doesn't mean anything. What amplifier. What are you trying
to say?
The PA amp white noise is irrelevant. Its way to low to be significant.
I have already explained. Stick your ear to a PA amp. It makes no noise.
It dont matter where its gain pot is. It still wont make any noise in
any practical situation.
>
> Back to examples. Say I have an amp who's input sensitivity is .775
> Vrms. A very common input sensitivity. This equates to 0dBu
> (obviously). What this means is that at 0dBu input, the amplifier
> will begin to clip beyond that point. It also has a 103 dB s/n
> ratio. Now our mixer, it clips at +22dBu. and has a signal to noise
> ratio of 97 dB. That puts the noise floor for the amp at -103 dB and
> the noise floor of the mixer at -75dB.
I assume what you mean here, is that if you back of the output of the
mixer by 22db its s/n will go to -75db. I have already explained this
will *not* happen. The fader will reduce both the signal and *noise*.
Only the noise of a post fader buffer/gain stage, if there, will stay
the same. This last stage will have very low noise relative to the
gained up mic amp noise that will add at this stage. Lets say it has a
10db gain, and an input noise of 20nv/rthz, this will be 3uv (*
sqrt(BW)). If the signal is say 1V, it will be 110 db s/n, therefore
negligable.
> Now, if we don't attenuate
> the amps, and I run a mix at 0dB on the console, the gig is going to
> be exteremely loud (maybe this is called for, but I doubt it if you
> can hear hiss when no music is playing). In order to keep the mix at
> a resonable volume, I would be forced to put the master faders at say
> -20 or -30. Especailly in situations where you're doing solely vocal
> reforcement for a corporate conference or something. These are the
> situations where hiss most rears it's ugly head. Now... I'll agree,
> that having the faders at -30 will also attenuate the noise -30
> coming from the mixer. So your s/n ratio shouldn't change.
So why are you claiming that the mixer s/n gets reduced by 22db above?
> However,
> the VU meters don't reflect where the amps are at in terms of their
> ability.
As the amp gain is max, the mixer and amp is calibrated, then the VU
meters are telling you what the amp is doing. End of story.
Sure, you might not have the full scale resolution, i.e the meters going
up to +22db typically, but this isnt relevent to the posters question. I
am only addressing hiss. Well, ok I did mention the bit about backing of
the pa amp gain is a good idea to reduce the effect of hum.
>
> If you instead attenuated the amps about 22 db at their input, the
> master VU meters would show red when the amps are about to clip AND
> the noise situation would be no worse than your solution.
Look, dude what part of "I am discussing hiss" do you have the trouble
with. You keep wanting to extend this to optimum setting system with
regard to matters other than hiss.
The posters question was one line.
>
> Back to the real world. MOST rigs DO have a graphic, limiter, delay,
> crossovers and possibly some other processing inserted between the
> mixer and amps. These all clip at different points, have different
> s/n ratios and add noise.
See above calculation on what sort of noise they should generate. Look,
all noise gets referred to the input divided by the gain. How many times
do I have to say this?
This is about practicalities like 1nv + 0.05nv.
As an engineer, this result is still 1nv.
Look, if the mic amp isn't dominating the noise floor, something is
wrong.
>
> IMO your assumption of a a mix-> mixer -> amp signal chain is a naive
> one
Oh...
>and is probably true in only the smallest of bar systems.
That is, the type that 100,000 bar bands use. Do you actually go to
bars?
> And
> even in THAT situation, I would expect the mixer and amps to be of
> poorer design and not have the noise performance you are so sure of.
Oh?
Mackies are dirt cheap, and very low noise, as are Behringer. Look, its
moved on from 20 years ago. Mixers and amps are a mature, saturated
market. There are plenty of good cheap, kit out there
You need to get out more and see what us players are actually using.
>
> So my assertion is that your method will screw the OP up even more,
> though we need more info to be sure.
My suggestion for noise was turn off unused mic channels and set the
gain to get the peak light to momentarily flash. That's about the size
of it.
Optimising the system for some other *reason*, is not being debated by
me in the slightest. It wasn't requested.
>
>>
>> I am not suggesting that the mixer output is attenuated at the output
>
>> after the jacks. I am simply stating that the mixer faders are
>> controlling the gain, as per usual.
>
> Most mixing console manuals I've read suggest and/or recommend the
> master faders be left at 0dB for best performance from the mixer.
This is the problem. With all due respect, you understanding is at
technician level, that is one that gets their information from
manufactures manuals. Look, dude I design the stuff, professionally. Who
do you think writes these manuals?
>>>
>>> Basically, my assertion is unity gain can not be acheived with your
>>> method. Your method may reduce PA hiss.
>>
>> Again, you have still lost it. You appear to be making an attempt to
>> contradict me, just for the sake of it. You are paying no attention
> to
>> the actual point of this thread or what I am actually claiming. I
> have
>> continuously claimed that setting the pa amp pot either to max or
>> say, -20db does not effect the white noise s/n in practice, so why
> you
>> indicted that I am claiming that leaving the amp pot at max, or
>> -20db (I don't even think you know what the claims is) would effect
>> the noise is beyond me.
>
> Maybe it doesn't. Let's forget that for a moment. You said in
> original reply to GeezerSonics:
>
>>> The first remedy, attenuate the input of your power amplifier(s) use
>>> the gain available from your console and processing equipment.
>
>> Yes, although I doubt if many amps nowadays generate relevant hiss.
>> This can be more useful for mitigating the effect of ground induced
>> hum.
>
> THAT is not the reason to turn down amplifier gain. The reason is
> with the pots up full, the amplifier will amplify all noise in signal
> chain by it's fixed amount of gain, period.
But irrelevant. It wont, to all intents and purposes, effect the s/n
ratio, so to suggest doing so to get a better s/n is nonsense. Of
course, they may be a reason to do this for some other reason, but that
isn't in debate by me.
>Additionally, if the
> noise floor is high, the amps will in turn amplify this to audible
> levels.
Oh dear...You just don't understand. It also amplifies the signal as
well so the s/n remains constant.
>
> So turning them down, does fix hiss problems,
No it doesn't. See above calculations.
I suppose I will have to add a noise paper to my design tutorials at
http://www.anasoft.co.uk/EE/index.html, as this seems to be not
understood by many.
> AND it achieves unity
> gain. YOu seem to think the issues are separate, but they are not.
They are.
> You need to achieve both goals (no hiss and unity gain). Leaving amps
> on full while it will achieve the first (if the preceeding equipment
> is attenuated) is not the way to acheive the second goal.
The second goal is irrelevant to noise.
>
>>
>>> However, the method that I
>>> use and have seen numerous other places reduces PA hiss
>>
>> Simply not possible for the majority of systems. A competent PA
> system
>> will have its noise floor dominated by the mic stage. As I have
> stated
>> many times. ein at any point in the chain is referred to the input by
>
>> dividing by the gain. If the mic gain is of the order of 40db, where
>> the pa amp pot is set, is pretty much irrelevant as far as white
>> noise is
>
>> concerned. Which is what is being discussed. You appear to be
>> discussing other things not in debate.
>
> What I am discussing is that your solution to the debate does not
> achieve the other goal of unity gain.
I am not debating unity gain. I am debating noise.
There is no unity gain goal for minimising noise. "Unity gain" has no
relevance to the posters question.
>
>>
>> With all due respect, do you actually understand technical details of
>
>> analogue design? It doesn't seem you do. Everything you say appears
> to
>> be all from a technician's point of view.
>
> With all do respect, have you ever stepped out of your box to look at
> the big picture? Do you even mix on a console on a regular basis?
>
>> From the Bio on your website
>> (http://www.anasoft.co.uk/band/index.htm)
> I can see that you are an Engineer who plays guitar. Wonderful.
Indeed. For many years.
> Maybe you can design a quiet guitar amp, yay.
Yep.
> I like how your top
> three favorite quotes are from yourself.
Yep.
>
>>
>>> AND achieves
>>> unity gain for the system.
>>
>> Irrelevent.
>
> It is most certianly relavent in the real world of sound
> reinforcement.
Not really.
Gene Sweeny
>
> Oh...what's an FOH engineer? Does that trump a B.Sc. degree, 25 years
> experience analogue design engineer who actually has designed
> professional audio equipment for a rather well known and respected
pro
> audio company? Hint: its the bit about Studiomaster on my bio.
>
> What's the probability, that such a someone who has been playing
guitar
> since 11, done loads of gigs, even doing many repairs for a music
shop
> during university for the odd 20 quid, is out to lunch?
Dude, I have a B.Sc. in software engineering/Computer Science. With
several years experience, currently employed as a signal analyst and
software/systems engineer. Additionally I have been doing the FOH
stuff on the side for a few years. The biggest advantage I have that
trumps you is that I've got a brain that actually works.
I've met lots of 25 year engineers that don't have a clue what they are
doing. I'll just add another to my list.
I refuse to carry on this argument any longer. Nothing you say will
convince me otherwise as the the things I've stated are facts and are
relavent to both noise, hiss, s/n ratio and other topics discussed.
You've chosen to describe methods that are non-standard and won't offer
ther results most ppl in this business would expect. I do not doubt
that any PA system that you would set up would not perform optimally
and/or would still hiss like an asp.
Kevin Aylward
> Kevin Aylward wrote:
>>
>> Oh...what's an FOH engineer? Does that trump a B.Sc. degree, 25 years
>
>> experience analogue design engineer who actually has designed
>> professional audio equipment for a rather well known and respected
> pro
>> audio company? Hint: its the bit about Studiomaster on my bio.
>>
>> What's the probability, that such a someone who has been playing
>> guitar since 11, done loads of gigs, even doing many repairs for a
>> music
> shop
>> during university for the odd 20 quid, is out to lunch?
>
> Dude, I have a B.Sc. in software engineering/Computer Science.
I see. That explains your lack of analogue processing knowledge then.
>With
> several years experience, currently employed as a signal analyst and
> software/systems engineer.
And this has exactly what relevance to analogue design?
>Additionally I have been doing the FOH
No idea what FOH means.
Stop press, I looked it up, front of house oh dear...another layman
calling himself an "engineer".
> stuff on the side for a few years.
On the side. You don't say.
>The biggest advantage I have that
> trumps you is that I've got a brain that actually works.
Oh...now down to the insults. I see from this that you now admit that
you lost the argument. Especially, since you have now sniped all my
technical content, as you are unable to refute any of it.
>
> I've met lots of 25 year engineers that don't have a clue what they
> are doing. I'll just add another to my list.
ROTFLMAO
>
> I refuse to carry on this argument any longer. Nothing you say will
> convince me otherwise
I sure it won't. You dont have the nessesary background to understand
the arguments.
>as the the things I've stated are facts and are
> relavent to both noise, hiss, s/n ratio and other topics discussed.
Its plainly obvious that you have next to zero understanding of analogue
design and analogue signal processing.
> You've chosen to describe methods that are non-standard and won't
> offer ther results most ppl in this business would expect.
Yeah, sure...
> I do not
> doubt that any PA system that you would set up would not perform
> optimally and/or would still hiss like an asp.
Strange, as my stuff dosnt do this.
Next time I would like a software engineer (and I use the term engineer
here loosely) to give me some advice on PA setting up, I'll let you
know, but don't hold your breath.
Look mate, your out of your depth.
Gene Sweeny
Kevin Aylward wrote:
> >
> > Dude, I have a B.Sc. in software engineering/Computer Science.
>
> I see. That explains your lack of analogue processing knowledge then.
Probably not much.
>
> >With
> > several years experience, currently employed as a signal analyst
and
> > software/systems engineer.
>
> And this has exactly what relevance to analogue design?
Probably not a whole lot except the RF signal analysis I do.
>
> >Additionally I have been doing the FOH
>
> No idea what FOH means.
hahahahahhahahahahahahaha. Proof you don't belong here.
>
> Stop press, I looked it up, front of house oh dear...another layman
> calling himself an "engineer".
Anybody can look something up. Same with analog design. Any skill can
be learned by any technically competent person. I work in an
environment where CS, EE, and ME backrounds work cohesively to produce
some of the best products for our clients and that you'll never see.
Just cause you have a BS in EE doesn't make you sound reinforcement
god. Neither does a million years of guitar playing. Electric guitar
players are the proverbial thorn in a sound reinforment
tech/engineer/persons side. With your amps turned to 11 and screaming
"MORE MONITOR, I can't hear myself sing, whaaaaaaaaaaaaaa." Stinking
crybabies.
>
> > stuff on the side for a few years.
>
> On the side. You don't say.
So you set up and design SR systems for a living? Didn't think so.
Wow, you designed a microphone pre-amp for Studiomaster console or
something a million years ago.. and they are still in business making
some of the best consoles ever. Hahahahahaha.
>
> >The biggest advantage I have that
> > trumps you is that I've got a brain that actually works.
>
> Oh...now down to the insults. I see from this that you now admit that
> you lost the argument. Especially, since you have now sniped all my
> technical content, as you are unable to refute any of it.
The only thing I admit is that you're brain is non-functional any
longer. Probably damaged from that loud guitar thing. I chose not to
include your technical stuff because it's nonsense.
>
> >
> > I've met lots of 25 year engineers that don't have a clue what they
> > are doing. I'll just add another to my list.
>
> ROTFLMAO
Glad to see the humor part of the brain is still working. Although,
that wasn't funny to anyone but you. Maybe what's funny is you
realising how washed up you really are.
>
> >
> > I refuse to carry on this argument any longer. Nothing you say
will
> > convince me otherwise
>
> I sure it won't. You dont have the nessesary background to understand
> the arguments.
This stuff isn't rocket science grand-pa. Maybe it was in 1932 when
you were born.
>
> > You've chosen to describe methods that are non-standard and won't
> > offer ther results most ppl in this business would expect.
>
> Yeah, sure...
We agree on something.
>
> > I do not
> > doubt that any PA system that you would set up would not perform
> > optimally and/or would still hiss like an asp.
>
> Strange, as my stuff dosnt do this.
Maybe because your aged ears can't hear PA hiss any longer. How sad.
LOL!
>
> Next time I would like a software engineer (and I use the term
engineer
> here loosely) to give me some advice on PA setting up, I'll let you
> know, but don't hold your breath.
A software engineer is every bit as much an engineer as any other
profession. The software *I* design is so infinitely complex that mere
mortals such as yourself can not always comprehend it. The software
*you* design makes monkey's look intelligent.
You're website is hilarious by the way.
Phildo
"Mike Dobony" <sw...@asarian-hostspamnot.net> wrote in message
news:d3fam4$b78$1...@news.netins.net...
>
> "mlcguy" <vthor...@comcast.net> wrote in message
> news:1113084880.0...@g14g2000cwa.googlegroups.com...
>> What causes PA hiss? Is there anyway to control it or reduce it?
>>
>
> Quite a long list of things can cause hiss. Here is a minute list.
>
> Bad mic cables
> bad connectors
> running cables next to power lines
> running cables along steel beams
> poor gain structure
> defective mics
> defective hardware
> Reversed phase outlets (hot and neutral reversed)
> using unbalanced cables to connect equipment
> equipment plugged in to mixed phases, especially with cheap equipment or
> blown filters
> bad speakers, specifically the crossover
> neon or florescent lights, especially with a bad ballast
> any faulty equipment on the same circuit, not necessarily your sound gear
>
> and the list goes on and on and on
Yet again Dobby shows how little he knows about sound. Three of those will
possibly cause hiss. The rest will not. Let's see if you can work it out.
Phildo
Phildo
> Surely *Monster* make a hiss filter ? ;-)
>
Phildo
Zigakly
> 20dB> sensitivity then needed. That's why many people recommend thaty
> attenuate the amplifier's input. This allows the entire signal chain to
> run at optimum, and allows the console and amplifier to have the same
> clip point.
Who the hell plugs a mixer straight into an amp these days? If you don't
have an EQ, speaker processor, and/or crossover between the mixer and the
amps, hiss is hardly a primary concern.
My assertion that the EQ is the most common source of hiss was under the
assumption that we're talking about properly configured systems here. If
you take any cheap analog 31-band EQ, turn the input up and output down, the
hiss level drops. Use the EQ output as the main volume, keep the mixer
output up. Under those circumstances you want the amp gain knobs pinned
wide open to keep the EQ output as low as possible. I know what the specs
say, but IME analog EQ's have high self-noise. Rane, Ashly, and
Klark-Teknik EQ's fare well in that regard, but DBX ones are no better than
ART's.
I guess I should have said 90% of the time I hear significant hiss I find a
cheap 31-band EQ in the rack. I've never had a hiss problem other than what
that trick fixes.
Zigakly
"Kevin Aylward" <see_w...@anasoft.co.uk> wrote in message
news:UON6e.56218$C12....@fe1.news.blueyonder.co.uk...
> Pooh Bear wrote:
> > Kevin Aylward wrote:
> >
> >> Zigakly wrote:
> >>>> What causes PA hiss? Is there anyway to control it or reduce it?
> >>>
> >>> 9 times out of 10 it's the EQ.
> >>
> >> Very unlikely. The EQ is usually after the initial gain stage. In
> >> which case the input referred noise of the EQ will be divided by the
> >> gain of that stage, say 40db.
> >
> > That's dB actually ! Go on - call me a pedant !
> >
> > You've never heard a noisy EQ section ?
>
> Yes. 20 years ago:-)
>
> >I assure you that they exist.
>
> I am talking about competently designed ones, I don't use Realisic
> mixers :-)
>
> Note, I am still focusing on the thread question. For me, "PA hiss",
> implicitly implies mic to mixer to power amp.
Now THAT'S 20 years ago. I was refering to the 31-band EQ commonly found
between the mixer and the crossover and/or amp(s), not the EQ section of a
mixer.
Bob Quintal
> Bob Quintal wrote:
>> Let us take an hypothetical example. say a mixer generates a
>> constant 0.228 mVRMS of hiss voltage,
>
> This seems a bit large for a competently designed mixer. The
> main output fader is usually at or very near the output, maybe
> an extra 10db max fixed gain after the fader. This fader is in
> the feedback loop. The fader will reduce the noise when fully
> down of all preceding stages and its own stage. The keep warm
> output noise of the output opamp, if there, will be of the
> order of 20nv*141*gain=15uV. This is 96db down on a 1V signal.
>
It is 80dB down from the nominal output. Not quite your quoted
96dB, but you forgot a few things when calculating your noise
floor.
> This is approximate worst case. Many mixers have no final
> gain, in which case the output noise will be a few uV.
Depends totally on the opamp used, the bias applied to the
opamp,other circuit components, etc., Real life is a lot more
complex than you admit.
>> irrespective of output
>> signal, in addition to any hiss amplified by the device. The
>> mixer has a +4dBm nominal output level with 24 dB of headroom
>> to clipping. It drives an amplifier that is set to produce
>> full output with a +4 dBm signal.
>>
>> With those facts, if we reduce the acoustic output of the
>> speakers by 20 dB, we can reduce the input sensitivity of the
>> amplifier by that amount, or the mixer output by that amount.
>>
>> What about S/N ratios in the two options?
>
> None of this is relevant. What *dominates* the noise is the
> mic stage. This is such a simple concept I am at a loss as to
> what the issue is in understanding it.
Just bear with me a minute: No mic stage hiss is passing the
fader when it is at maximum attenuation. Every stage in a cascade
of circuits adds some noise to the output.
>
> Each noise souce must be appropriately added. If there is one
> source that dominates, then we can effectively ignore the
> rest.
True on a very broad level. And more specifically, any subsequent
electronics downstream will add to the noise
>
> Sure, if a later stage is *dreadfully* noisy it could be
> significant, but I don't see how this can occur for the output
> amp, knowing what the noise of typical opamp is, and the noise
> of mic amps.
What about the shot noise of the resistors and capacitors in the
circuit?.
In most of the low-noise circuits I've had the obligation to
analyse, They contribute much more than the op-amp.
Even in a proper P-Spice and Venable analysis.
>>
>> since the hiss voltage from the output of the mixer is
>> constant, reducing the output of the mixer reduces the ratio
>> by the 20 dB of gain reduction. Turning down the amplifier's
>> gain maintains the ratio at whatever it was.
>
> Sure, there is a constant noise at various parts of the chain,
> but this noise is usually way way less than the noise
> generated by the mic stage so we can ignore it.
>
BS. the noise of a preamp is approximately the same as that of a
final output buffer. What changes is the amount of amplification
in the cascade following that stage. Preamps being the first in
the cascade have their noise amplified by the channel output
driver, by the summing matrix, the output buffer,
AND ADD TO THAT the fact that one usually has 8,12,16,24 pre-
amplifiers contributing their noise to a summing matrix, which
sums the noise exactly as it is designed to do.
>
>> Why did I specify the 0.228 mV noise output? 1st it is close
>> to real post master fader noise levels of several mixers I've
>> measured, and it's 80dB down from nominal input.
>
> I would suggest that you have measured incorrectly. Probably
> measuring hum.
No, just shot noise. As confirmed by my trusty HP wave analyser.
>
> Fact. I just now turned down the faders pots on my mackie
> mixer (and studiomaster as a 2nd check), with my mackie pa amp
> gain full up. The hiss is completly insignificant when the
> faders are down. And oh, dear, and I do this with my 15 year
> old Tascam 8 track driving MOSFET 1000, a bit more hiss, but
> still insignificant. In all case the hiss is way less then the
> fan noise!
>
Fan noise is not an issue. Amps are in a rack case backstage,
What an old fart with 10dB of hearing loss at frequencies above
5KHz may consider insignificant may be considered blatant to a 16
year-old.
What about outboard processing: EQ, Limiting, Frequency division?
Each of these adds white noise to the signal.
And how good are the speakers? All these things need to be
considered in any objective test. You are not being objective
elsewhere, so I suspect your objectivity here.
> Are you using Realistic/Tandy gear by chance?
Dont knock Rat Shack gear. Some (most) of it is better than any
of the StudioMaster boards I've used.
>>
>> If the maximum SPL from the speakers is 120dBa, that means
>> that the 20 dB reduction reduces the sound level to 100 dBa,
>> a comfortable (rock and roll venue) level. Doing it by
>> turning down the amplifier means that the 80dB Signal to hiss
>> ratio is maintained giving 20 dBa of hiss. That is barely
>> noticable during silences. But turning down the amp faders
>> means that there is 40 dba of hiss coming out of the speakers
>> when there is no signal, and most people can hear that quite
>> well.
>
> This is nonsense.
Not in the least.
>
> With the mics *on* the total noise at the output will be the
> rms sum of mic_gain*mic_input noise and the post fader noise.
> The post fader noise will be insignificant as the typical mic
> gain from that point may be say 50db-60db.
>
No argument with your numbers, but you haven't got a clue as to
what they mean.
> I suggest you actually go and do the test. Keep the total gain
> constant with and without the pa amp backed of to -20db. If
> you notice a difference, you have shit gear.
Total lack of objectivity here. I can hear a 20dB change. Few
rock guitarists can.
>>
>> Add to that, say, 6 to 10 dB of hiss from preamps and sources
>> such as keyboards, Effects Units etc, and you have evident
>> hiss coming out of the speakers.
>> .
>>>
>>>> 20dB encroachment of amplfier
>>>> sensitivity on S/N ratio.
>>>
>>> It doesn't effect the s/n ratio, as far as hiss goes of the
>>> system at all. The fact that the gain at the amp may be
>>> reduced by 20db, does not mean that the s/n is so effected.
>>
>> Read above. It dows too. absolutely, and unequivocably.
>
> Unequivocally not, unless your output stage has an equivalent
> input noise that is truly horrendous.
>
>>That's
>> the law (of physics).
>
> Yes its simple physics. RMS Add each noise source divided by
> the gain up to that noise source.
>
>> The noise level of the amp itself isn't the problem, but the
>> amp's available gain over the required gain does contribute
>> dB for dB to the audible level of hiss coming out of the
>> speaker connected to that amp.
>
> Not in practice it don't. You should really get rid of that
> Tandy kit mate.
>
I obviously have a lot more practice than you do.
Pooh Bear
Phildo wrote:
Wot ? Meeee ? ;-)
Graham
shannon
>
>
> THAT is not the reason to turn down amplifier gain. The reason is with
> the pots up full, the amplifier will amplify all noise in signal chain
> by it's fixed amount of gain, period. Additionally, if the noise floor
> is high, the amps will in turn amplify this to audible levels.
>
> So turning them down, does fix hiss problems, AND it achieves unity
> gain. YOu seem to think the issues are separate, but they are not.
> You need to achieve both goals (no hiss and unity gain). Leaving amps
> on full while it will achieve the first (if the preceeding equipment is
> attenuated) is not the way to acheive the second goal.
>
>
Its not really practical for an arena rig
We run all the amps at a fixed voltage gain so that the relative levels
of crossover bands remains constant, and calibration of amp pots isn't
an issue.
If the amps aren't contributing noise, then as long as the previous link
in the chain is also noise free, controlling the total system gain at
that point will also attenuate the source noise floor as much as
reducing the amp gains.
I run consoles at their design gain structure where adjusting a PFL
level according to the console meter means something, then I use the
matrix output to adjust the nominal SPL I am going to mix at.
If the console doesnt have a matrix, I'll use the input trim of a
limiter or eq, or as last resort the crossover output gains
With my preferred rig I tweak the DSPs in the amp racks with a FOH laptop.
The crossover indicators tell me what headroom I have with fixed amp
gain, and the console meters tell me how I am doing relative to my
nominal SPL.
Any hiss is usually a source
Phil Allison
"Gene Sweeny"
> New argument...
( snip)
>
> The only thing I admit is that you're brain is non-functional any
> longer. Probably damaged from that loud guitar thing. I chose not to
> include your technical stuff because it's nonsense.
>
** Well, I did make virtually the same observations a few days ago.
Kev has zero first hand knowledge of live sound reinforcement systems and is
arguing from a position of ignorance and a some major false assumptions. The
number 1 fallacy is that mic pre-amp noise is the dominant source of "hiss"
in such systems when that rarely is the case.
The reason is that the maximum signal level arriving from mics used in live,
popular music is much higher than he has naively assumed - mic pre-amps
only supplying only 15 to 25 dB of gain in most situations. This is
particularly so if condenser mics are used but even mics like the SM58 will
output 100mV to 500 mV in the hands of a strong voiced singer or parked in
front of an instrument amp speaker box.
At such low gain settings, mic pre amp noise falls to its residual level and
hence is swamped by noise generated in the all the succeeding stages of
virtual earth summing, eqing and processing.
In some cases, the effective voltage gain from the XLR mic input on a desk
(when being used for live rock music ) through to the input on a 1kW per
channel power amp at the end of the chain may be as low as unity.
........... Phil
Pooh Bear
Hi Phil,
you made the very point I was about to myself. So I'll just second it.
I don't recall ever using more than 40dB of gain on any mic in a live situation
( quiet ballady stuff ). 30 - 35 dB is probably more typical and sometimes 20dB
is enough !
As you say, the mic amp's contribution is miniscule at these gains.
What you *do* have is a system that will likely produce about 130dB ( acoustic )
close to the stacks. An 80dB s/n ratio ( not too bad really ) means 50dB of
hiss. You're sure going to hear that.
I've seen *Volts* come out of an AKG D190E ( specifically that was the one being
demonstrated ) used close up.
Graham
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in message
> news:UON6e.56218$C12....@fe1.news.blueyonder.co.uk...
>> Pooh Bear wrote:
>>> Kevin Aylward wrote:
>>>
>>>> Zigakly wrote:
>>>>>> What causes PA hiss? Is there anyway to control it or reduce it?
>>>>>
>>>>> 9 times out of 10 it's the EQ.
>>>>
>>>> Very unlikely. The EQ is usually after the initial gain stage. In
>>>> which case the input referred noise of the EQ will be divided by
>>>> the gain of that stage, say 40db.
>>>
>>> That's dB actually ! Go on - call me a pedant !
>>>
>>> You've never heard a noisy EQ section ?
>>
>> Yes. 20 years ago:-)
>>
>>> I assure you that they exist.
>>
>> I am talking about competently designed ones, I don't use Realisic
>> mixers :-)
>>
>> Note, I am still focusing on the thread question. For me, "PA hiss",
>> implicitly implies mic to mixer to power amp.
>
> Now THAT'S 20 years ago.
Oh...
> I was refering to the 31-band EQ commonly
> found between the mixer and the crossover and/or amp(s),
But not as common as seems to be implied here, imo. There seems to be
this "This is what pro sound is" implying that what major acts do is
standard for the small timers. I would say the majority of systems
simply don't do this. The majority are the soloists, dues, trios, small
bands playing in small bars and clubs, that essentially have a mixer and
amp, or a powered mixer, or powered speakers. I would be surprised if
more than 5% of actual working acts use 31-band EQ in line with a desk.
Its just another bit of agro that is hardly worth the effort.
> not the EQ
> section of a mixer.
The first thing I should mention here, is that I am addressing the
orginal question:
"What causes PA hiss? Is there anyway to control it or reduce it?"
I don't read this as, "what bad bit of kit is most likely to cause
noise". The assumption is that one has reasonable gear as a given. If
the gear is bad, then the answer is buy better gear. I don't see that
this solution is the intent of the original question, although maybe it
was.
Since by assumption, the input to the eq is going to be at the 1v level,
if the eq is dominating the noise, then it has to be a pretty bad eq.
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in
> news:PK57e.62372$C12....@fe1.news.blueyonder.co.uk:
>
>> Bob Quintal wrote:
>
>>> Let us take an hypothetical example. say a mixer generates a
>>> constant 0.228 mVRMS of hiss voltage,
>>
>> This seems a bit large for a competently designed mixer. The
>> main output fader is usually at or very near the output, maybe
>> an extra 10db max fixed gain after the fader. This fader is in
>> the feedback loop. The fader will reduce the noise when fully
>> down of all preceding stages and its own stage. The keep warm
>> output noise of the output opamp, if there, will be of the
>> order of 20nv*141*gain=15uV. This is 96db down on a 1V signal.
>>
> It is 80dB down from the nominal output. Not quite your quoted
> 96dB,
Oh?
s/n = 20.log(1V/15uv) = -96.48dB.
> but you forgot a few things when calculating your noise
> floor.
Well, at least I didn't forget how to use a calculator.
>
>> This is approximate worst case. Many mixers have no final
>> gain, in which case the output noise will be a few uV.
>
> Depends totally on the opamp used,
Sure, but name me one op amp that is used in audio eqipment that has
noise as large as say, 50nV/rthz.
>the bias applied to the
> opamp,
Pardon?
Op amp bias doesn't effect noise. I suspect that you aint no analogue
design engineer.
>other circuit components, etc., Real life is a lot more
> complex than you admit.
Sure, but I have made an upper bound estimate accounting for all of
these factors. Op amp noise typically varies from 1nv to 30nv. A 100ohm
resister makes about 1.3nv. Plug it all in, and worst case, and e.g., if
there is no gain post fader, the noise will only be a few uv. end of
story.
>
>>> irrespective of output
>>> signal, in addition to any hiss amplified by the device. The
>>> mixer has a +4dBm nominal output level with 24 dB of headroom
>>> to clipping. It drives an amplifier that is set to produce
>>> full output with a +4 dBm signal.
>>>
>>> With those facts, if we reduce the acoustic output of the
>>> speakers by 20 dB, we can reduce the input sensitivity of the
>>> amplifier by that amount, or the mixer output by that amount.
>>>
>>> What about S/N ratios in the two options?
>>
>> None of this is relevant. What *dominates* the noise is the
>> mic stage. This is such a simple concept I am at a loss as to
>> what the issue is in understanding it.
>
> Just bear with me a minute: No mic stage hiss is passing the
> fader when it is at maximum attenuation. Every stage in a cascade
> of circuits adds some noise to the output.
Yes. What your point?
>
>>
>> Each noise souce must be appropriately added. If there is one
>> source that dominates, then we can effectively ignore the
>> rest.
>
> True on a very broad level. And more specifically, any subsequent
> electronics downstream will add to the noise
Sure, on an rms basis, divided by the gain at that point to refer that
noise to the input.
>
>>
>> Sure, if a later stage is *dreadfully* noisy it could be
>> significant, but I don't see how this can occur for the output
>> amp, knowing what the noise of typical opamp is, and the noise
>> of mic amps.
>
> What about the shot noise of the resistors and capacitors in the
> circuit?.
Oh dear...confirmed, no you aint no qualified analogue engineer.
Resisters and capacitors do not generate shot noise. Shot noise is due
to the statistical variation of electrons crossing a potential barrier.
There are no potential barriers in capacitors or resisters.
> In most of the low-noise circuits I've had the obligation to
> analyse, They contribute much more than the op-amp.
Sure, the mic stage is actually dominated by the resistance of the mic,
not the amp.
Lets say we hang a large 100k resister on an opamp. Its another
40nv/rthz. Its still only around 5uv. At the 1v signal level,
insignificant.
>
> Even in a proper P-Spice
Oh dear...there are more user friendly spices, which are a dam bit
cheaper:-)
>and Venable analysis.
No idea what this is.
>
>>>
>>> since the hiss voltage from the output of the mixer is
>>> constant, reducing the output of the mixer reduces the ratio
>>> by the 20 dB of gain reduction. Turning down the amplifier's
>>> gain maintains the ratio at whatever it was.
>>
>> Sure, there is a constant noise at various parts of the chain,
>> but this noise is usually way way less than the noise
>> generated by the mic stage so we can ignore it.
>>
> BS. the noise of a preamp is approximately the same as that of a
> final output buffer.
Actually, no it isn't always, but you misunderstood my statement. See
after this for clarification.
Mic preamp noise is typically 1nv-3nv/rt hz. Typical op amps are at the
10-20nv. Sure, there are opamps at the mixer summer that sometimes have
around 1nv-2nv, but I doubt if the output devices and tone controls have
such ones. Its a cost thing. Additionally, they have high current noise
so not that useful for eq sections that use higher value resisters.
>What changes is the amount of amplification
> in the cascade following that stage. Preamps being the first in
> the cascade have their noise amplified by the channel output
> driver, by the summing matrix, the output buffer,
Yes, and that was what was *obviously* meant.
I am discussing evething as referred to the input implicitly. The
assumption is always that the mic noise is going to be amplified by the
gain when it is compared to later stages.
>
> AND ADD TO THAT the fact that one usually has 8,12,16,24 pre-
> amplifiers contributing their noise to a summing matrix, which
> sums the noise exactly as it is designed to do.
>>
>>> Why did I specify the 0.228 mV noise output? 1st it is close
>>> to real post master fader noise levels of several mixers I've
>>> measured, and it's 80dB down from nominal input.
>>
>> I would suggest that you have measured incorrectly. Probably
>> measuring hum.
>
> No, just shot noise. As confirmed by my trusty HP wave analyser.
Oh dear...it now so obvious that you are not a qualified electronics
engineer, so your measurements mean dddly squat.
Look, dude, you have obviously picked up this "shot noise" thing from
the back of an envelope. The dominate noise of an optimally designed
system is not shot noise, but the *thermal* noise of resisters and
active devices. Most op amps in a mixer are probably fet input ones,
which have essentially, no shot noise.
>
>>
>> Fact. I just now turned down the faders pots on my mackie
>> mixer (and studiomaster as a 2nd check), with my mackie pa amp
>> gain full up. The hiss is completly insignificant when the
>> faders are down. And oh, dear, and I do this with my 15 year
>> old Tascam 8 track driving MOSFET 1000, a bit more hiss, but
>> still insignificant. In all case the hiss is way less then the
>> fan noise!
>>
> Fan noise is not an issue.
This was for just for reference. The point being that the noise is not
relevant.
>Amps are in a rack case backstage,
Sometimes. 1000s and 1000s of duos, trios, bands playing in bars and
clubs have a powered mixers on stage right in front of them. I would say
only 5% of working musicians have this "pro audio" big mixer running on
a multi core to the back of the venue sort of thing. I go wow, whenever
I happen to see such a set-up in my weekly trips to bars. Its just so
unusual. So many bars are too small for all that gear.
>
> What an old fart with 10dB of hearing loss at frequencies above
> 5KHz may consider insignificant may be considered blatant to a 16
> year-old.
And you point would be?
>
> What about outboard processing: EQ, Limiting, Frequency division?
> Each of these adds white noise to the signal.
Sure. Now divide their noise by the gain up that point, e.g. 50db.
>
> And how good are the speakers? All these things need to be
> considered in any objective test. You are not being objective
> elsewhere, so I suspect your objectivity here.
Oh dear...
>
>> Are you using Realistic/Tandy gear by chance?
>
> Dont knock Rat Shack gear. Some (most) of it is better than any
> of the StudioMaster boards I've used.
Nothing I have used from Tandy was quiet, well except when it stopped
working.
I'll leave the refutation of the rest to Graham:-)
>
>>
>> With the mics *on* the total noise at the output will be the
>> rms sum of mic_gain*mic_input noise and the post fader noise.
>> The post fader noise will be insignificant as the typical mic
>> gain from that point may be say 50db-60db.
>>
> No argument with your numbers, but you haven't got a clue as to
> what they mean.
And this is from a dude that thinks capacitors and resistors have shot
noise. Look dude, like many here, you are simply out of your depth.
Hey, if you want a short course on electronics, try here
http://www.anasoft.co.uk/EE/index.html, you surely need it.
>
>> I suggest you actually go and do the test. Keep the total gain
>> constant with and without the pa amp backed of to -20db. If
>> you notice a difference, you have shit gear.
>
> Total lack of objectivity here.
Oh dear, this is hardly rocket science. So I might be off by 6db, so
what. The point remains the same.
>I can hear a 20dB change. Few
> rock guitarists can.
I am not a rock guitarist.
>>
>>> The noise level of the amp itself isn't the problem, but the
>>> amp's available gain over the required gain does contribute
>>> dB for dB to the audible level of hiss coming out of the
>>> speaker connected to that amp.
>>
>> Not in practice it don't. You should really get rid of that
>> Tandy kit mate.
>>
> I obviously have a lot more practice than you do.
Certainly not in electronics or audio, maybe in basket weaving you do.
Kevin Aylward
> "Gene Sweeny"
>
>> New argument...
> ( snip)
>>
>> The only thing I admit is that you're brain is non-functional any
>> longer. Probably damaged from that loud guitar thing. I chose not
>> to include your technical stuff because it's nonsense.
>>
>
>
> ** Well, I did make virtually the same observations a few days ago.
Yeah, the blind leading the blind.
>
> Kev has zero first hand knowledge of live sound reinforcement systems
Yeah sure. What part of playing for 35 years did you fail to understand?
http://www.anasoft.co.uk/band/equipment.htm
Proved wrong again Phil, Hint: I got the Studimatser 15 years ago.
> and is arguing from a position of ignorance and a some major false
> assumptions. The number 1 fallacy is that mic pre-amp noise is the
> dominant source of "hiss" in such systems when that rarely is the
> case.
Of course it is. Its trivial to test. I do it *every* day. Seriously.
Its a no brainer mate. If I have the mic channels off on my mixer, and
the faders at 0VU, the total amp noise is less than the power amp fan.
Its not debatable dude. I just did it. Dah....
Your simply clueless. Go and try it before opening you mouth and putting
you foot in it.
It clear just who is arguing form ignorance. Any fool can go and the
test.
>
> The reason is that the maximum signal level arriving from mics used
> in live, popular music is much higher than he has naively assumed -
> mic pre-amps only supplying only 15 to 25 dB of gain in most
> situations.
Well, I just checked where mine are set, and they are around 40db. This
"most" thing is quite nonsense. If this was the case why even bother
having 50db-60db mic gains available?
>This is particularly so if condenser mics
Oh... One of my capacitor mics needs around 45db at least.
Secondly, there is no such thing as a "condenser mic". A condenser is
something one uses in fridges.
>are used but
> even mics like the SM58 will output 100mV to 500 mV in the hands of
> a strong voiced singer or parked in front of an instrument amp
> speaker box.
The dominant noise source is the weakest link in the chain. Lots of
singers, especially female ones, don't eat the mic, or sing that loudly.
If just *one* mic needs say, 10db more gain then say an instrument mic,
it will completely dominate the noise of all mics, i.e it is 5 times
more nosier then any other source.
>
> At such low gain settings, mic pre amp noise falls to its residual
> level and hence is swamped by noise generated in the all the
> succeeding stages of virtual earth summing, eqing and processing.
Sure, at low gain settings this would be the case, and the succeeding
stages would indeed be significant in principle. Like this is new to me?
However, this case is completely irrelevant in practise to the posters
question, as in this case, as the gain is so low, so is the final output
noise. Dah...There is no relevant noise to complain about for PA
applications. Sure, in recording one might want a bit better, but for a
pub PA, no way Jose.
Look, dude, Its irrelevant what the dominant noise source is if the hiss
to too low to bother about.
And yes, I just checked, with low mic gain, and line gains typically
adjusted for my brothers 5 keyboards, my midi sound sequencer etc, for
full output, the noise was of no practical concern. You should try it
sometime. It removes one ignorance.
>
> In some cases, the effective voltage gain from the XLR mic input on a
> desk (when being used for live rock music ) through to the input on a
> 1kW per channel power amp at the end of the chain may be as low as
> unity.
I don't care about special extreme cases. I care about typical, common
occurrence, e.g 40db of mic gain, and more on some. Even at 30db gain,
its 30 times the basic noise of the mic and its amp. Again, as I noted,
at low gains, there is not much noise to complain about. Been there.
Every day.
Steve White
news:SCq7e.66554$C12....@fe1.news.blueyonder.co.uk...
> >I can hear a 20dB change. Few
> > rock guitarists can.
>
> I am not a rock guitarist.
http://www.tribute-to-blondie.pwp.blueyonder.co.uk/status.htm
"Kevin is very much interested in joining an existing Glam Rock band,"
??????
Steve W
Phil Allison
"Kevin Aylward"
** Is your wife still alive Kev ??
.............. Phil
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in message
> news:SCq7e.66554$C12....@fe1.news.blueyonder.co.uk...
>>> I can hear a 20dB change. Few
>>> rock guitarists can.
>>
>> I am not a rock guitarist.
>
>
> "Kevin is very much interested in joining an existing Glam Rock band,"
>
That don't make me a "rock" guitarist. Sure, I play many styles, some of
which is "rock", but for me Glam Rock is really "pop". You can't really
class it along with say, Led Zeppelin and Black Sabbath etc. In the
contest of the thread, the use of the word "rock" appears to be more of
an implication to me playing heavy rock/metal etc, loud with massive
crunch. I absolutely detest that sort of stuff. I align myself with the
likes of Steely Dan and Santana, and while Santana might be associated
with the word "rock" I don't think of his music that way.
The demo I have is more of my style
http://www.anasoft.co.uk/band/media/NewBeginning.mp3
Steve White
> >> I am not a rock guitarist.
> >
> > http://www.anasoft.co.uk/band/index.htm
> > "Kevin is very much interested in joining an existing Glam Rock band,"
>
> That don't make me a "rock" guitarist. Sure, I play many styles, some of
> which is "rock", but for me Glam Rock is really "pop". You can't really
> class it along with say, Led Zeppelin and Black Sabbath etc. In the
> contest of the thread, the use of the word "rock" appears to be more of
> an implication to me playing heavy rock/metal etc, loud with massive
> crunch. I absolutely detest that sort of stuff. I align myself with the
> likes of Steely Dan and Santana, and while Santana might be associated
> with the word "rock" I don't think of his music that way.
And from what you've said I guess you'll *hate* stuff like this.
http://www.sjwhite.plus.com/mp3s/toccata_lp_steve_white.mp3
Although this is more representative of what I really get up to
http://www.sjwhite.plus.com/mp3s/SteveWhite_AuldLangSine.mp3
> The demo I have is more of my style
> http://www.anasoft.co.uk/band/media/NewBeginning.mp3
Which is playing now.
I see you're a Pat Metheny fan. Me too - and Blondie. Clem Burke was one of
my drum idols. I caught them last year and he was superb. Debbie Harry was
... erm .... how do I say this ... off her t*ts.
Steve W
Kevin Aylward
> "Gene Sweeny"
>
>> New argument...
> ( snip)
>>
>> The only thing I admit is that you're brain is non-functional any
>> longer. Probably damaged from that loud guitar thing. I chose not
>> to include your technical stuff because it's nonsense.
>>
>
>
> ** Well, I did make virtually the same observations a few days ago.
>
> Kev has zero first hand knowledge of live sound reinforcement systems
> and is arguing from a position of ignorance and a some major false
> assumptions. The number 1 fallacy is that mic pre-amp noise is the
> dominant source of "hiss" in such systems when that rarely is the
> case.
> The reason is that the maximum signal level arriving from mics used
> in live, popular music is much higher than he has naively assumed -
> mic pre-amps only supplying only 15 to 25 dB of gain in most
> situations. This is particularly so if condenser mics are used
And oh dear oh dear....you just don't see the *global* picture. There is
no such thing as a free lunch. More output from the source means more
noise from the source. Its sods law.
First capacitor mic I looked at
http://www.shure.com/microphones/recording/ksm/default.asp
KSM141 -37db/Pa @ 94db spl (speaking 1 inch away), i.e. 14mv nominal.
s/n = 80db
Translates to 10nv/rthz noise.
With 20db of front end gain this is 100nv at the line level stage. Since
op-amps are at the 16nv (TLO71) range, once again the mic stage
dominates. For the post mixer outboard kit, there is going to be say,
probably 10db gain around or post fader, making outboard stuff totally
insignificant.
What you dudes are trying to claim is effectively, that unless one gets
10V driving the output graphic, there will be unacceptable noise. This
just aint on mate.
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in message
> news:Zzt7e.67391$C12....@fe1.news.blueyonder.co.uk...
>>>> I am not a rock guitarist.
>>>
>>> http://www.anasoft.co.uk/band/index.htm
>>> "Kevin is very much interested in joining an existing Glam Rock
>>> band,"
>>
>> That don't make me a "rock" guitarist. Sure, I play many styles,
>> some of which is "rock", but for me Glam Rock is really "pop". You
>> can't really class it along with say, Led Zeppelin and Black Sabbath
>> etc. In the contest of the thread, the use of the word "rock"
>> appears to be more of an implication to me playing heavy rock/metal
>> etc, loud with massive crunch. I absolutely detest that sort of
>> stuff. I align myself with the likes of Steely Dan and Santana, and
>> while Santana might be associated with the word "rock" I don't think
>> of his music that way.
>
> And from what you've said I guess you'll *hate* stuff like this.
> http://www.sjwhite.plus.com/mp3s/toccata_lp_steve_white.mp3
Tocca and Fugue in Dm by Bach. I sometimes play bits of this. Its not a
bad tune.
John Williams (the classical one) used to do it in his band Sky. I do a
reasonable decent version of Cavatina myself, well bits of it anyway.
>
> Although this is more representative of what I really get up to
> http://www.sjwhite.plus.com/mp3s/SteveWhite_AuldLangSine.mp3
Playings quite neat. I hate the song though.
>
>
>> The demo I have is more of my style
>> http://www.anasoft.co.uk/band/media/NewBeginning.mp3
>
> Which is playing now.
>
> I see you're a Pat Metheny fan. Me too - and Blondie. Clem Burke was
> one of my drum idols. I caught them last year and he was superb.
> Debbie Harry was ... erm .... how do I say this ... off her t*ts.
>
Yep.
Steve White
> Tocca and Fugue in Dm by Bach. I sometimes play bits of this. Its not a
> bad tune.
> John Williams (the classical one) used to do it in his band Sky.
... which is where I lifted it from.
> I do a reasonable decent version
> of Cavatina myself, well bits of it anyway.
Snap. I spent a whole week wearing out a video of Mr. Williams playing this
at The Secret Policeman's Ball ... a good few years before I could afford
the price of the sheet music ... and long before the internet was in full
flight :-)
> > Although this is more representative of what I really get up to
> > http://www.sjwhite.plus.com/mp3s/SteveWhite_AuldLangSine.mp3
>
> Playings quite neat. I hate the song though.
Not my favourite either but I play in a ceilidh band so it goes with the
territory. This was a recording put together specially for a christmas
charity compilation CD on uk.music.guitar. The only thing "not me" is the
bagpipes.
Steve W
Gene Sweeny
Kevin Aylward wrote:
>
> What you dudes are trying to claim is effectively, that unless one
gets
> 10V driving the output graphic, there will be unacceptable noise.
This
> just aint on mate.
No, that's not true.
Another example, mic -> mixer -> graphic -> amp.
graphic has a peak of +18 and s/n ratio of 97 dBu. Noise floor of
graphic is -79 then.
Forget the noise coming from the mixer for a moment. For every dBu I
attenuate the output of the mixer, the nominal signal gets closer to
the noise floor of the graphic. So if the master faders on the mixer
are set to -20 and my VU meters show a nominal level of -20, then the
signal going into the graphic is -20dBu and the since noise floor of
the graphic is at -79 that gives us a 59dB s/n ratio. If I ran the mix
at 0 dBu on the mixer, with master faders up, the input to the graphic
would be 0 dBu, same noise floor as before now we have 79 db of s/n
ratio. 59 vs 79.
YOu can't argue that attenuating a previous device reduces the overall
s/n ratio of the system. The only exception, is if you ran an even
hotter mix (at say 0db) with the master faders still down to -20.
However, that would be damn near impossible with out clipping previous
stages somewhere (mic pre, channel fader, group fader).
Another reason not to run the master faders lower than normal... the
sensitivy is not linear near the bottom of a fader. Small movements
equate to large changes in amplitude. Bad mojo.
Phildo
> Oh...what's an FOH engineer?
For fucks sake don't ask Arny. He posted here for years without ever
figuring out what FOH meant.
Phildo
Kevin Aylward
> Kevin Aylward wrote:
>>
>> What you dudes are trying to claim is effectively, that unless one
>> gets 10V driving the output graphic, there will be unacceptable
>> noise.
> This
>> just aint on mate.
>
>
> No, that's not true.
>
> Another example, mic -> mixer -> graphic -> amp.
I dont use graphics live:-)
More trouble then they're worth. I would say the majority of working
soloists, duos, bands in pubs don't use them either.
>
> graphic has a peak of +18 and s/n ratio of 97 dBu. Noise floor of
> graphic is -79 then.
ok.
>
> Forget the noise coming from the mixer for a moment. For every dBu I
> attenuate the output of the mixer, the nominal signal gets closer to
> the noise floor of the graphic.
Yes.
>So if the master faders on the mixer
> are set to -20 and my VU meters show a nominal level of -20, then the
> signal going into the graphic is -20dBu and the since noise floor of
> the graphic is at -79 that gives us a 59dB s/n ratio.
I assume that what you are saying here is that suppose one wants to
operate the system not at full power, but at -20 from full power. Well,
sure the s/n will go down. In the limit, when the faders are fully off,
there is no signal at all!!!
I don't see that this is relevant. If the final output noise level is
acceptable at full power, the total noise at reduced power will be less,
therefore also acceptable. Its the *absolute* level of background hiss,
*not* the s/n by itself that people are complaining about.
> If I ran the
> mix at 0 dBu on the mixer, with master faders up, the input to the
> graphic would be 0 dBu, same noise floor as before now we have 79 db
> of s/n ratio. 59 vs 79.
>
> YOu can't argue that attenuating a previous device reduces the overall
> s/n ratio of the system.
I can. It depends on the net input referred noise. If the later stage
equivalent input noise, divided by the gain to the input stage is less
than 1/3 of the input stage noise, it will only increase the totally
noise by 0.37db. (sqrt(1+1/3^2))
Only if the ein of the later device is very noisy will it make a
difference. One has to do the actual numbers. At low gain it don't
matter that the s/n goes off, you wont hear the absolute value of the
noise.
>The only exception, is if you ran an even
> hotter mix (at say 0db) with the master faders still down to -20.
> However, that would be damn near impossible with out clipping previous
> stages somewhere (mic pre, channel fader, group fader).
>
> Another reason not to run the master faders lower than normal... the
> sensitivy is not linear near the bottom of a fader. Small movements
> equate to large changes in amplitude. Bad major.
I agree that from a practical point of view one likes to have the fader
somewhere midway, so one can adjust them up or down as needed. I do this
in the house. My MOSFET 1000 is, set around at around 1/3.
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in message
> news:6Mv7e.67417$C12....@fe1.news.blueyonder.co.uk...
>> Tocca and Fugue in Dm by Bach. I sometimes play bits of this. Its
>> not a bad tune.
>> John Williams (the classical one) used to do it in his band Sky.
>
> ... which is where I lifted it from.
>
>
I did see Sky in concert about 25 years ago.
>
>> I do a reasonable decent version
>> of Cavatina myself, well bits of it anyway.
>
> Snap. I spent a whole week wearing out a video of Mr. Williams
> playing this at The Secret Policeman's Ball ... a good few years
> before I could afford the price of the sheet music ... and long
> before the internet was in full flight :-)
I learnt it from the sheet music myself. It took me a full 3 weeks.
Nowadays, I find a midi file that someone has already done. I have been
working on and off on the Guns&Roses child of mine. Its got a really
interesting bit at the main solo. Really strange notes. Unfamiliar
fingering.
>
>
>
>>> Although this is more representative of what I really get up to
>>> http://www.sjwhite.plus.com/mp3s/SteveWhite_AuldLangSine.mp3
>>
>> Playings quite neat. I hate the song though.
>
> Not my favourite either but I play in a ceilidh band so it goes with
> the territory. This was a recording put together specially for a
> christmas charity compilation CD on uk.music.guitar. The only thing
> "not me" is the bagpipes.
>
Have you tried a guitar synth yet?
Zigakly
> I would be surprised if
> more than 5% of actual working acts use 31-band EQ in line with a desk.
> Its just another bit of agro that is hardly worth the effort.
Are we to assume you mean 95% of actual working acts use speakers with
perfectly flat response in rooms tuned for perfectly balanced ambience?
What I do know is that you won't convince 5% of the folks here that a
powered board and speakers constitutes a professional sound system. The
fact that many moonlighting musicians get by with such a rig without a
qualified tech doesn't mean it's of any significance to discussions here, no
more than the popularity of a Ford Focus should be reflected in a sports car
forum.
I'm not a fan of 31-band EQ's specifically, but something has to neutralize
the drivetrain to let the mixer do its job. That's where the 20-yrs-old
comment came from. Nowadays there are far more effective devices starting
around $250. And they don't hiss...
Steve White
> Nowadays, I find a midi file that someone has already done.
Yup. Sometimes the interenet is wonderful. However I find a lot of midis and
almost all tab to be very inaccurate. Sometimes still useful but not very
reliable.
> I have been working on and off
> on the Guns&Roses child of mine. Its got a really
> interesting bit at the main solo. Really strange
> notes. Unfamiliar fingering.
That's what foxed me about Cavatina - it took a lot of work to figure out
the fingerings for some of the chords in the mid-section.
> Have you tried a guitar synth yet?
Only the one on my Boss GT3. It's OK for the odd thing but not much more.
I'm quite interested in a hex pickup though, especially having seen Bill
Bailey use one on his latest DVD. I'm not really that keen on Pat Metheny's
guitar synth work though. It's OK but it leaves me a bit cold. But then I
love guitar music to sound .... welll ... guitary.
Cheers,
Steve W
Bob Quintal
> Bob Quintal wrote:
>> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in
>> news:PK57e.62372$C12....@fe1.news.blueyonder.co.uk:
>>
>>> Bob Quintal wrote:
>>
>>>> Let us take an hypothetical example. say a mixer generates
>>>> a constant 0.228 mVRMS of hiss voltage,
>>>
>>> This seems a bit large for a competently designed mixer. ...
will be of the order of 20nv*141*gain=15uV. This is
>>> 96db down on a 1V signal.
>>>
>> It is 80dB down from the nominal output. Not quite your
>> quoted 96dB,
>
> Oh?
>
> s/n = 20.log(1V/15uv) = -96.48dB.
I don't give a R@ts@$$ about your numbers. From my numbers posted
several messages ago, 20 log (0.228E-3V/2.28V) is 80dB. You are
so hung up in your hypothetical numbers you cannot accept anybody
else's.
1 Volt is the nominal output of your beloved Radio Shack consumer
gear, not pro audio equipment. I'm using voltages found in stuff
a pro-sound technician would deal with, when driven like most
techs would drive it.
>> but you forgot a few things when calculating your noise
>> floor.
>
> Well, at least I didn't forget how to use a calculator.
>
Yeah right... You got the number right, but the sign wrong.
That's not a good thing for a competent engineer.
>>
>>the bias applied to the
>> opamp,
>
> Pardon?
>
> Op amp bias doesn't effect noise. I suspect that you aint no
> analogue design engineer.
>
You suspect wrong. And you are factually wrong too, because bias
will affect nose figure.
... A 100ohm resister makes about 1.3nv. ..., the
> noise will only be a few uv. end of story.
And a 1K resistor makes 10 times as much. I see more 1K resistors
on schematics for audio gear than 100 ohm ones.
>
>>
>>>
>>> Each noise souce must be appropriately added. If there is
>>> one source that dominates, then we can effectively ignore
>>> the rest.
>>
>> True on a very broad level. And more specifically, any
>> subsequent electronics downstream will add to the noise
>
> Sure, on an rms basis, divided by the gain at that point to
> refer that noise to the input.
We are not talking EIN here, we are talking S/N.
>>>
>> What about the shot noise of the resistors and capacitors in
>> the circuit?.
>
> Oh dear...confirmed, no you aint no qualified analogue
> engineer.
>
> Resisters and capacitors do not generate shot noise. Shot
> noise is due to the statistical variation of electrons
> crossing a potential barrier. There are no potential barriers
> in capacitors or resisters.
>
A site that yours links to says otherwise:
http://www.aikenamps.com/ResistorNoise.htm
besides you keep misspelling resistor....
>> In most of the low-noise circuits I've had the obligation to
>> analyse, They contribute much more than the op-amp.
>
> Sure, the mic stage is actually dominated by the resistance of
> the mic, not the amp.
And the feedback resistors, and the bias resistors,and... and...
>> Even in a proper P-Spice
>
> Oh dear...there are more user friendly spices, which are a dam
> bit cheaper:-)
>
But not trustworthy. And not acceptable to NASA, afaik.
>
>> BS. the noise of a preamp is approximately the same as that
>> of a final output buffer.
>
> Actually, no it isn't always, but you misunderstood my
> statement. See after this for clarification.
>
...
>>What changes is the amount of amplification
>> in the cascade following that stage. Preamps being the first
>> in the cascade have their noise amplified by the channel
>> output driver, by the summing matrix, the output buffer,
>
> Yes, and that was what was *obviously* meant.
>
> I am discussing evething as referred to the input implicitly.
Which is your big mistake.
> The assumption is always that the mic noise
Microphones contribute no noise when the fader is at infinite
attenuation.
>
> Look, dude, you have obviously picked up this "shot noise"
> thing from the back of an envelope.
Yeah, the one with my paycheque in it.
The dominate noise of an
> optimally designed system is not shot noise, but the *thermal*
> noise of resisters and active devices. Most op amps in a mixer
> are probably fet input ones, which have essentially, no shot
> noise.
True, but noise is noise, none the less.
>>
>> What an old fart with 10dB of hearing loss at frequencies
>> above 5KHz may consider insignificant may be considered
>> blatant to a 16 year-old.
>
> And you point would be?
>
You are the hot-shot engineer, derive the conclusion. You
graduated in 1980, from your website. You are not 16 anymore. ..
>>
>> What about outboard processing: EQ, Limiting, Frequency
>> division? Each of these adds white noise to the signal.
>
> Sure. Now divide their noise by the gain up that point, e.g.
> 50db.
>
Why, that's totally irresponsible, and masks the true S/N we are
trying to calculate. I say again it's not EIN, it's S/N that we
are dealing with.
> Nothing I have used from Tandy was quiet, well except when it
> stopped working.
>
Maybe you should have bought some of their items, not taken them
from the rubbish tip at the factory gate ;).
> And this is from a dude that thinks capacitors and resistors
> have shot noise. Look dude, like many here, you are simply out
> of your depth.
Look to the mirror....
> Hey, if you want a short course on electronics, try here
> http://www.anasoft.co.uk/EE/index.html, you surely need it.
Yep: Despite much literature that implies other wise, a
transistor is a voltage controlled device not a current
controlled device.
That's it for me. Plonk another @$$hole.
Phil Allison
"Kevin Aylward"
> Phil Allison wrote:
>> "Gene Sweeny"
>>
>>> New argument...
>> ( snip)
>>>
>>> The only thing I admit is that you're brain is non-functional any
>>> longer. Probably damaged from that loud guitar thing. I chose not
>>> to include your technical stuff because it's nonsense.
>>>
>>
>> ** Well, I did make virtually the same observations a few days ago.
>>
>> Kev has zero first hand knowledge of live sound reinforcement systems
>> and is arguing from a position of ignorance and a some major false
>> assumptions. The number 1 fallacy is that mic pre-amp noise is the
>> dominant source of "hiss" in such systems when that rarely is the
>> case.
>> The reason is that the maximum signal level arriving from mics used
>> in live, popular music is much higher than he has naively assumed - mic
>> pre-amps only supplying only 15 to 25 dB of gain in most
>> situations. This is particularly so if condenser mics are used
>
>
> And oh dear oh dear....you just don't see the *global* picture. There is
> no such thing as a free lunch. More output from the source means more
> noise from the source. Its sods law.
** This has no connection to the point above about mic output levels and is
in any case completely wrong - as expected from the incurably insane Kev
Aylward. The s/n ratio of a mic depends on diaphragm size - larger
diaphragm mics having more output with no noise penalty.
> First capacitor mic I looked at
> http://www.shure.com/microphones/recording/ksm/default.asp
>
> KSM141 -37db/Pa @ 94db spl (speaking 1 inch away), i.e. 14mv nominal.
> s/n = 80db
** 14 mV output @ 94 dB SPL is some 20 dB greater than for an SM58 or
similar - so pre-amp gain will be set 20 dB lower for the same result.
The spec given by Shure for self noise is 14 dB SPL equivalent, *A
ighted* - which is virtually identical to an SM58 or similar used with a
low noise pre-amp.
> With 20db of front end gain this is 100nv at the line level stage. Since
> op-amps are at the 16nv (TLO71) range, once again the mic stage dominates.
** The "noise gain" of op-amps in EQ circuits and virtual earth summing
stages etc is not unity.
One of the many **idiotic fallacies** that Mad Kev is purveying.
........... Phil
Pooh Bear
Atek Rools wrote:
> Agreed, but in my example, which I find to be more the norm (mixer ->
> graphic -> amp) this situation could be remedied by leaving master
> faders at 0 and attenuating amps 20 dB. problem solved.
Alternatively you could 'pull' *all* the frequencies on the graphic.
;-)
I've seen this situation for real too !
Graham
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in message
> news:iTp7e.66306$C12....@fe1.news.blueyonder.co.uk...
>> I would be surprised if
>> more than 5% of actual working acts use 31-band EQ in line with a
>> desk. Its just another bit of agro that is hardly worth the effort.
>
> Are we to assume you mean 95% of actual working acts use speakers with
> perfectly flat response in rooms tuned for perfectly balanced
> ambience?
Nope. Like, dude, people actually bought the Sex Pistols music.
>What I do know is that you won't convince 5% of the folks
> here that a powered board and speakers constitutes a professional
> sound system.
Its what 1000s and 1000s of actual gigging musicians do. Sure, sound
engineers would like to justify their existance, but that dont change
the facts.
> The fact that many moonlighting musicians get by with
> such a rig without a qualified tech
By and large, the all do. Very, very, few working musicaians have sound
engineers. They are simply are not required for the average bar band,
due, trio, soloist etc.
The numbers are not set by the 0.01% of fully professional err..
musicians, i.e. Brittany Spears, but by 100,000s, of *semi-pro* small
time outfits who have a day job. To all an intents and purposes, there
are no "pro" musicians.
>.doesn't mean it's of any
> significance to discussions here, no more than the popularity of a
> Ford Focus should be reflected in a sports car forum.
Well, for sound engineers trying to justify the existence, probably not,
but for non technical people who what a quick and simple solution to
setting up their powered mixer I would say my approach fits the bill.
>
> I'm not a fan of 31-band EQ's specifically, but something has to
> neutralize the drivetrain to let the mixer do its job.
Not really. No one in the audience cares or notices. As long as the
singer is blond and wears a very short skirt, the sound don't matter
diddly squat. This is "Living In The Real world" - Blondie.
>That's where
> the 20-yrs-old comment came from. Nowadays there are far more
> effective devices starting around $250. And they don't hiss...
Sure. I can't see any valid reason why modern day, line level stuff
should generate objectionable hiss.
Pooh Bear
Phil Allison wrote:
> ** The "noise gain" of op-amps in EQ circuits and virtual earth summing
> stages etc is not unity.
Indeed - the noise gain of an inverting stage with unity gain is *2x* !
Graham
Phil Allison
"Pooh Bear"
** The noise gain on the virtual earth summing amp in a 32 channel desk is
32.
The noise gain of the summing stage in a typical 31 band eq is 50 - 100.
The noise gain of the "gain cell" in a FET comp/limiter is 30 - 100.
While all the above have signal gains of unity.
.......... Phil
Kevin Aylward
> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in
> news:SCq7e.66554$C12....@fe1.news.blueyonder.co.uk:
>
>> Bob Quintal wrote:
>>> "Kevin Aylward" <see_w...@anasoft.co.uk> wrote in
>>> news:PK57e.62372$C12....@fe1.news.blueyonder.co.uk:
>>>
>>>> Bob Quintal wrote:
>>>
>>>>> Let us take an hypothetical example. say a mixer generates
>>>>> a constant 0.228 mVRMS of hiss voltage,
>>>>
>>>> This seems a bit large for a competently designed mixer. ...
> will be of the order of 20nv*141*gain=15uV. This is
>>>> 96db down on a 1V signal.
>>>>
>>> It is 80dB down from the nominal output. Not quite your
>>> quoted 96dB,
>>
>> Oh?
>>
>> s/n = 20.log(1V/15uv) = -96.48dB.
>
> I don't give a R@ts@$$ about your numbers. From my numbers posted
> several messages ago, 20 log (0.228E-3V/2.28V) is 80dB. You are
> so hung up in your hypothetical numbers you cannot accept anybody
> else's.
Well, your numbers don't appear to be valid.
>
> 1 Volt is the nominal output of your beloved Radio Shack consumer
> gear, not pro audio equipment.
+4db is a pretty much of a standard for pro gear. Its 1.23V.
Sure, typically gear can give out a lot more.
> I'm using voltages found in stuff
> a pro-sound technician would deal with, when driven like most
> techs would drive it.
Yeah, I know, you deal at the tech level.
>
>>> but you forgot a few things when calculating your noise
>>> floor.
>>
>> Well, at least I didn't forget how to use a calculator.
>>
> Yeah right... You got the number right, but the sign wrong.
> That's not a good thing for a competent engineer.
>
>>>
>>> the bias applied to the
>>> opamp,
>>
>> Pardon?
>>
>> Op amp bias doesn't effect noise. I suspect that you aint no
>> analogue design engineer.
>>
> You suspect wrong.
Oh?
>And you are factually wrong too, because bias
> will affect nose figure.
No it wont.
This is getting a bit pointless. Its obvious that you are simple cluless
about electronic design to make this disscuion pointless.
Look, here's my trumpet blowing. I have been a professional, B.Sc.
degreed, design engineer for 25 years. For instance, I was Analogue IC
design team leader at Texas Instruments for 3 years, among other
positions. I designed op-amps for a living dude. I have spent 10,000's
of hours designing high transistor count analogue circuitry. I know
op-amp design inside out.
As a bacground check out my tutorial papers
http://www.anasoft.co.uk/EE/index.html
and a minor bio http://www.anasoft.co.uk/band/bio.htm
Look dude. Now what's wrong here? I think your emotions are getting in
the way of your objectivity. You have two options. Accept that I do
actually know what is indicated from the above, or keep sticking your
head in the sand and learn nothing.
You comment on bias effcting op-amp noise is simply bizzar. No
compentant analogue designer would use such a phrase. "Bias" on an
op-amp is just offsetting one of the inputs by a DC voltage. There is no
other way to "bias" an op-amp. DC offsets on op-amps don't effect it
noise one iota.
For a discrete transistor design, yes bias, as in the value of the
collector current does effect noise. For example, the typical optimim is
to set:
re = Rsource/sqrt(hfe), where re=1/40Ic
To minimise the sum of shot noise and thermal noise.
>
> ... A 100ohm resister makes about 1.3nv. ..., the
>> noise will only be a few uv. end of story.
>
> And a 1K resistor makes 10 times as much. I see more 1K resistors
> on schematics for audio gear than 100 ohm ones.
Here we go again. Look, mate, you don't understand even the basics. The
noise of a resister is thermal noise. The equation is:
Vn = sqrt(4kTBR)
k=boltzmans constant. B is the bandwidth, T is temperature.
Increasing the resistance by 10, only increses the noise by sqrt(10),
i.e. 3.16 times
With all dur respect, your way out of you depth. This is 101 EE stuff
dude.
>>>>
>>> What about the shot noise of the resistors and capacitors in
>>> the circuit?.
>>
>> Oh dear...confirmed, no you aint no qualified analogue
>> engineer.
>>
>> Resisters and capacitors do not generate shot noise. Shot
>> noise is due to the statistical variation of electrons
>> crossing a potential barrier. There are no potential barriers
>> in capacitors or resisters.
>>
> A site that yours links to says otherwise:
> http://www.aikenamps.com/ResistorNoise.htm
Yes, and its absolutely wrong. You cant expect to get good information
from any old web site. And, oh dear, its an acquaintance of mine's site.
I will email him and correct him. Randell, does not have a degree in EE
or physics.
Shot noise is given by:
In = sqrt(2Iq)
Again, this is 101 EE stuff.
This does seem to be a somewhat commen erroneos notion. I did find this
(National Instruments)
http://zone.ni.com/devzone/conceptd.nsf/webmain/8DE42E13BD089D8B86256816006545CD?OpenDocument
With a rebuttal to its shot noise statements, here:
http://zone.ni.com/devzone/dzcomments.nsf/webmain/2C5E6097649DBD5186256DFD006BC7C7?OpenDocument
It will take some time to find a proper academic reference, but trust me
in this, shot noise is due to charges crossing a potential barrier, as I
originally stated, and agreed upon by the above. Its basic physics and
is not debatable.
> besides you keep misspelling resistor....
Speeling is not a strong point of mine.
>
>>> In most of the low-noise circuits I've had the obligation to
>>> analyse, They contribute much more than the op-amp.
>>
>> Sure, the mic stage is actually dominated by the resistance of
>> the mic, not the amp.
>
> And the feedback resistors, and the bias resistors,and... and...
>
>>> Even in a proper P-Spice
>>
>> Oh dear...there are more user friendly spices, which are a dam
>> bit cheaper:-)
>>
> But not trustworthy.
Of course they are. They are all based on the Spice/XSpice. For the most
part, results are pretty much identical.
> And not acceptable to NASA, afaik.
Nonsense.
>
>>
>>> BS. the noise of a preamp is approximately the same as that
>>> of a final output buffer.
>>
>> Actually, no it isn't always, but you misunderstood my
>> statement. See after this for clarification.
>>
> ...
>
>>> What changes is the amount of amplification
>>> in the cascade following that stage. Preamps being the first
>>> in the cascade have their noise amplified by the channel
>>> output driver, by the summing matrix, the output buffer,
>>
>> Yes, and that was what was *obviously* meant.
>>
>> I am discussing evething as referred to the input implicitly.
>
> Which is your big mistake.
Oh dear. Its what we pro analogue engineers do as a mater of course. We
discuss in terms of the net input referred noise.
>
>> The assumption is always that the mic noise
>
> Microphones contribute no noise when the fader is at infinite
> attenuation.
Sure. But at low gain, noise isnt an issue.
>
>>
>> Look, dude, you have obviously picked up this "shot noise"
>> thing from the back of an envelope.
>
> Yeah, the one with my paycheque in it.
You can be wrong and still get a paycheck.
>
> The dominate noise of an
>> optimally designed system is not shot noise, but the *thermal*
>> noise of resisters and active devices. Most op amps in a mixer
>> are probably fet input ones, which have essentially, no shot
>> noise.
>
> True, but noise is noise, none the less.
But it aint shot noise, as you are cliamming.
>
>>>
>>> What an old fart with 10dB of hearing loss at frequencies
>>> above 5KHz may consider insignificant may be considered
>>> blatant to a 16 year-old.
>>
>> And you point would be?
>>
> You are the hot-shot engineer, derive the conclusion. You
> graduated in 1980, from your website. You are not 16 anymore. ..
>>>
>>> What about outboard processing: EQ, Limiting, Frequency
>>> division? Each of these adds white noise to the signal.
>>
>> Sure. Now divide their noise by the gain up that point, e.g.
>> 50db.
>>
> Why, that's totally irresponsible, and masks the true S/N we are
> trying to calculate. I say again it's not EIN, it's S/N that we
> are dealing with.
Not far as objectionable noise to the punters in a pub. They care about
the *absolute* level of hiss, not the s/n. When the music is playing,
even with as low as 40db s/n, no one will notice the noise. Its only
when the music is off that people notice the hiss.
If the system is low enough noise, i.e. no music with background hiss,
at full output, reducing the output will invariable reduce the total
output noise.
But anyway, it don't mater whether one calculate using ein or s/n , the
final answer is the same.
>
>> Nothing I have used from Tandy was quiet, well except when it
>> stopped working.
>>
> Maybe you should have bought some of their items, not taken them
> from the rubbish tip at the factory gate ;).
>
>> And this is from a dude that thinks capacitors and resistors
>> have shot noise. Look dude, like many here, you are simply out
>> of your depth.
>
> Look to the mirror....
>
>> Hey, if you want a short course on electronics, try here
>> http://www.anasoft.co.uk/EE/index.html, you surely need it.
>
> Yep: Despite much literature that implies other wise, a
> transistor is a voltage controlled device not a current
> controlled device.
>
> That's it for me. Plonk another
You've just plonked the best advice that you will receive in this
lifetime.
These papers, explains in pictorial terms the true voltage controlled
nature of a bipolar transistor. Its also 101 Physics. You have been
reading too many non academic text books.
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/info/comp/active/BiPolar/page2.html
http://www.mtmi.vu.lt/pfk/funkc_dariniai/transistor/bipolar_transistor.htm
You *are* out of your depth, and its a shame that you don't see this.
Its going to always hold you back. You don't have a degree in EE or
physics so why do you think that you understand the physics of
transistor?
Kevin Aylward
> "Kevin Aylward"
>> Phil Allison wrote:
>>> "Gene Sweeny"
>>>
>>>> New argument...
>>> ( snip)
>>>>
>>>> The only thing I admit is that you're brain is non-functional any
>>>> longer. Probably damaged from that loud guitar thing. I chose not
>>>> to include your technical stuff because it's nonsense.
>>>>
>>>
>>> ** Well, I did make virtually the same observations a few days ago.
>>>
>>> Kev has zero first hand knowledge of live sound reinforcement
>>> systems and is arguing from a position of ignorance and a some
>>> major false assumptions. The number 1 fallacy is that mic pre-amp
>>> noise is the dominant source of "hiss" in such systems when that
>>> rarely is the case.
>>> The reason is that the maximum signal level arriving from mics used
>>> in live, popular music is much higher than he has naively assumed -
>>> mic pre-amps only supplying only 15 to 25 dB of gain in most
>>> situations. This is particularly so if condenser mics are used
>>
>>
>> And oh dear oh dear....you just don't see the *global* picture.
>> There is no such thing as a free lunch. More output from the source
>> means more noise from the source. Its sods law.
>
>
Of course it does. You just don't have the sufficient technical
background to understand the point.
> ** This has no connection to the point above about mic output levels
Of course it does.
>
> and is in any case completely wrong
Not at all.
>- as expected from the
> incurably insane Kev Aylward.
Another excellent technical argument from Phil.
> The s/n ratio of a mic depends on
> diaphragm size - larger diaphragm mics having more output with no
> noise penalty.
Sure, tecnically the trade off is given by e.g
http://www.anasoft.co.uk/EE/cmospafl/cmospafl.html.
That is, in *any* system there is an inherent constraint of f(sigma,
frequency, power) = 0
where: Sigma = accuary = noise.
That is, one always has to trade off power, frequency and accuracy
against each other.
The "sods law" reference is a simplified, all things being equal sort of
thing.
In practise, getting the *same* performance but more signal, will
automatically mean more noise. So, what do you think the negative of a
large diaphragm is?
What was traded off? There is no escape. Something has to give. The nice
thing about the above general physics result, is that one doesn't have
to know say, the specifics of mic design, to know what parameter must
get worse.
What your comment does tell me is that you are an amateur. You don't
really understand how deep all of this is in the big scheme of things.
You've lernt a few things here and there, but have too many gaps in
your knowledge to see the big picture. As I said, you can't a free
lunch. Its that simple.
>
>
>> First capacitor mic I looked at
>> http://www.shure.com/microphones/recording/ksm/default.asp
>>
>> KSM141 -37db/Pa @ 94db spl (speaking 1 inch away), i.e. 14mv nominal.
>> s/n = 80db
>
>
> ** 14 mV output @ 94 dB SPL is some 20 dB greater than for an SM58 or
> similar - so pre-amp gain will be set 20 dB lower for the same
> result.
As I noted the normal speaking volume voltage is 14mv for this mic. Even
if singing was 20db above this, one would still need 20db of gain to get
the signal up to reasonable line levels. This means 100nv noise.
>The spec given by Shure for self noise is 14 dB SPL
> equivalent, *A ighted* - which is virtually identical to an SM58
> or similar used with a low noise pre-amp.
>
>
>> With 20db of front end gain this is 100nv at the line level stage.
>> Since op-amps are at the 16nv (TLO71) range, once again the mic
>> stage dominates.
>
>
>
> ** The "noise gain" of op-amps in EQ circuits and virtual earth
> summing stages etc is not unity.
And you point would be?
In EQ circuits the noise gain and signal gain is the same, so your point
is meaningless.
I have already addressed the excess noise gain of the summing amp
elswhere (i.e noise gain is >> signal gain). Its irrelevant in this
context as I am not discussing summing noise, but the noise effect of
outboard kit noise.
You seem to be blathering here. Your lack of formal electronic education
is becoming quite apparent. You have no idea what your talking about.
>
> One of the many **idiotic fallacies** that Mad Kev is purveying.
>
You do seem to have this habit of insults as a substitute for any
credible technical arguments.
Look, Phil, you aint gonna win. I am an expert in analogue design, and
have been so for 25 years. This should be obvious from my tutorial
papers http://www.anasoft.co.uk/EE/index.html.
Your letting you emotions of, "I hate this dude", get the best of you.
All your doing is embarrassing yourself. Until you refrain from your
barrage of personal abuse, your view will never mean anything.
Kevin Aylward
> Kevin Aylward wrote:
>>
>> I dont use graphics live:-)
>>
>> More trouble then they're worth. I would say the majority of working
>> soloists, duos, bands in pubs don't use them either.
>
> are using some sort of graphic. And I live in a rural area.
Well not here in the UK. I would say by far the majority are using
powered mixers or powered speakers. As I noted, I don't often see bands
with desks at the back of the hall. Its all small stuff. I am sceptical
of your claim in the US. I think you are probably ignoring all the
soloists, duos, and small outfits playing small venues that form the
majority of the working musicians. "Rock" bands are the exceptions, imo.
> Agreed, but in my example, which I find to be more the norm (mixer ->
> graphic -> amp) this situation could be remedied by leaving master
> faders at 0 and attenuating amps 20 dB. problem solved.
>
>>
>> I don't see that this is relevant. If the final output noise level is
>
>> acceptable at full power, the total noise at reduced power will be
>> less, therefore also acceptable. Its the *absolute* level of
>> background hiss, *not* the s/n by itself that people are complaining
>> about.
>
> By chaining several devices together all with mis matched input
> sensitivity adjustments, the s/n can get squashed further and further
> until there is as little as 40 to 45 db of of s/n ratio. At this
> point, if the amps are cranked, and the avg output is as little as
> 90db, the noise floor will be audible at 45 to 50 db period! (walking
> directly in front of speakers)
Flogging a dead horse.
>
> I realize that in a mixer -> amp situation it may not matter. But
> teaching ppl the wrong solutions will lead them to problems if and
> when they grow there system to include several devices in the signal
> chain.
>
More flogging a dead horse.
> You changed mojo to major. You suck.
Doesn't occur to you that I always spell check? I don't pay much
attention to what comes up i the dialog box, or what exact piece of text
gets selected.
>
>> I agree that from a practical point of view one likes to have the
>> fader somewhere midway, so one can adjust them up or down as needed.
>> I do this in the house. My MOSFET 1000 is, set around at around 1/3.
>
> in general... wrong.
Oh?
> YOu generally don't run them midway and the
> ability to move up and down is not why you'd run them there.
It is for me.
>YOu
> should run faders at or near their 0db point.
You are simply just regurgitating stuff from manufactures app notes.
Sure, its not a bad seting but by and large, in the big scheme of
things, it don't make much difference in bars.
>99% of faders are most
> senstive at this point, allowing very fine control of amplitude.
I don't care about fine control dude. This is live mate. A few dbs
change means shit in bars.
What I want to do is to have range to say, have max for PA and some
lower level for backing music, and so forth.
> Running them at -20 or -30 or whatever, a minor adjustmeant can be +6
> dBu.
My mid range comment is about running at this -20 or -30 for playing
background music during breaks. But again, as long as they is some room
to move the faders up and down, and not to attenuated, it don't matter
too much where they are.
>
> These are accepted SR standards that nearly everyone uses.
Yeah sure, many people parrot copy what others do with no concept that
doing things a little different can be useful.
Overall, there seems to be this idea of "studio perfectionist" that
without which, things sound crap. Its the, well "I'm so f'ing clever, I
know how it should be done", anything else is bogus. Well perfectionist
thinking is a well know sign of Schizophrenia. Well, mate, in the real
world those that actual play, just get on with the job. Within quite a
wide range, the set-up of a mixer don't matter squat. Get the mic gain
up, and your done. Thats what the majority do, and it works. Live with
it.
Kevin Aylward
Yep. Part of my point on graphics being more trouble then they're worth.
Its just another item for the the majority of the non technical
musicians to twiddle knobs wrongly.
Overall this NG seems to be full of bloody purists with no appreciation
for what really goes on on. The audience just don't care much about the
sound at all. They aren't musicians and are usually drunk as a skunk.
The singer can be completely of tune, as well as the guitarist, and most
just don't seem to notice.
The point you make is the crux of the matter.
The 1000' and 1000's of soloists, karacheists, duos, bands are non
technical and don't use a sound engineer. They never set the graphic and
other stuff correctly, so its pointless that stuff being there. It just
the way it is, despite all these "sound engineers" trying their best to
justify their existence. They aint required. That's why you don't see
them much. There ones you do see get paid f'all and are only there on
there on their one recourse in order to pick up the straggler women that
the rest of the band don't want. Trust me, I've been there:-)
Phil Allison
"Kevin Aylward"
= one very sick manic/depressive puppy
> ** The "noise gain" of op-amps in EQ circuits and virtual earth
> summing stages etc is not unity.
> In EQ circuits the noise gain and signal gain is the same,
** A massive piece of technical insanity.
> Look, Phil, you aint gonna win. I am an expert in analogue design, and
> have been so for 25 years.
** Dear Kev - you are very simply a mentally ill person.
Is your very ill wife still alive ??
Has her death from cancer totally rattled you ???
Professional help is available - but you need to realise that you need it.
Carrying on like a demented chimpanzee on Usenet is only going to destroy
you.
If you were not such an unmitigated pompous pommy ass folk might feel sorry
for you.
........... Phil
Kevin Aylward
> "Kevin Aylward"
>
> = one very sick manic/depressive puppy
>
>
>> ** The "noise gain" of op-amps in EQ circuits and virtual earth
>> summing stages etc is not unity.
>
>> In EQ circuits the noise gain and signal gain is the same,
>
>
> ** A massive piece of technical insanity.
Still with your usual brilliant technical arguments I see Phil.
The noise gain of a typical TB is (1+RF/RS). The signal gain is RF/RS.
So lets say we have 10db boost for the signal. Its RF/RS=2.16 for the
signal, and 3.16 for the noise. So, a gain differance of 3.3 db. Good
enough for me to be round about the same.
>
>
>> Look, Phil, you aint gonna win. I am an expert in analogue design,
>> and have been so for 25 years.
>
>
> ** Dear Kev - you are very simply a mentally ill person.
>
> Is your very ill wife still alive ??
>
> Has her death from cancer totally rattled you ???
>
> Professional help is available - but you need to realise that you
> need it.
> Carrying on like a demented chimpanzee on Usenet is only going to
> destroy you.
>
> If you were not such an unmitigated pompous pommy ass folk might feel
> sorry for you.
>
>
Well, your technical arguments are indeed compelling Phil.
Phil Allison
= one very sick manic/depressive puppy
> ** The "noise gain" of op-amps in EQ circuits and virtual earth
> summing stages etc is not unity.
> In EQ circuits the noise gain and signal gain is the same,
** A massive piece of technical insanity.
> Look, Phil, you aint gonna win. I am an expert in analogue design, and
> have been so for 25 years.
** Dear Kev - you are very simply a mentally ill person.
Is your very ill wife still alive ??
Has her death from cancer totally rattled you ???
Professional help is available - but you need to realise that you need it.
Carrying on like a demented chimpanzee on Usenet is only going to destroy
you.
If you were not such an unmitigated pompous pommy ass folk might feel sorry
for you.
........... Phil
Kevin Aylward
Well, there you go. More of the same amazingly compelling arguments from
Phil. Anyone with any improvements on this?
Gene Sweeny
>
> >What I do know is that you won't convince 5% of the folks
> > here that a powered board and speakers constitutes a professional
> > sound system.
>
> Its what 1000s and 1000s of actual gigging musicians do. Sure, sound
> engineers would like to justify their existance, but that dont change
> the facts.
This isn't alt.music.bar.old-geezer-bands
>
> > The fact that many moonlighting musicians get by with
> > such a rig without a qualified tech
>
> By and large, the all do. Very, very, few working musicaians have
sound
> engineers. They are simply are not required for the average bar band,
> due, trio, soloist etc.
They may not, however I would say that the overwhelming majority of the
musicians that don't, don't have a PA either. They'll rely on house
sound and or have (what I refuse to call a PA) one of those fender
passport things.
>
> The numbers are not set by the 0.01% of fully professional err..
> musicians, i.e. Brittany Spears, but by 100,000s, of *semi-pro* small
> time outfits who have a day job. To all an intents and purposes,
there
> are no "pro" musicians.
I think there are. Musicians that treat sound professionals with
respect, unlike you.
>
> >.doesn't mean it's of any
> > significance to discussions here, no more than the popularity of a
> > Ford Focus should be reflected in a sports car forum.
>
> Well, for sound engineers trying to justify the existence, probably
not,
> but for non technical people who what a quick and simple solution to
> setting up their powered mixer I would say my approach fits the bill.
If they have a powered mixer.... why would they have external amps with
input sensitivity knobs? Hmmmm?
> >
> > I'm not a fan of 31-band EQ's specifically, but something has to
> > neutralize the drivetrain to let the mixer do its job.
>
> Not really. No one in the audience cares or notices. As long as the
> singer is blond and wears a very short skirt, the sound don't matter
> diddly squat. This is "Living In The Real world" - Blondie.
You'd be surprised how many times I go to see a band with a poorly
adjusted PA that ppl come up to me and say, "wow, did they sound bad."
The audience cares more than you'd think. Especially when they paid a
cover at the door to see some good music. And even more so when the
lead singer isn't a blonde in a short skirt but an old geezer such as
yourself. Either look good, or sound good, that's what I always say.
>
> >That's where
> > the 20-yrs-old comment came from. Nowadays there are far more
> > effective devices starting around $250. And they don't hiss...
>
> Sure. I can't see any valid reason why modern day, line level stuff
> should generate objectionable hiss.
Well it does. I'm here to tell you that I've delt with all types of
crap ass consumer and MI gear and some "pro-level" stuff that hisses
like an asp. And if you don't set up your gain structure properly it
will be audible to those that still have their hearing intact.
Maybe you should go get your hearing checked out Kevy.
Phil Allison
= one very sick manic/depressive puppy
> ** The "noise gain" of op-amps in EQ circuits and virtual earth
> summing stages etc is not unity.
> In EQ circuits the noise gain and signal gain is the same,
** A massive piece of technical insanity.
> Look, Phil, you aint gonna win. I am an expert in analogue design, and
> have been so for 25 years.
** Dear Kev - you are very simply a mentally ill person.
Is your very ill wife still alive ??
Has her death from cancer totally rattled you ???
Professional help is available - but you need to realise that you need it.
Carrying on like a demented chimpanzee on Usenet is only going to destroy
you.
If you were not such an unmitigated pompous pommy ass folk might feel sorry
for you.
........... Phil
Steve White
I've been gigging in the UK for 20 years. In that time I must have played in
or guested with over 100 bands and watched 1,000s of amateur or semi-pro
bands.
In my experience, in the last say 10 years, around half of the bands I've
seen use an EQ unit on their main mix. Of the local venues I've been to
which have house PAs almost all have an EQ unit for this purpose.
Some of the bands I've played in had Kev's outlook on sound quality but more
than half didn't. The ones that were successful certainly didn't. They
believed the quality of their sound to be very important. By successful I
mean consistently pleased the audience, the venues and their customers.
I will however say that I don't tend to watch duos or people using backing
music. I have nothing against them but it's not what I like to see in a
band. The proportions there could be different. However, from my experience
there are many more "real" bands than duos etc. with backing.
For my part I came here to learn how the pros do it so I can use that
knowledge to help improve my ability and the sound of the bands I play in.
If I wasn't a pro or wasn't interested in these things I don't know what I'd
be here for.
Cheers,
Steve W
Kevin Aylward
> Kevin Aylward wrote:
>>
>> from Phil. Anyone with any improvements on this?
>
> anything about running a PA system.
You are entitled your opinion.
>All newbies and ppl of lesser
> experience please do not listen to his crackpot advice on running a
> sound reinforcement system.
Look mate, running the mixer to the amp, with the amp flat out will be
perfectly acceptable for 99% of live applications. End of story.
>
> I think everyone who has experience in this arena, would agree that
> there are ways to set up a PA system
What we have here are protectionist alledged "sound engineers" trying to
make all sorts of vacuous claims that to get a decent sound requires all
sort of special masons handshakes, only understood by said alledged
"sound engineers". Get real. Its not credible. PAs are shit simple. Plug
it in twiddle the knobs and go. You have to be a right wasy to not be
able to get a reasonable sound. Sure, I have listened to some bands with
less then perfect sound, but by and large, when I go and see sound
engineerless, layman musician bands, the sound is perfectly adequate.
Period.
>and ways *not* to set up a PA
> system. Kev has a monthly subscription to "Stupid Muso trying to run
> a PA."
>
The only thing that really appears to be in debate here is whether or
not it makes sense to back off the amp gain and drive the mixer harder
into the graphic, if there. The claim is, from your implication of "I
don't know shit", is that if one don't do this, the sound must be shit.
This aint credible mate.
> Here's an analogy. There are many competent Electrical Engineers that
> design power transformers, rectifiers, power distribution systems, and
> the like for a living. Would I trust them to wire my home with
> household AC. Absolutely not. In this analogy, Kev would be the
> engineer, and the PA system would be my home.
Yep, you are still entitled your opinion.
Kevin Aylward
> For what it's worth.
>
> I've been gigging in the UK for 20 years. In that time I must have
> played in or guested with over 100 bands and watched 1,000s of
> amateur or semi-pro bands.
>
> In my experience, in the last say 10 years, around half of the bands
> I've seen use an EQ unit on their main mix. Of the local venues I've
> been to which have house PAs almost all have an EQ unit for this
> purpose.
My mackie mixer has an internal graphic. My powered, name not memtioned
mixer has an internal graphic. Only my Studiomaster big desk don't.
Do you actually see those with the 31 band graphics actually do room
acoustic set-ups to set them up?
>
> Some of the bands I've played in had Kev's outlook on sound quality
> but more than half didn't.
I am not sure what you mean by my outlook on quality. I am stating
objective *facts* as I see them. This does not mean that *I* don't
personally try an get a decent sound better then the norm.
Sure I don't use an outboard graphic. Its pointless unless its in a
fixed installation. There is no time to piss about trying to "equalise
the bar acoustics". Its like, load the van at 5:00PM, drive 20 miles,
have the bass player call saying his car wont start, so back to pick him
up. Drive 70 bloody miles. Get to the pub, haul the gear up two flights
of stairs, put the wig on and silver boots, etc... plug the shit in.
Fuck we're late. Start playing dudes. Shit who didn't tune up...
>The ones that were successful certainly
> didn't.
I can't agree. I don't see much correlation between successful bands and
the quality of the PA set up.
>They believed the quality of their sound to be very
> important.
>By successful I mean consistently pleased the audience,
> the venues and their customers.
And this is achievable by very minimal PA set-up, with minimal
knowledge. Its simply not rocket science. Jesus wept dude, its just a
mixer and an amp.
One don't need the *best* sound, one just needs a *good* sound. And
absolutely, that don't require a 31 band graphic.
>
> I will however say that I don't tend to watch duos or people using
> backing music. I have nothing against them but it's not what I like
> to see in a band. The proportions there could be different. However,
> from my experience there are many more "real" bands than duos etc.
> with backing.
How many Karaoke pubs do you think there are? The deal about duos in the
UK is that the licensing laws are such that pubs can engage them without
a licence. This makes a big difference. However, this is being phased
out by November, when all entertainment will require a licence.
>
> For my part I came here to learn how the pros
>do it so I can use that
But one does not know who the real pros are in NGs. Anyone can say
anything they like, and usually do. How many of these alleged pros have
a B.Sc. in Audio engineering? How many actual really set up sound, in a
pub rather then in their bedrooms?
> knowledge to help improve my ability and the sound of the bands I
> play in. If I wasn't a pro or wasn't interested in these things I
> don't know what I'd be here for.
I'm, here just to pass the time. It can be quite amusing reading Phil's
rants and raves. But...I am now off to my local pub to see the
band...guaranteed to be powered mixer, probably the band with the
guitarist that uses the 4 of 4 by 12 Marshal cabinets!!!
Joe Kotroczo
SdU7e.72384$C12....@fe1.news.blueyonder.co.uk, "Kevin Aylward"
<see_w...@anasoft.co.uk> wrote:
(...)
> You are entitled your opinion.
>
>> All newbies and ppl of lesser
>> experience please do not listen to his crackpot advice on running a
>> sound reinforcement system.
>
> Look mate, running the mixer to the amp, with the amp flat out will be
> perfectly acceptable for 99% of live applications. End of story.
/Quote/ "You are entitled your opinion." /Endquote/
And some people even believe that there is a god in heaven...
I can agree that some people, who don't know how to use an EQ properly, are
better off without one.
>> I think everyone who has experience in this arena, would agree that
>> there are ways to set up a PA system
>
> What we have here are protectionist alledged "sound engineers" trying to
> make all sorts of vacuous claims that to get a decent sound requires all
> sort of special masons handshakes, only understood by said alledged
> "sound engineers". Get real. Its not credible. PAs are shit simple. Plug
> it in twiddle the knobs and go. You have to be a right wasy to not be
> able to get a reasonable sound. Sure, I have listened to some bands with
> less then perfect sound, but by and large, when I go and see sound
> engineerless, layman musician bands, the sound is perfectly adequate.
> Period.
There is a difference between "reasonable" sound and "good" sound. If your
goal is to achieve "reasonable" then you're in the wrong newsgroup, mate.
Most people here have to achieve "good" if not "very good" sound
consistently to avoid losing customers. "Reasonable" is not an option.
BTW. I'm most willing to further discuss this over a pint next time I'm in
the Basildon area. I'm sure so is Phildo, but his will be non-alcoholic.
That would be roundabout June...
Joe
--
Joe Kotroczo kotr...@mac.com
Gene Sweeny
>
> Look mate, running the mixer to the amp, with the amp flat out will
be
> perfectly acceptable for 99% of live applications. End of story.
This just shows how "out of your depth" you are when it comes to live
sound.
>
> >
> > I think everyone who has experience in this arena, would agree that
> > there are ways to set up a PA system
>
> What we have here are protectionist alledged "sound engineers" trying
to
> make all sorts of vacuous claims that to get a decent sound requires
all
> sort of special masons handshakes, only understood by said alledged
> "sound engineers". Get real. Its not credible. PAs are shit simple.
Plug
> it in twiddle the knobs and go. You have to be a right wasy to not be
> able to get a reasonable sound. Sure, I have listened to some bands
with
> less then perfect sound, but by and large, when I go and see sound
> engineerless, layman musician bands, the sound is perfectly adequate.
> Period.
Here, you show that you don't really care about live sound. The rest
of us are here to increase our knowledge of the subject, pass on
advice, and strive for as much perfection as can be achieved in a live
setting. You, "mate," are a simple minded muso with an EE degree.
This hardly qualifies you as an expert in the live sound profession.
Just go away.
>
> >and ways *not* to set up a PA
> > system. Kev has a monthly subscription to "Stupid Muso trying to
run
> > a PA."
> >
>
> The only thing that really appears to be in debate here is whether or
> not it makes sense to back off the amp gain and drive the mixer
harder
> into the graphic, if there. The claim is, from your implication of "I
> don't know shit", is that if one don't do this, the sound must be
shit.
> This aint credible mate.
I didn't claim it would sound like sh*t. My claim was, there is a
proper way to do things and teaching nOObs the wrong stuff will screw
them up later.
Gene Sweeny
>
> My mackie mixer has an internal graphic. My powered, name not
memtioned
> mixer has an internal graphic. Only my Studiomaster big desk don't.
This is what I don't get... you said in previous post that most acts
that are small time and don't care about sound have a powered mixer
type deal. What need would there be to adjust amp gains on a powered
mixer?
>
> Do you actually see those with the 31 band graphics actually do room
> acoustic set-ups to set them up?
They're great for notching out feedback with out destroying the sound.
Hard to do with a wimpy 9 band graphic eq included on most MI mixers.
>
> >
> > Some of the bands I've played in had Kev's outlook on sound quality
> > but more than half didn't.
>
> I am not sure what you mean by my outlook on quality. I am stating
> objective *facts* as I see them. This does not mean that *I* don't
> personally try an get a decent sound better then the norm.
If you were really trying, you'd learn proper practices. We don't say
this stuff to blow hot air. These methods truely enhance the
performance of a PA system.
>
> Sure I don't use an outboard graphic. Its pointless unless its in a
> fixed installation. There is no time to piss about trying to
"equalise
> the bar acoustics". Its like, load the van at 5:00PM, drive 20 miles,
> have the bass player call saying his car wont start, so back to pick
him
> up. Drive 70 bloody miles. Get to the pub, haul the gear up two
flights
> of stairs, put the wig on and silver boots, etc... plug the shit in.
> Fuck we're late. Start playing dudes. Shit who didn't tune up...
This shows how truely *unprofessional* you really are. Muso with an EE
degree.
>
> >The ones that were successful certainly
> > didn't.
>
> I can't agree. I don't see much correlation between successful bands
and
> the quality of the PA set up.
Again, showing your naivety in the world of professional live sound.
>
> >They believed the quality of their sound to be very
> > important.
> >By successful I mean consistently pleased the audience,
> > the venues and their customers.
>
> And this is achievable by very minimal PA set-up, with minimal
> knowledge. Its simply not rocket science. Jesus wept dude, its just a
> mixer and an amp.
>
> One don't need the *best* sound, one just needs a *good* sound. And
> absolutely, that don't require a 31 band graphic.
All completely true... If you're playing for a bunch of drunks in a
pub.
> >
> > For my part I came here to learn how the pros
> >do it so I can use that
>
> But one does not know who the real pros are in NGs. Anyone can say
> anything they like, and usually do. How many of these alleged pros
have
> a B.Sc. in Audio engineering? How many actual really set up sound, in
a
> pub rather then in their bedrooms?
"Ehhhh, look at me. I'm an old geezer with a EE degree eehhhhhhhh. I
design analogue circuits ehhhhh. Where's my pocket protector."
I don't care if you have a DOCTORATE in audio engineering. I'd like to
see you go a day without mentioning "I'm Kevin Alyward, I have a B.Sc
in Audio engineering."
>
> > knowledge to help improve my ability and the sound of the bands I
> > play in. If I wasn't a pro or wasn't interested in these things I
> > don't know what I'd be here for.
>
> I'm, here just to pass the time. It can be quite amusing reading
Phil's
> rants and raves. But...I am now off to my local pub to see the
> band...guaranteed to be powered mixer, probably the band with the
> guitarist that uses the 4 of 4 by 12 Marshal cabinets!!!
Cause THAT'S appropriate in a pub. People are going to be deaf.
shannon
>
> What we have here are protectionist alledged "sound engineers" trying to
> make all sorts of vacuous claims that to get a decent sound requires all
> sort of special masons handshakes, only understood by said alledged
> "sound engineers". Get real. Its not credible. PAs are shit simple. Plug
> it in twiddle the knobs and go. You have to be a right wasy to not be
> able to get a reasonable sound. Sure, I have listened to some bands with
> less then perfect sound, but by and large, when I go and see sound
> engineerless, layman musician bands, the sound is perfectly adequate.
> Period.
>
Thats the difference between Blondie and Blon-DEE.
shannon
What are you doing here then ?
This group is for people who do care, who work with acts who do care,
playing to audiences who do care.
I'm off to set up a small show in a bar. They normally just have a DJ but
they want this set up right because the act is a guy from REM and we don't
often get a chance to see good musicians playing original material in an
intimate setting. These gigs really go on, you're just not part of it.
There's no market here for part timers in 80s tribute bands.
Bob Quintal
> This is getting a bit pointless. Its obvious that you are
> simple cluless about electronic design to make this disscuion
> pointless.
>
> Look, here's my trumpet blowing. I have been a professional,
> B.Sc. degreed, design engineer for 25 years. For instance, I
> was Analogue IC design team leader at Texas Instruments for 3
> years, among other positions. I designed op-amps for a living
> dude. I have spent 10,000's of hours designing high transistor
> count analogue circuitry. I know op-amp design inside out.
>
> As a bacground check out my tutorial papers
> http://www.anasoft.co.uk/EE/index.html
> and a minor bio http://www.anasoft.co.uk/band/bio.htm
>
I dont know why I'm bothering to answer a twit like you, but, I'm
so clueless that I have held progressively more responsible
engineering positions with the same aerospace design and
manufacturing firm for the last 27 years,
You, on the other hand, have bounced (or been bounced) from job
to job, changing your area of (probably in)competence radically
each time. You have designed op-amps. Which ones would I possibly
have DPA'ed (That's Destructive Physical Analysis for any readers
not familiar with the term.)
I did read some of the absolute bullshit on your site, like the
nonsense about bjts being voltage controlled devices.
Or was the publication date of your scientifucked paper April 1?
You are the weakest link-- Goodbye.
Chas Gill
"Kevin Aylward" <see_w...@anasoft.co.uk> wrote in message
news:SdU7e.72384$C12....@fe1.news.blueyonder.co.uk...
> Gene Sweeny wrote:
>> Kevin Aylward wrote:
>>>
>>> Well, there you go. More of the same amazingly compelling arguments
>>> from Phil. Anyone with any improvements on this?
>>
>> I do. Kev may know how to design a mic pre, but he doesn't know
>> anything about running a PA system.
>
> You are entitled your opinion.
>
>>All newbies and ppl of lesser
>> experience please do not listen to his crackpot advice on running a
>> sound reinforcement system.
>
> Look mate, running the mixer to the amp, with the amp flat out will be
> perfectly acceptable for 99% of live applications. End of story.
>
>>
>> I think everyone who has experience in this arena, would agree that
>> there are ways to set up a PA system
>
> What we have here are protectionist alledged "sound engineers" trying to
> make all sorts of vacuous claims that to get a decent sound requires all
> sort of special masons handshakes, only understood by said alledged "sound
> engineers". Get real. Its not credible. PAs are shit simple. Plug it in
> twiddle the knobs and go. You have to be a right wasy to not be able to
> get a reasonable sound. Sure, I have listened to some bands with less then
> perfect sound, but by and large, when I go and see sound engineerless,
> layman musician bands, the sound is perfectly adequate. Period.
Friggin' well said.
It seems to me that most so-called knowledge of how to set up and run a
sound reinforcement system is based on handed down knowledge that is more to
do with folklore than sound (in both senses of the word) engineering
principles. Like, do one really good gig (by accident) and live off the
kudos for the rest of your career.
Fact is the Pareto principle applies to virtually everything, including
sound reinforcement. You get 80% of the result for 20% of the effort and
the achievement of the last 20% requires 80% of the effort. Trouble is your
average punter won't notice the last 20%. For those that do, this NG
exists.
A small audience in a very small pond.
A real sound enginner for whom I had the greatest respect once, in the early
days, said something that I've never forgotten and which guides me to this
day. Pay attention to the detail - don't ever skip anything and never
assume anything. It has stood me in good stead up until now and I suspect
it always will.
Having said all of this I guess it's pointless to come back asking for
advice. Having watched the recent slag matches and particularly foul
language in recent debates I don't think I'm gonna learn anything more from
this NG, except the obvious point that each and every expert out there seems
to think that the only way to prove their expertise is to piss all over
anyone that might also be an expert. There are an awful of sad bastards out
there for whom I feel very sorry.
Thanks, chaps, I'll just toddle off and learn it all the hard way - and
hopefully I won't develop an inferiority/superiority/megalomaniac complex or
Turettes syndrome whilst I'm doing it.
Be wonderful at what you do - and I hope (but fear otherwise) that the
paying punter will notice the difference.
Au Revoir.
Chas
Gene Sweeny
trying to
> > make all sorts of vacuous claims that to get a decent sound
requires all
> > sort of special masons handshakes, only understood by said alledged
"sound
> > engineers". Get real. Its not credible. PAs are shit simple. Plug
it in
> > twiddle the knobs and go. You have to be a right wasy to not be
able to
> > get a reasonable sound. Sure, I have listened to some bands with
less then
> > perfect sound, but by and large, when I go and see sound
engineerless,
> > layman musician bands, the sound is perfectly adequate. Period.
>
>
> Friggin' well said.
I think not.
>
> It seems to me that most so-called knowledge of how to set up and run
a
> sound reinforcement system is based on handed down knowledge that is
more to
> do with folklore than sound (in both senses of the word) engineering
> principles. Like, do one really good gig (by accident) and live off
the
> kudos for the rest of your career.
I will agree, there are a ton of dumba** "sound engineers" out there.
However, I still think that the advice that others and myself have
offered to all are based on sound principles that the majority of
*knowledgable* sound professionals use day in and day out. Minus the
foul language.
>
> Fact is the Pareto principle applies to virtually everything,
including
> sound reinforcement. You get 80% of the result for 20% of the effort
and
> the achievement of the last 20% requires 80% of the effort. Trouble
is your
> average punter won't notice the last 20%. For those that do, this NG
> exists.
100% agree. That's why I strive for a "perfect" show, albiet, there
are always problems. C'est la vie.
>
> A small audience in a very small pond.
>
> A real sound enginner for whom I had the greatest respect once, in
the early
> days, said something that I've never forgotten and which guides me to
this
> day. Pay attention to the detail - don't ever skip anything and
never
> assume anything. It has stood me in good stead up until now and I
suspect
> it always will.
Seems like good advice. To me, it sounds like striving for the truth.
>
> Having said all of this I guess it's pointless to come back asking
for
> advice. Having watched the recent slag matches and particularly
foul
> language in recent debates I don't think I'm gonna learn anything
more from
> this NG, except the obvious point that each and every expert out
there seems
> to think that the only way to prove their expertise is to piss all
over
> anyone that might also be an expert. There are an awful of sad
bastards out
> there for whom I feel very sorry.
Different strokes for differet folks.