Shunt Resistance For Mics
David Aguilera
You mentioned this the other day:
> [snip]
>
>Buy some of the old classics. Everyone should have an SM-57, an EV
>635A, and maybe an RE-20. The first two are very cheap, and all three
>can be found used. The first two are very touchy about loading and
>will benefit from a shunt resistor in a barrel connector if you are
>using a modern transformerless preamp.
>
> [snip]
I'm just getting into the electrical side of this field and am really
curious as to how to go about building the shunt resistor. If you have
the time to enlighten me I would greatly appreciate it.
The preamp I am using is not giving me what I want out of my 57's and
EV635A. I've always thought it was an impedance problem, but If I could
build a box with switchable resistance...well--wow! That's
straightforward and downright useful!
I don't have a anything that will measure impedance, so I am unsure of
the impedance of my preamps. Is there a way to measure it with a my
multimeter sans one?
After that it's like connecting resistance in serial right (I'm really
weak in AC electronics)?
Thanks,
David
John Williamson
>
> I don't have a anything that will measure impedance, so I am unsure of
> the impedance of my preamps. Is there a way to measure it with a my
> multimeter sans one?
>
be made up of capacitive and resisitive components for a transformerless
preamp, or capacitive, resistive and inductive components if there is a
transformer present. It's normally just quoted as "X ohms", though, as
if it were purely resistive.
> After that it's like connecting resistance in serial right (I'm really
> weak in AC electronics)?
>
No, a shunt resistor goes in parallel with the load. The easiest way to
make the unit you need is to connect a resistor across the signal. Make
up a short XLR-XLR extension lead, and solder a resistor between pins 2
and 3. Changing the resistor value will change the sound of older
microphones. For setting up, start with a 2 Kilohm variable resistor,
adjust it until you get the sound you like, then replace that with the
nearest value of fixed resistor. Use a low inductance metal film
resistor, preferably a one percent tolerance type. Carbon resistors have
more noticeable self noise.
--
Tciao for Now!
John.
David Aguilera
Thanks John!
Mike Rivers
> I'm just getting into the electrical side of this field and am really
> curious as to how to go about building the shunt resistor. If you have
> the time to enlighten me I would greatly appreciate it.
Make up a short (1 foot or so) mic cable with a resistor connected between
pins 2 and 3 of the XLR connector on one end.
> The preamp I am using is not giving me what I want out of my 57's and
> EV635A. I've always thought it was an impedance problem, but If I could
> build a box with switchable resistance...well--wow! That's
> straightforward and downright useful!
Actually, at the Summer NAMM show, I saw someone who made a box
just like that. But honestly, if you're not getting what you want form your
SM57s or 635s, it's probably not the preamp. The 600 or so ohm load
across an SM57 cleans it up noticeably but it doesn't make it "warmer" or
"more transparent" or any of those other feelgood terms.
> I don't have a anything that will measure impedance, so I am unsure of
> the impedance of my preamps. Is there a way to measure it with a my
> multimeter sans one?
Yes, but it's more complicated than just hooking it across the input
terminals.
What you need to do is put a variable resistor (pot) in series with a
generator and
connect it to the input. You'll want to connect the generator between
pins 2 and 3, or
between pins 2 and 1, with pin 3 grounded (connected to pin 1).
Fire up the generator and measure the output voltage with the variable
resistor at
zero ohms, making sure you're not driving the preamp into clipping with the
generator. Then adjust the pot so that the voltage drops to half what
you measured
with no resistance in the circuit. Then measure the resistance of the
pot. That's
half the input impedance.
LAB
measured with no resistance in the circuit. Then measure the resistance of
the
pot. That's half the input impedance.
If Vgen=Vpot+Vin, when Vin is the half (-6dB) of Vgen Rin=Rpot. Why the
half?
Note: generator has his output impedance (50...600 Ohms). You have to
consider it.
--
Gianluca
Wecan do it
"LAB" <n...@mail.com> wrote in message
news:4af8767d$0$702$5fc...@news.tiscali.it...
> >> Then adjust the pot so that the voltage drops to half
> what you measured with no resistance in the circuit. Then
> measure the resistance of the
> pot. That's half the input impedance.
>
> If Vgen=Vpot+Vin, when Vin is the half (-6dB) of Vgen
> Rin=Rpot. Why the half?
Because if it is half then you can read Zin directly from the
pot with your multimeter.
peace
dawg
geoff
> David Aguilera wrote:
>
>> I'm just getting into the electrical side of this field and am really
>> curious as to how to go about building the shunt resistor. If you
>> have the time to enlighten me I would greatly appreciate it.
>
> Make up a short (1 foot or so) mic cable with a resistor connected
> between pins 2 and 3 of the XLR connector on one end.
An XLR-M to XLR-F 'barrel' adaptor is tidier.
geoff
geoff
Not if there is a coupling capacitor in the way.
geoff
Mike Rivers
> If Vgen=Vpot+Vin, when Vin is the half (-6dB) of Vgen Rin=Rpot. Why
> the half?
Which half? When the voltage drops in half, half the voltage is dropped
across
the preamp input and the pot, so their impedances are equal. This is of
course
ignoring the source resistance of the generator, which you may not be able
to do, depending on the generator or what you do about it.
If it's the half of the impedance you're asking about, the other leg of
the differential
input has (presumably) the same impedance, and both are across the
microphone,
essentially in series.
> Note: generator has his output impedance (50...600 Ohms). You have to
> consider it.
There's always some wise guy in the audience. You're correct of course,
but I
didn't mention that in my reply for two reasons:
1 - the guy probably isn't going to make the measurement as I described
anyway
2 - He's going to ask "what's a good generator that I can buy from eBay?
I'd probably suggest that he use a DAW program to generate a sine wave
and use his sound card as a generator. If it's something fancier than the
sound card built into his computer, it will probably have a source impedance
of around 20 ohms.
Alternately, he can build an attenuator which he may need anyway to get the
generator output low enough not to clip the preamp, and make the shunt
resistor
low enough so the generator source impedance is insignificant.
Mike Rivers
> An XLR-M to XLR-F 'barrel' adaptor is tidier.
But you'd have to buy it. You can make a short cable out of
a long cable, even one that you've junked because it's faulty,
as long as you don't use the faulty part.
PStamler
As the guy who started some of this discussion with an article called
"The Taming of the Shure" in Recording Magazine, let me weigh in with
a couple of suggestions.
1. I don't recommend using barrel connectors; if you put them on the
preamp/mixer end of the cable, one yank of the cable can badly torque
the XLR on the mixer/preamp, possibly causing expensive damage. A
short cable is safer.
2. Solder the resistor between pins 2 and 3 of the male connector on
the cable. I recommend the male connector because there's a bit more
room inside the shell. Even so, you'll need to snip the leads off real
short. Make sure that, when you solder in the resistor, the cable
doesn't come unsoldered. This is a lot easier to do properly with a
Neutrik connector rather than Switchcraft.
3. If you look at your mixer/preamp's manual, you should find the
input impedance listed and can compute the proper parallel resistor.
In the absence of more data, though, for an SM57 a 681 ohm (NOT 681k!)
resistor will usually give good results.
4. The improvement is mostly in the area of cleanness and clarity, and
lack of high-frequency "hashiness". In the comparison I originally
did, we listened to the SM57 on a drumkit; with a standard load (2k)
the cymbals sounded more like white noise than the metallic ring a
cymbal should have. With the parallel resistor, it sounded much more
like a real cymbal. On acoustic guitar switching to the parallel
resistor channel took away a lot of the "ringy" quality at high
frequencies.
5. I've only found a few microphones that are helped by this
technique; the most important was the SM57 family (SM56 and SM58
included). The AKG D112 got a lot less "pillowy" and a lot tighter on
the bottom, but the jury's out whether that was an improvement. Among
the mics for which this makes little or no difference are Electro-
Voice RE15 and RE20 and Shure Beta57 It's not recommended for
condenser mics.
Peace,
Paul
geoff
Then you need to have (or buy) an XLR-M and XLR-F which you can now only use
for that purpose.
Whatever works for you. I have a labelled set of 6 of the aforementioned
barrel adaptors, and another pair wired as polarity changers. For a few
bucks each (unless one is addicted to Neutrik) it's really neither here nor
there....
geoff
geoff
> On Nov 9, 6:15 pm, Mike Rivers <mriv...@d-and-d.com> wrote:
>> geoff wrote:
>>> An XLR-M to XLR-F 'barrel' adaptor is tidier.
>>
>> But you'd have to buy it. You can make a short cable out of
>> a long cable, even one that you've junked because it's faulty,
>> as long as you don't use the faulty part.
>
> As the guy who started some of this discussion with an article called
> "The Taming of the Shure" in Recording Magazine, let me weigh in with
> a couple of suggestions.
>
> 1. I don't recommend using barrel connectors; if you put them on the
> preamp/mixer end of the cable, one yank of the cable can badly torque
> the XLR on the mixer/preamp, possibly causing expensive damage. A
> short cable is safer.
Just like any mic XLR that is plugged in then, but admittedly a little more
so because of the fulcrum/leverage/thing. But as I routinely gaffer the
attached cables to something solid, have never had a problem.
>
> 4. The improvement is mostly in the area of cleanness and clarity, and
> lack of high-frequency "hashiness". In the comparison I originally
> did, we listened to the SM57 on a drumkit; with a standard load (2k)
> the cymbals sounded more like white noise than the metallic ring a
> cymbal should have. With the parallel resistor, it sounded much more
> like a real cymbal. On acoustic guitar switching to the parallel
> resistor channel took away a lot of the "ringy" quality at high
> frequencies.
I find it takes the harsh splashiness off the 57s, and makes the bottom end
firmer. Nothing like a nice firm bottom. With the loader, 57s more usable
on toms.
geoff
Mike Rivers
> Then you need to have (or buy) an XLR-M and XLR-F which you can now only use
> for that purpose.
No you don't. Everybody has a useless mic cable lying around somewhere.
That contains
all the parts you need except for the resistor, and you'd have to buy
that anyway. I fully
agree with Paul's example of too much torque on the chassis connector,
It would probably
be OK to put a barrel adapter on the mic end if the mic stayed on a
stand and you secured the
cable. But he was talking about improved sound on cymbals, which
suggests drum overheads,
which suggests high, which probably doesn't need to be any higher.
> Whatever works for you.
That works for me.
Ian Bell
Since when has a 'standard' load been 2K??
> the cymbals sounded more like white noise than the metallic ring a
> cymbal should have. With the parallel resistor, it sounded much more
> like a real cymbal. On acoustic guitar switching to the parallel
> resistor channel took away a lot of the "ringy" quality at high
> frequencies.
>
Dynamic mics are basically inductive which means their impedance rises
with frequency. By loading it with 680R instead of 2K you effectively
hit the high frequency response - something you can do with infinitely
more precision using EQ.
Cheers
Ian
LAB
> with no resistance in the circuit. Then measure the resistance of
If Vgen=Vpot+Vin, when Vin is the half (-6dB) of Vgen Rin=Rpot. Why *the
half*?
Note: generator has his output impedance (50...600 Ohms). You have to
consider it.
>> Because if it is half then you can read Zin directly from the pot with
>> your multimeter.
>>> Not if there is a coupling capacitor in the way.
I was assuming a balanced 1kHz signal...
You could measure -6dB by a VUmeter on the preamp/mixer out.
--
Gianluca
Don Pearce
wrote:
>Dynamic mics are basically inductive which means their impedance rises
>with frequency. By loading it with 680R instead of 2K you effectively
>hit the high frequency response - something you can do with infinitely
>more precision using EQ.
>
>Cheers
>
>Ian
You also change the Q of the resonances, not quite so simple with EQ.
d
Scott Dorsey
Since cheap consoles with transformerless input stages became standard.
It's depressing.
>Dynamic mics are basically inductive which means their impedance rises
>with frequency. By loading it with 680R instead of 2K you effectively
>hit the high frequency response - something you can do with infinitely
>more precision using EQ.
That's not all that is going on here. You're adjusting the mechanical
damping of the diaphragm, and moving the resonance down and widening it,
as you put electrical (and thus mechanical) load on. Many dynamic elements
will ring badly if they are insufficiently loaded.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Mike Rivers
> Since when has a 'standard' load been 2K??
It's sort of a roughly weighted average of the input impedance of
a good many modern transformerless preamps and mixing consoles.
This is typical of the gear to which an SM57 is likely to be connected,
at least among the Recording Magazine readership, where Paul's
article was published.
Perhaps "typical" would have been a better term, unless (and I don't
recall the details of the study) a specific load resistor for each preamp
was chosen so that the mic would be loaded by 2K for any preamp
in the study. This would allow the preamp to be the major source of
differences in sound when used with an SM57.
Further listening experiments indicated that the mic sounded better
when loaded with 600 ohms (adjusted for the preamp used for testing).
> Dynamic mics are basically inductive which means their impedance rises
> with frequency. By loading it with 680R instead of 2K you effectively
> hit the high frequency response - something you can do with infinitely
> more precision using EQ.
More important, you change the damping of the diaphragm assembly,
which broadens resonant peaks and may even change the frequency
of some resonances. That's not easily done with EQ.
Wecan do it
> If Vgen=Vpot+Vin, when Vin is the half (-6dB) of Vgen
> Rin=Rpot. Why *the > half*?
>>> Because if it is half then you can read Zin directly from
>>> the pot with your multimeter.
>>>> Not if there is a coupling capacitor in the way.
True but not applicable.
If you ever measured impedance using the voltage divider
method you would know the pot is outside of the device under
test ( the preamp whos impedance you are measuring) and once
you adjust the pot for half Vgen you remove the pot from the
circuit and measure it with your multimeter.
The pot (or more correctly a rheostat in the properly wired
configuration) will be equal to the input impedance at the
frequency Vgen is set at. It is impossible for a coupling
capacitor to be in the way without you knowing it.
If I was to measure the impedance I would do the same voltage
devider method but use my FFT analyzer to graph the impedance
at a few thousand frequencies to be sure nothing funny is
going on.
peace
dawg
Ian Bell
I suspect the Q is so low in most mics it makes no difference.
Cheers
Ian
Ian Bell
Personally I doubt that is a factor. The Q of the average dynamic mic is
so low that changing the load this way will make little difference.
Cheers
ian
Ian Bell
Indeed, but as I said elsewhere, I suspect the Q of the average dynamic
mic is so low that this is not a factor.
Cheers
ian
Scott Dorsey
>Scott Dorsey wrote:
>> Ian Bell <ruffr...@yahoo.com> wrote:
>>> Since when has a 'standard' load been 2K??
>>
>> Since cheap consoles with transformerless input stages became standard.
>> It's depressing.
>>
>>> Dynamic mics are basically inductive which means their impedance rises
>>> with frequency. By loading it with 680R instead of 2K you effectively
>>> hit the high frequency response - something you can do with infinitely
>>> more precision using EQ.
>>
>> That's not all that is going on here. You're adjusting the mechanical
>> damping of the diaphragm, and moving the resonance down and widening it,
>> as you put electrical (and thus mechanical) load on. Many dynamic elements
>> will ring badly if they are insufficiently loaded.
>
>so low that changing the load this way will make little difference.
You'd be shocked. Take an SM-57 and look at an impulse with a 600 ohm
load and with an unterminated output. It's not a subtle difference.
You'd think the losses through the internal transformer would reduce the
effects of loading changes, but they are still very significant.
On an RE-20 you won't notice it so much, mind you. Maybe not at all.
geoff
>>
>
> Personally I doubt that is a factor. The Q of the average dynamic mic
> is so low that changing the load this way will make little difference.
>
Given the easily-heard results with a humble SM57, a simple change in
loading clearly does make a difference.
If you are skeptical, try it !
geoff
geoff
> "LAB" <n...@mail.com> wrote in message
> news:hdbu5j$9et$1...@serv1.iunet.it...
>
>> If Vgen=Vpot+Vin, when Vin is the half (-6dB) of Vgen
>> Rin=Rpot. Why *the > half*?
>
>>>> Because if it is half then you can read Zin directly from
>>>> the pot with your multimeter.
>
>>>>> Not if there is a coupling capacitor in the way.
>
>
> True but not applicable.
Thought you were advocating testing in input impedence of a preamp by
hooking a multimeter to it and simply reading the R of the scale..
geoff
Mike Rivers
> Thought you were advocating testing in input impedence of a preamp by
> hooking a multimeter to it and simply reading the R of the scale..
I think that's what the original poster was dreaming about, but he's
been told
that he can't. But he can still make use of his multimeter to make the
measurement.
It just takes some extra stuff.
Mike Rivers
> I suspect the Q is so low in most mics it makes no difference.
Suspect all you want, but test it before you poo-poo the concept.
Mike Rivers
> Indeed, but as I said elsewhere, I suspect the Q of the average dynamic
> mic is so low that this is not a factor.
We're not talking about an "average dynamic mic" here (whatever that
is), we're
talking about a specific mic, an SM57. Given the large market
penetration, it
may be the most popular mic in its class. That might make it "average"
in some
circles.
As Scott pointed out, an RE20 doesn't exhibit the same resonant
characteristics
as an SM57. and doesn't have the same problem. A loading experiment with an
RE20 probably wouldn't yield much useful data.
Mark
>
> - Show quoted text -
Ian,
you don't want to let facts and science get in the way of
mythology :-)
Mark
Scott Dorsey
>
>As Scott pointed out, an RE20 doesn't exhibit the same resonant
>characteristics
>as an SM57. and doesn't have the same problem. A loading experiment with an
>RE20 probably wouldn't yield much useful data.
My guess is that the coupling in the RE-20 just is a lot lower, so load
changes have less effect on the mechanical behaviour of the system.
If that were the case, you'd expect the output of the RE-20 to also be
a lot lower, which it is.
PStamler
Which perhaps is why the change in loading had little or no effect on
most of the microphones we tried -- Sennheisers, E-Vs, Shure Beta
series, etc.. But the SM57 family was designed in a different era, and
the designers clearly relied on electrical damping more than
mechanical damping, on the assumption of a lower preamp input
impedance than is used nowadays.
As Scott noted, this is not only about the flattening of resonance
curves; it's also about the damping of the motor, and it manifests
itself as cleaner sound.
Peace,
Paul
Ian Bell
Yes but we are not talking about the difference between unloaded and
loaded but between a 2K load and a 680R load. In both case I bet the Q
is below 1.
Cheers
Ian
Ian Bell
You miss the point. I am suggesting that loading does nothing that EQ
will not achieve and with greater flexibility.
Cheers
Ian
Ian Bell
We are not disputing facts i.e that loading changes the response of an
SM57 - we are discussing the mechanism that causes it.
Cheers
Ian
Ian Bell
No, now you are assuming damping is the mechanism and there is no
evidence of that yet.
>
> As Scott noted, this is not only about the flattening of resonance
> curves; it's also about the damping of the motor, and it manifests
> itself as cleaner sound.
No, he suggests it may be, no one has done any tests to show if this is
the case.
Cheers
Ian
>
> Peace,
> Paul
Ian Bell
Agreed, but no one has made any measurements at all. All we have to go
on so far are some subjective listening tests which show that it sound
'different'. They tell you nothing about why. However, I do know the
inductance of tape replay heads is much higher than that of dynamic mics
and tape heads resonate in the over 20KHz region so it is reasonable to
suppose dynamic mics will have a resonance much higher than this. Damped
with 2K or 680R they will not affect the response in the audio band.
The simple rise of impedance with frequency due to the inductance of the
mic does explain what is 'heard' and this could equally be achieved with EQ.
Cheers
Ian
Scott Dorsey
>
>Yes but we are not talking about the difference between unloaded and
>loaded but between a 2K load and a 680R load. In both case I bet the Q
>is below 1.
Didn't measure it directly, but I sure saw a difference in ringing on
a scope. If you want to measure it, I'd be curious what you come up with.
geoff
And we are suggesting that it indeed does. Damping of the diaphram that has
an audible ( and beneficiary) effect on the unwanted results physical
excitation of the diaphram.
geoff
Scott Dorsey
>
>You miss the point. I am suggesting that loading does nothing that EQ
>will not achieve and with greater flexibility.
EQ doesn't stop ringing. It's not the right tool for that. It can change
ringing, mind you.
Scott Dorsey
>Agreed, but no one has made any measurements at all. All we have to go
>on so far are some subjective listening tests which show that it sound
>'different'. They tell you nothing about why. However, I do know the
>inductance of tape replay heads is much higher than that of dynamic mics
>and tape heads resonate in the over 20KHz region so it is reasonable to
>suppose dynamic mics will have a resonance much higher than this.
Not at all. You're missing the point, this is a MECHANICAL resonance.
It's being mechanically altered by the mechanical load of the coil in
the magnetic field, which is what is changing the MECHANICAL damping of
the system. It's not like a simple coil at all.
Mark
Ian,
I agree with you..
there are audible effects of load changes but they are electrical in
nature as discussed here:
http://www.jensen-transformers.com/an/an005.pdf
and I also agree with you that these audible effects have little to do
with mechanical resonances in the mic but rather have everything to do
with electrical loading effects and that these electrical effects can
be well approximated with standard EQ.
A lightly loaded SM57 with a piece of cable creates an ELECTRICAL
resonance as discussed in the paper above that creates high end
peaking. A lower load resistance can damp this ELECTRICAL resonance
and make the mic sound "warmer" just the same as EQ can.
The mechanical resonances in the mic are lightly coupled to the
electrical load so changes in the electrical load have almost no
effect on the MECHANICAL resonances.
Just to be clear, Ian, I agree with you.
Mark
hank alrich
But it does, and that is the point of suggesting you TRY it.
You assume this and that - so much for empirical empirical.
--
ha
shut up and play your guitar
hank alrich
Suspect in one hand and shit in the other. Which one fills up first?
PStamler
500 ohms, respectively. The 681 ohm resistor in parallel with a
typical tranformerless-preamp impedance of 1800-2200 gives about 500
ohms, which is why I suggested it as a first try.
As Scott says, we're dealing here with a mechanical resonance, NOT an
electrical resonance.
Mr. Bell should try the experiment before belittling it further. It
won't cost much; heck, he could add the resistor to a mic cable he
isn't using, and snip it out afterwards.
One thing that's important, though, is to match levels very carefully.
After setting the level on channel 1, we put the mic in front of a
guitar amplifier that had a 700Hz tone running through it. We used
that to match the levels of the two preamp channels (with their
different input Zs) to within a pointer-width on a standard VU meter,
or about 0.1dB. If you don't do the match that closely, the test is
meaningless.
Our two test instruments were a stripped-down drumkit (snare, cymbal
and kick) and an acoustic guitar, played with fingerpicks in a
percussive blues style (gentle bare-finger picking didn't tell us
squat). The difference was most audible on cymbal and guitar.
Peace,
Paul
Ian Bell
No, all you said was changing the loading makes a difference. I am not
disputing that.
Cheers
Ian
Ian Bell
> Ian Bell <ruffr...@yahoo.com> wrote:
>
>> geoff wrote:
>>> Ian Bell wrote:
>>>
>>>> Personally I doubt that is a factor. The Q of the average dynamic mic
>>>> is so low that changing the load this way will make little difference.
>>>>
>>> Given the easily-heard results with a humble SM57, a simple change in
>>> loading clearly does make a difference.
>>>
>>> If you are skeptical, try it !
>>>
>>> geoff
>>>
>>>
>> You miss the point. I am suggesting that loading does nothing that EQ
>> will not achieve and with greater flexibility.
>>
>> Cheers
>>
>> Ian
>
> But it does, and that is the point of suggesting you TRY it.
There is no evidence it does. All we have so far is subjective
statements. No one has made any measurements.
Cheers
Ian
Ian Bell
Well, if you want to throw mechanical resonances into the pot as well
that's fine by me. However, dynamic mic manufacturers aim to get the
main resonance as low as possible because the response drops like a
stone below it - they aim for as much of the response as possible to be
above resonance. Looking at the SM57 frequency response it clearly
drops off steeply below 200Hz so my guess it that is where the main
resonance is.
The response rises from 3KHz to 6KHz which looks to me like it is
proportional to rate of change of flux, then falls off rapidly due to
mechanical losses (mainly stiffness).
The only way this issues will be resolved is if someone cares to make
some measurements. Until then it is all speculation.
Cheers
Ian
Ian Bell
Thanks Mark
Cheers
Ian
Ian Bell
> Ian Bell <ruffr...@yahoo.com> wrote:
>> Yes but we are not talking about the difference between unloaded and
>> loaded but between a 2K load and a 680R load. In both case I bet the Q
>> is below 1.
>
> Didn't measure it directly, but I sure saw a difference in ringing on
> a scope. If you want to measure it, I'd be curious what you come up with.
> --scott
>
So when you say 'saw a difference in ringing on the scope' what exactly
were the test conditions?
Cheers
Ian
Ian Bell
Indeed, but until someone makes an actual measurement my shit is a good
as yours.
Cheers
Ian
Ian Bell
> And, just to be clear, the two channels had input impedances of 2k and
> 500 ohms, respectively. The 681 ohm resistor in parallel with a
> typical tranformerless-preamp impedance of 1800-2200 gives about 500
> ohms, which is why I suggested it as a first try.
>
> As Scott says, we're dealing here with a mechanical resonance, NOT an
> electrical resonance.
>
> Mr. Bell should try the experiment before belittling it further. It
> won't cost much; heck, he could add the resistor to a mic cable he
> isn't using, and snip it out afterwards.
>
I'll say it again, I do not dispute that loading an SM57 alters the way
it sounds. What we are discussion is the MECHANISM that causes that change.
> One thing that's important, though, is to match levels very carefully.
> After setting the level on channel 1, we put the mic in front of a
> guitar amplifier that had a 700Hz tone running through it. We used
> that to match the levels of the two preamp channels (with their
> different input Zs) to within a pointer-width on a standard VU meter,
> or about 0.1dB. If you don't do the match that closely, the test is
> meaningless.
>
> Our two test instruments were a stripped-down drumkit (snare, cymbal
> and kick) and an acoustic guitar, played with fingerpicks in a
> percussive blues style (gentle bare-finger picking didn't tell us
> squat). The difference was most audible on cymbal and guitar.
>
I never disagreed that there is an audible difference.
Cheers
Ian
> Peace,
> Paul
anahata
> PStamler wrote:
>
>> As Scott noted, this is not only about the flattening of resonance
>> curves; it's also about the damping of the motor, and it manifests
>> itself as cleaner sound.
>
> No, he suggests it may be, no one has done any tests to show if this is
> the case.
So somebody should repeat Paul's experiment with a good EQ in the chain,
and try and determine what EQ settings are needed on an SM57 loaded with
2k to make it sound identical to the unequalised SM57 with 600 ohm load.
Or indeed, whether such a match is possible, and whether it's repeatable
if you switch the mic for another SM57.
The resonances may be narrowband, high-Q affairs that are hard to cancel
with EQ, and even if you managed it, the same EQ setting would have to be
finely adjusted for each SM57 unit it's used with. The beauty of
resistive damping, when it works, is it doesn't have to be tuned to
anything.
--
Anahata
ana...@treewind.co.uk ==//== 01638 720444
http://www.treewind.co.uk ==//== http://www.myspace.com/maryanahata
Mike Rivers
> Yes but we are not talking about the difference between unloaded and
> loaded but between a 2K load and a 680R load. In both case I bet the Q
> is below 1.
Borrow an SM57 and try it. Tell me the input impedance of the preamp you'll
be using and I'll send you a resistor. Then perhaps you can better
analyze what
you hear. Or better yet, measure it.
Mike Rivers
> But it does, and that is the point of suggesting you TRY it.
> You assume this and that - so much for empirical empirical.
Paul's conclusion was based on listening. The effect of the resistor
that Scott describes is consistent with what Paul and his students
(and others who have tried it) heard. As to whether it's correct, or
the complete explanation, could be verified by measurement. I don't
know that this has been done by anyone yet.
Ferchrissake, it's a $100 mic. It costs a lot more than that to properly
test it in a suitable acoustic test chamber. You're certainly free to
mess with EQ to try to clean up the mic, or to not use it at all if you
feel that it's too much trouble.
Mike Rivers
> All we have so far is subjective
> statements. No one has made any measurements.
Including you. Unless you've tried "the magic load" you can't
even subjectively say that you can duplicate the results with
equalization.
I understand, though, that you like to argue and debate. We all
do, around here. We have no lives. It's what we do.
Mike Rivers
> Agreed, but no one has made any measurements at all. All we have to go
> on so far are some subjective listening tests which show that it sound
> 'different'. They tell you nothing about why.
Perhaps this is worth a more scientific study. Or perhaps not. There are
many
things in audio that have been accepted without knowing the reason why they
work. Sometimes the scientific explanation is discovered, sometimes not.
> However, I do know the
> inductance of tape replay heads is much higher than that of dynamic mics
> and tape heads resonate in the over 20KHz region so it is reasonable to
> suppose dynamic mics will have a resonance much higher than this.
Somewhere on the shelf I have an issue of R-E-P magazine with an article by
Peter Butt in which he described a very similar experiment in which he put
shunt resistors across the playback head of a 3M M-79 recorder. He was
looking at frequency response using a flux loop and found a resistor value
that flattened out the top end of the frequency response (as he measured it)
and he concluded by listening that his recorder sounded better with the
resistor
in place. He described it (perhaps incorrectly) as damping.
Sometimes we use familiar terms that aren't technically precise, but it
does help
to explain things to the layman. And most people in this field today are
laymen.
Shall we discuss "phase" now? Or "latency?" ;)
George's Pro Sound Co.
"Mike Rivers" <mri...@d-and-d.com> wrote in message
news:hde8se$l0i$2...@news.eternal-september.org...
actuially I almost had a life, then I got married --)
I love this thread
g
Scott Dorsey
>
>The mechanical resonances in the mic are lightly coupled to the
>electrical load so changes in the electrical load have almost no
>effect on the MECHANICAL resonances.
Do the measurements and see. I think you'll be shocked at what you see on
the scope. It really does ring.
While you're at it, do a swept impedance curve on the SM-57. That's an
eye-opener too.
Scott Dorsey
>
>Well, if you want to throw mechanical resonances into the pot as well
>that's fine by me. However, dynamic mic manufacturers aim to get the
>main resonance as low as possible because the response drops like a
>stone below it - they aim for as much of the response as possible to be
>above resonance. Looking at the SM57 frequency response it clearly
>drops off steeply below 200Hz so my guess it that is where the main
>resonance is.
If the mike is loaded properly, yes. The response plots that you see
on the data sheet are assuming about a 600 ohm load (Shure M67 or
the like).
>The response rises from 3KHz to 6KHz which looks to me like it is
>proportional to rate of change of flux, then falls off rapidly due to
>mechanical losses (mainly stiffness).
>The only way this issues will be resolved is if someone cares to make
>some measurements. Until then it is all speculation.
I have, as I said, made some basic qualitative measurements showing
ringing. If you want to do a full impulse response, I'd love to see it.
Scott Dorsey
>Scott Dorsey wrote:
>> Ian Bell <ruffr...@yahoo.com> wrote:
>>> Yes but we are not talking about the difference between unloaded and
>>> loaded but between a 2K load and a 680R load. In both case I bet the Q
>>> is below 1.
>>
>> Didn't measure it directly, but I sure saw a difference in ringing on
>> a scope. If you want to measure it, I'd be curious what you come up with.
>
>were the test conditions?
Little baby IEC chamber, tone burst source, 600 and 10K loads into
a Neff differential amp, old Tek 565 scope. Don't recall the burst
frequency but it was probably 5KC or so.
I'd use an acoustical square wave if such a thing existed, but it doesn't.
Ian Bell
I think the term 'magic load' you used sums it up. I really am not
enamoured of these simple changes that 'work magic' in a purely
subjective way. I much prefer to understand why and how the effect is
produced but that unfortunately demotes from magic to mere science.
Cheers
Ian
Ian Bell
> Ian Bell <ruffr...@yahoo.com> wrote:
>> Well, if you want to throw mechanical resonances into the pot as well
>> that's fine by me. However, dynamic mic manufacturers aim to get the
>> main resonance as low as possible because the response drops like a
>> stone below it - they aim for as much of the response as possible to be
>> above resonance. Looking at the SM57 frequency response it clearly
>> drops off steeply below 200Hz so my guess it that is where the main
>> resonance is.
>
> If the mike is loaded properly, yes. The response plots that you see
> on the data sheet are assuming about a 600 ohm load (Shure M67 or
> the like).
Are you certain. Looking at Shure's own data sheet all it says is:
"Rated impedance is 150Ω (310Ω actual) for connection to
microphone inputs rated low impedance"
There is no mention at all of the test cinditions under which frequency
response is measured.
>
>> The response rises from 3KHz to 6KHz which looks to me like it is
>> proportional to rate of change of flux, then falls off rapidly due to
>> mechanical losses (mainly stiffness).
>
>> The only way this issues will be resolved is if someone cares to make
>> some measurements. Until then it is all speculation.
>
> I have, as I said, made some basic qualitative measurements showing
> ringing. If you want to do a full impulse response, I'd love to see it.
And I have asked you to specify exactly how those measurements were made.
Cheers
Ian
> --scott
>
Ian Bell
> Ian Bell <ruffr...@yahoo.com> wrote:
>> Scott Dorsey wrote:
>>> Ian Bell <ruffr...@yahoo.com> wrote:
>>>> Yes but we are not talking about the difference between unloaded and
>>>> loaded but between a 2K load and a 680R load. In both case I bet the Q
>>>> is below 1.
>>> Didn't measure it directly, but I sure saw a difference in ringing on
>>> a scope. If you want to measure it, I'd be curious what you come up with.
>> So when you say 'saw a difference in ringing on the scope' what exactly
>> were the test conditions?
>
> Little baby IEC chamber, tone burst source, 600 and 10K loads into
> a Neff differential amp, old Tek 565 scope. Don't recall the burst
> frequency but it was probably 5KC or so.
So this was a burst of how many cycles of 5KHz and precisely what did
you see on the scope?
CHeers
Ian
Ian Bell
I agree, it needs to be measured. Until then no one is 'right'.
Cheers
Ian
Scott Dorsey
>
>I think the term 'magic load' you used sums it up. I really am not
>enamoured of these simple changes that 'work magic' in a purely
>subjective way. I much prefer to understand why and how the effect is
>produced but that unfortunately demotes from magic to mere science.
Then make the measurement.
Scott Dorsey
>Scott Dorsey wrote:
>> Ian Bell <ruffr...@yahoo.com> wrote:
>>> Well, if you want to throw mechanical resonances into the pot as well
>>> that's fine by me. However, dynamic mic manufacturers aim to get the
>>> main resonance as low as possible because the response drops like a
>>> stone below it - they aim for as much of the response as possible to be
>>> above resonance. Looking at the SM57 frequency response it clearly
>>> drops off steeply below 200Hz so my guess it that is where the main
>>> resonance is.
>>
>> If the mike is loaded properly, yes. The response plots that you see
>> on the data sheet are assuming about a 600 ohm load (Shure M67 or
>> the like).
>
>Are you certain. Looking at Shure's own data sheet all it says is:
>
>"Rated impedance is 150Ω (310Ω actual) for connection to
>microphone inputs rated low impedance"
>
>There is no mention at all of the test cinditions under which frequency
>response is measured.
That's true, and that's why all the confusion has occurred over the years.
But you'd expect that the companion mike mixer that Shure sold at the
time, for instance, would be a good match for it.
"Low impedance" would be something in the 500-600 ohm region. Olsen's
book has a good description of how bridging loads on microphones work.
Scott Dorsey
>Scott Dorsey wrote:
>> Ian Bell <ruffr...@yahoo.com> wrote:
>>> Scott Dorsey wrote:
>>>> Ian Bell <ruffr...@yahoo.com> wrote:
>>>>> Yes but we are not talking about the difference between unloaded and
>>>>> loaded but between a 2K load and a 680R load. In both case I bet the Q
>>>>> is below 1.
>>>> Didn't measure it directly, but I sure saw a difference in ringing on
>>>> a scope. If you want to measure it, I'd be curious what you come up with.
>>> So when you say 'saw a difference in ringing on the scope' what exactly
>>> were the test conditions?
>>
>> Little baby IEC chamber, tone burst source, 600 and 10K loads into
>> a Neff differential amp, old Tek 565 scope. Don't recall the burst
>> frequency but it was probably 5KC or so.
>
>So this was a burst of how many cycles of 5KHz and precisely what did
>you see on the scope?
Probably around a .1 second burst. What I saw on the scope was a nice
ringing on the trailing edge of the waveform, and that ringing was not at
the excitation frequency.
Again, it's been quite a few years since I did any of this.... I did some
basic testing when Paul Stamler's article first came out. A quick
qualitative test made it clear to me that something was going on which
was consistent with Paul's article and that's all I really cared about at
the time. If that's not enough for you, I urge you to make your own
measurements.
Don Pearce
If I can lay my hands on a 57 or 58 in the next few days, I will use
my spark impulse generator on it and take a response, with and without
a heavy resistive load. I'll have to hope that the main coupling mode
is acoustic and not electrical.
d
Ian Bell
> Ian Bell wrote:
>
>> Agreed, but no one has made any measurements at all. All we have to go
>> on so far are some subjective listening tests which show that it sound
>> 'different'. They tell you nothing about why.
>
> Perhaps this is worth a more scientific study. Or perhaps not. There are
> many
> things in audio that have been accepted without knowing the reason why they
> work. Sometimes the scientific explanation is discovered, sometimes not.
>
> > However, I do know the
>> inductance of tape replay heads is much higher than that of dynamic
>> mics and tape heads resonate in the over 20KHz region so it is
>> reasonable to suppose dynamic mics will have a resonance much higher
>> than this.
>
> Somewhere on the shelf I have an issue of R-E-P magazine with an article by
> Peter Butt in which he described a very similar experiment in which he put
> shunt resistors across the playback head of a 3M M-79 recorder. He was
> looking at frequency response using a flux loop and found a resistor value
> that flattened out the top end of the frequency response (as he measured
> it)
> and he concluded by listening that his recorder sounded better with the
> resistor
> in place. He described it (perhaps incorrectly) as damping.
I have a very similar article in a 1940s edition of Wireless World which
goes into a lot more technical (and mathematical) detail but the bottom
line is that there is an optimum load resistance for best frequency and
phase response and this is related to the head resonance.
>
> Sometimes we use familiar terms that aren't technically precise, but it
> does help
> to explain things to the layman. And most people in this field today are
> laymen.
>
> Shall we discuss "phase" now? Or "latency?" ;)
>
NO, not again!!
Cheers
Ian
Arkansan Raider
> Ian Bell <ruffr...@yahoo.com> wrote:
>> I think the term 'magic load' you used sums it up. I really am not
>> enamoured of these simple changes that 'work magic' in a purely
>> subjective way. I much prefer to understand why and how the effect is
>> produced but that unfortunately demotes from magic to mere science.
>
> Then make the measurement.
> --scott
>
Sounds like work to me. And that's an ugly four-letter word.
/popcorn, sodypop, theater seat, watching the fireworks.
---Jeff
Mark
>
> Our two test instruments were a stripped-down drumkit (snare, cymbal
> and kick) and an acoustic guitar, played with fingerpicks in a
> percussive blues style (gentle bare-finger picking didn't tell us
> squat). The difference was most audible on cymbal and guitar.
>
> Paul
Yes you can HEAR a difference, that is not in question.
What is in question is the cause of those differences.
Changing the loading can make changes to the frequency response that
you can hear due to electrical effects that can also be accomplished
by EQ. That was the key question.
There is no evidence that the audible change has anything to do with
mechanical resonances.
Mark
David Aguilera
> There's always some wise guy in the audience. You're correct of course,
> but I
> didn't mention that in my reply for two reasons:
>
> 1 - the guy probably isn't going to make the measurement as I described
> anyway
> 2 - He's going to ask "what's a good generator that I can buy from eBay?
>
> I'd probably suggest that he use a DAW program to generate a sine wave
> and use his sound card as a generator. If it's something fancier than the
> sound card built into his computer, it will probably have a source
> impedance
> of around 20 ohms.
>
> Alternately, he can build an attenuator which he may need anyway to get the
> generator output low enough not to clip the preamp, and make the shunt
> resistor
> low enough so the generator source impedance is insignificant.
>
Mike,
I was thinking of something along the lines of this:
http://www.zen22142.zen.co.uk/Theory/inzoz.htm.
I have a MOTU 896HD and would use Logic 9's utilities for a generator.
Would that be adequate?
David
David Aguilera
> This is just Th�venin's Equivalent. Write it out.
>
> If you want a 600 ohm load, solve for Z as the shunt knowing Zpre:
>
> 600 = 1/((1/Zpre)+(1/Z))
> or
> Z = 1/(0.001667 + 1/Zpre)
Very helpful!
Thank you!
David Aguilera
> geoff wrote:
>
>> Thought you were advocating testing in input impedence of a preamp by
>> hooking a multimeter to it and simply reading the R of the scale..
>
> I think that's what the original poster was dreaming about, but he's
> been told
> that he can't. But he can still make use of his multimeter to make the
> measurement.
> It just takes some extra stuff.
I understood that I could not beforehand, which was why I mentioned AC
electronics.
David Aguilera
> On Nov 9, 6:15 pm, Mike Rivers <mriv...@d-and-d.com> wrote:
>> geoff wrote:
>>> An XLR-M to XLR-F 'barrel' adaptor is tidier.
>> But you'd have to buy it. You can make a short cable out of
>> a long cable, even one that you've junked because it's faulty,
>> as long as you don't use the faulty part.
>
> As the guy who started some of this discussion with an article called
> "The Taming of the Shure" in Recording Magazine, let me weigh in with
> a couple of suggestions.
>
> 1. I don't recommend using barrel connectors; if you put them on the
> preamp/mixer end of the cable, one yank of the cable can badly torque
> the XLR on the mixer/preamp, possibly causing expensive damage. A
> short cable is safer.
>
> 2. Solder the resistor between pins 2 and 3 of the male connector on
> the cable. I recommend the male connector because there's a bit more
> room inside the shell. Even so, you'll need to snip the leads off real
> short. Make sure that, when you solder in the resistor, the cable
> doesn't come unsoldered. This is a lot easier to do properly with a
> Neutrik connector rather than Switchcraft.
>
> 3. If you look at your mixer/preamp's manual, you should find the
> input impedance listed and can compute the proper parallel resistor.
> In the absence of more data, though, for an SM57 a 681 ohm (NOT 681k!)
> resistor will usually give good results.
>
> 4. The improvement is mostly in the area of cleanness and clarity, and
> lack of high-frequency "hashiness". In the comparison I originally
> did, we listened to the SM57 on a drumkit; with a standard load (2k)
> the cymbals sounded more like white noise than the metallic ring a
> cymbal should have. With the parallel resistor, it sounded much more
> like a real cymbal. On acoustic guitar switching to the parallel
> resistor channel took away a lot of the "ringy" quality at high
> frequencies.
>
> 5. I've only found a few microphones that are helped by this
> technique; the most important was the SM57 family (SM56 and SM58
> included). The AKG D112 got a lot less "pillowy" and a lot tighter on
> the bottom, but the jury's out whether that was an improvement. Among
> the mics for which this makes little or no difference are Electro-
> Voice RE15 and RE20 and Shure Beta57 It's not recommended for
> condenser mics.
>
> Peace,
> Paul
Thanks!
zero
"Ian Bell" <ruffr...@yahoo.com> wrote in message
news:hde267$cbj$1...@localhost.localdomain...
> geoff wrote:
>> Ian Bell wrote:
> No, all you said was changing the loading makes a difference. I am not
> disputing that.
Ian,
What you appear to be saying, is applying EQ can produce
the same results.
Ok then. Loading a 1/4 ton in the back of your
pick-up truck will change the behavior of the suspension
in a way that air shocks and an air-ride chair will clearly
NOT make you feel the truck is now without weight.
The 1/4 ton is the parallel resistor, the air-ride &
air-seat are your EQ. You can't simulate a 1/4 ton
of load, without a 1/4 ton of real load. Nor can you
eliminate it.
Say you have the perfect photo of person/object, and the
deal breaker is the background or foreground lighting,
do you fix the lighting,,, or fix the picture in Photoshop?
(btw, I know of no EQ/plug-in that can function as remotely close
in audio, as a photo-editing software does with regards
to pictures.)
So. As to what says you.
Now, I can take the stock crossover in
my one 15" Tannoy, add or remove components, and
then simply patch an EQ into it, to match the
other 15" Tannoy? ...Really? :-P
--
-zero
zero
> I think the term 'magic load' you used sums it up. I really am not
> enamoured of these simple changes that 'work magic' in a purely subjective
> way. I much prefer to understand why and how the effect is produced but
> that unfortunately demotes from magic to mere science.
Please read my response to your post earlier in this thread.
--
-zero
David Aguilera
> Scott,
>
> You mentioned this the other day:
>
> > [snip]
> >
> >Buy some of the old classics. Everyone should have an SM-57, an EV
> >635A, and maybe an RE-20. The first two are very cheap, and all three
> >can be found used. The first two are very touchy about loading and
> >will benefit from a shunt resistor in a barrel connector if you are
> >using a modern transformerless preamp.
> >
> > [snip]
>
> I'm just getting into the electrical side of this field and am really
> curious as to how to go about building the shunt resistor. If you have
> the time to enlighten me I would greatly appreciate it.
>
> The preamp I am using is not giving me what I want out of my 57's and
> EV635A. I've always thought it was an impedance problem, but If I could
> build a box with switchable resistance...well--wow! That's
> straightforward and downright useful!
>
> I don't have a anything that will measure impedance, so I am unsure of
> the impedance of my preamps. Is there a way to measure it with a my
> multimeter sans one?
>
> After that it's like connecting resistance in serial right (I'm really
> weak in AC electronics)?
>
> Thanks,
>
> David
I didn't think this would open the can of worms it seems to have. :)
zero
>> Suspect in one hand and shit in the other. Which one fills up first?
>>
>
>
> Indeed, but until someone makes an actual measurement my shit is a good as
> yours.
Remember...
Speakers>crossover>swap-o-rama>fix with EQ>not so much.
--
-zero
zero
"Mark" <mako...@yahoo.com> wrote in message
news:b727f2ba-e954-4a86...@p28g2000vbi.googlegroups.com...
>
> Yes you can HEAR a difference, that is not in question.
>
> What is in question is the cause of those differences.
>
> Changing the loading can make changes to the frequency response that
> you can hear due to electrical effects that can also be accomplished
> by EQ. That was the key question.
>
> There is no evidence that the audible change has anything to do with
> mechanical resonances.
And if the SM-57 was a loudspeaker for toy solders,
How would you properly correct the crossover network
solely using EQ?
--
-zero
drichard
I look forward to seeing your results. I did the load change mod on a
mic cord a couple of years ago, and have used it with my SM58. To my
"layman" ears it sounds like more than an EQ change to me, but I've
been fooled before. To me it sounds like the mic is "tighter" which is
consistent with what Scott saw.
Question for all: Doesn't the level change as well, when using the
shunt resistor?
Thanks,
Dean
On Nov 11, 8:11 am, s...@spam.com (Don Pearce) wrote:
> On 11 Nov 2009 09:05:59 -0500, klu...@panix.com (Scott Dorsey) wrote:
>
>
>
>
>
> >Ian Bell <ruffreco...@yahoo.com> wrote:
> >>Scott Dorsey wrote:
> >>> Ian Bell <ruffreco...@yahoo.com> wrote:
> >>>> Scott Dorsey wrote:
> d- Hide quoted text -
>
> - Show quoted text -
Scott Dorsey
>
>If I can lay my hands on a 57 or 58 in the next few days, I will use
>my spark impulse generator on it and take a response, with and without
>a heavy resistive load. I'll have to hope that the main coupling mode
>is acoustic and not electrical.
If you do it with a 58, take the ball off. The on-axis response is
changed dramatically by the ball and not in a good way.
If possible, I'd be curious if you could do it with the internal
transformer removed as well. That would tell for sure if the ringing
issue was due to the poor quality transformer or the mechanical issues.
drichard
and to me it seems enlightening. The difference between an EQ change
and a change in the damping is worth discussing.
On Nov 11, 10:49 am, David Aguilera
> I didn't think this would open the can of worms it seems to have. :)- Hide quoted text -
Don Pearce
<DRic...@wi.rr.com> wrote:
>Question for all: Doesn't the level change as well, when using the
>shunt resistor?
Yes, but a very minor tweak on the preamp gain fixes that.
d
Scott Dorsey
>
>Question for all: Doesn't the level change as well, when using the
>shunt resistor?
Yes, and so the S/N is reduced a little. A better way to do this is
to use a step-up transformer, which gets you proper loading without
losing signal. Problem is, good transformers are expensive, and
resistors are cheap. You pays your money and you takes your chance.
A good transformer input preamp like the John Hardy is going to give
you a lower noise floor than a Mackie with a shunt resistor, and the
tone won't change when you adjust the gains like it does on the Mackie
either. But you can buy a lot of channels of Mackie for the cost of
one channel of the John Hardy. Life's just like that.
Don Pearce
Even a good transformer will lose you about 1dB of signal, which is
probably no more than the increased loading will cause.
d
David Aguilera
> I don't consider it a bad thing. The discussion seems mostly civil,
> and to me it seems enlightening. The difference between an EQ change
> and a change in the damping is worth discussing.
>
I'm just happy to get all of this information from what I previously
thought was a much simpler question.
hank alrich
No matter how many Mackie preamp channels one buys for their money, they
will not get even in the same ballpark as a Hardy preamp if intended use
includes plenty of dynamic mics.
--
ha
shut up and play your guitar
hank alrich
At some point, just to get audio work of the recording kind accomplished
some of us have to make decisions based on what we hear, on the spot. It
is a simple thing to hear the difference loading makes on those old
Shures. Move one from a 1202 to a Great River and one could think one
switched mics instead of preamps.
hank alrich
> Mike Rivers wrote:
> > Ian Bell wrote:
> >
> >> All we have so far is subjective statements. No one has made any
> >> measurements.
> >
> > Including you. Unless you've tried "the magic load" you can't
> > even subjectively say that you can duplicate the results with
> > equalization.
> >
> > I understand, though, that you like to argue and debate. We all
> > do, around here. We have no lives. It's what we do.
>
>
> I think the term 'magic load' you used sums it up. I really am not
> enamoured of these simple changes that 'work magic' in a purely
> subjective way. I much prefer to understand why and how the effect is
> produced but that unfortunately demotes from magic to mere science.
>
> Cheers
>
> Ian
Figuring out why the effect results from a change in load in no way
negates the outright sonic MAGIC of the change.
hank alrich
> And, just to be clear, the two channels had input impedances of 2k and
> 500 ohms, respectively. The 681 ohm resistor in parallel with a
> typical tranformerless-preamp impedance of 1800-2200 gives about 500
> ohms, which is why I suggested it as a first try.
>
> As Scott says, we're dealing here with a mechanical resonance, NOT an
> electrical resonance.
>
> Mr. Bell should try the experiment before belittling it further. It
> won't cost much; heck, he could add the resistor to a mic cable he
> isn't using, and snip it out afterwards.
>
> One thing that's important, though, is to match levels very carefully.
> After setting the level on channel 1, we put the mic in front of a
> guitar amplifier that had a 700Hz tone running through it. We used
> that to match the levels of the two preamp channels (with their
> different input Zs) to within a pointer-width on a standard VU meter,
> or about 0.1dB. If you don't do the match that closely, the test is
> meaningless.
>
> Our two test instruments were a stripped-down drumkit (snare, cymbal
> and kick) and an acoustic guitar, played with fingerpicks in a
> percussive blues style (gentle bare-finger picking didn't tell us
> squat). The difference was most audible on cymbal and guitar.
>
> Peace,
> Paul
Transients seem to bring out the differences IME, based on 57's and 58's
through Mackies and A&H's, versus Great River MP2-MH and MP4, Neve
modules, the Phoenix/Aurora GTQ2, etc.
hank alrich
> hank alrich wrote:
> > Ian Bell <ruffr...@yahoo.com> wrote:
> >
> >> geoff wrote:
> >>> Ian Bell wrote:
> >>>
> >>>> Personally I doubt that is a factor. The Q of the average dynamic mic
> >>>> is so low that changing the load this way will make little difference.
> >>>>
> >>> Given the easily-heard results with a humble SM57, a simple change in
> >>> loading clearly does make a difference.
> >>>
> >>> If you are skeptical, try it !
> >>>
> >>> geoff
> >>>
> >>>
> >> You miss the point. I am suggesting that loading does nothing that EQ
> >> will not achieve and with greater flexibility.
> >>
> >> Cheers
> >>
> >> Ian
> >
> > But it does, and that is the point of suggesting you TRY it.
>
>
> There is no evidence it does. All we have so far is subjective
> statements. No one has made any measurements.
>
> Cheers
>
> Ian
> >
> > You assume this and that - so much for empirical empirical.
> >
Deaf mo'fo's don't work long in the studio. Blind? Yes. Deaf? No.
Shit, if I had to stop and measure for evey decision I made last week,
let alone for each collective sonic choice made by all involved, we'd
still be trying to get around to tracking the second song. As it is
we'll do three tiny tweaks of 1 dB or less this afternoon and we're
wrapped in production but for mastering, tomorrow.
I don't dismiss science or the value of testing, ABX or otherwise. But
when work needs to get done few of us can stop the show to drag out the
test gear unless we aspire to be labelled an "Unproducer".
hank alrich
> hank alrich wrote:
> > Ian Bell <ruffr...@yahoo.com> wrote:
> >
> >> Mike Rivers wrote:
> >>> Ian Bell wrote:
> >>>
> >>>> Since when has a 'standard' load been 2K??
> >>> It's sort of a roughly weighted average of the input impedance of
> >>> a good many modern transformerless preamps and mixing consoles.
> >>> This is typical of the gear to which an SM57 is likely to be connected,
> >>> at least among the Recording Magazine readership, where Paul's
> >>> article was published.
> >>>
> >>> Perhaps "typical" would have been a better term, unless (and I don't
> >>> recall the details of the study) a specific load resistor for each preamp
> >>> was chosen so that the mic would be loaded by 2K for any preamp
> >>> in the study. This would allow the preamp to be the major source of
> >>> differences in sound when used with an SM57.
> >>>
> >>> Further listening experiments indicated that the mic sounded better
> >>> when loaded with 600 ohms (adjusted for the preamp used for testing).
> >>>
> >>>> Dynamic mics are basically inductive which means their impedance rises
> >>>> with frequency. By loading it with 680R instead of 2K you effectively
> >>>> hit the high frequency response - something you can do with infinitely
> >>>> more precision using EQ.
> >>> More important, you change the damping of the diaphragm assembly,
> >>> which broadens resonant peaks and may even change the frequency
> >>> of some resonances. That's not easily done with EQ.
> >>>
> >>
> >> Indeed, but as I said elsewhere, I suspect the Q of the average dynamic
> >> mic is so low that this is not a factor.
> >>
> >>
> >> Cheers
> >>
> >> ian
> >
> > Suspect in one hand and shit in the other. Which one fills up first?
> >
>
>
> Indeed, but until someone makes an actual measurement my shit is a good
> as yours.
>
> Cheers
>
> Ian
I got one hand full of experience and the other full of music.
Arkansan Raider
Preach it.
---Jeff
Arkansan Raider
Okay, I feel like this might be a dumb question based on the price
level, but I've been looking at the ART MPA and I'm wondering where its
quality falls on the Mackie/John Hardy continuum.
The thing that got my attention was the variable input impedance, and it
seems to have gotten some good reviews.
Any takers?
---Jeff
Mike Rivers
> I much prefer to understand why and how the effect is
> produced but that unfortunately demotes from magic to mere science.
Really, I do too. But I don't let the lack of pure scientific understanding
get in the way of using sometking that works for me.
I'm glad hummingbirds can fly, too.
Don Pearce
wrote:
>Mike Rivers wrote:
>> Ian Bell wrote:
>>
>>> All we have so far is subjective statements. No one has made any
>>> measurements.
>>
>> Including you. Unless you've tried "the magic load" you can't
>> even subjectively say that you can duplicate the results with
>> equalization.
>>
>> I understand, though, that you like to argue and debate. We all
>> do, around here. We have no lives. It's what we do.
>
>
>I think the term 'magic load' you used sums it up. I really am not
>enamoured of these simple changes that 'work magic' in a purely
>subjective way. I much prefer to understand why and how the effect is
>produced but that unfortunately demotes from magic to mere science.
>
You mean promotes mere magic to real science, surely?
d
Mike Rivers
> I have a very similar article in a 1940s edition of Wireless World which
> goes into a lot more technical (and mathematical) detail but the bottom
> line is that there is an optimum load resistance for best frequency and
> phase response and this is related to the head resonance.
It's easier to understand and explain when dealing with tape heads, because
there's no (at least there should be no) mechanical part to the frequency
and phase response anomalies. But this is a different kind of source than a
microphone.