[Elecraft] SSB transmit audio - Where's the punch?
Dick, WN3R
his FT-1000MP. He claims the K-3 audio will never be as good as Yaesu,
Kenwood, or ICOM.
Yesterday we spent several hours between ourselves and several European
stations tuning and testing. I never got a bad audio report, BUT I never
got a report I had the same PUNCH!!! He reduced power and still overtalked
me.
I have tried HC-4, HC-5, and HC-6 using W2IHY eight band EQ with the EQPlus
with some compression from the K-3.
I need some ideas before the next contest.
73, Dick, WN3R
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elec...@mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
John Ragle
> I need some advice in how to get the same punch my fellow contester gets on
> his FT-1000MP. He claims the K-3 audio will never be as good as Yaesu,
> Kenwood, or ICOM.
Dick...
What level of compression are you using? I use a setting of around
20 or higher, as suggested by some folks at Elecraft. For a
peanut-whistle, the K3 has good audio "punch" at this setting.
Playing with the equilizers is probably not too useful, but for
most punch, you want a good fraction of your audio in the treble
mid-range, without wasting power on the lows, which don't translate into
intelligibility. On average, a male speaker's voice peaks in spectral
power around 500 Hz, within a broad band between 250 Hz and 1000 Hz.
These are the parts you want to emphasize wrt "punch," e.g.
intelligibility. Obviously the opposite of "high fidelity" reproduction,
which seems to be a current fad, at least on this reflecftor.
John Ragle -- W1ZI
w5...@w5ov.com
You use "some compression"? Crank it up!
Dump the external EQ, and use either the HC-4 or HC-5 (no bias) and crank
the TX EQ bands up or down until it sounds nasty enough. You'll generally
reduce the lows and increase the highs.
Crank up the compression and it'll sound as bad as the Yaesu does.
Your fellow contester's claim reveals his lack of knowledge about the K3
as it can be adjusted to sound any way you want, with nearly any
microphone.
Generally speaking, if it doesn't sound right, crank it up.
73,
Bob W5OV
P.S. I know that audio purists will not like the above advice. But, you
are looking to bust pileups, not get an audio quality award.
Rob May
The K3 has sufficient equalization to tailor the sound of any microphone, the external equalizer is completely unneeded in my opinion. My cheap-o computer gaming headset sounds really good (through the monitor) and gets great reports on the air. When chasing DX or during a contest I'll turn the compression up to 25. I've heard my signal through a couple of different radios that have their output online. The audio is surprisingly good. Setting the compression that high really makes a huge difference in average output and it makes a big difference on the other end.
Rob
NV5E
Tim Tucker
output "
It makes *some* difference in average output. There is nothing I have been
able to do to get the K3's average output anywhere close to what my Icom 746
Pro would do. The best average power numbers I have seen on my K3 with
various mic and settings combos are about 40-45 watts on specific syllables
or words. Typically I see about 30 watts with normal speaking. I have
tried with the built-in processing, with outboard compression, a combination
of the two, various mics, etc. On my Icom 746 Pro, I see 65 watts AVG with
the same mic and outboard compression. I spent hours doing the tests into
a dummy load, even comparing adding a Yaesu 897D into the mix for comparison
purposes.
The K3 sounds great IMO, but it does not have the same AVG SSB output power
of some other rigs available. That may or may not be important to you, so
decide for yourself how much stock you put in that. There are more details
of my findings here:
http://www.worldwidedx.com/hf-bands-hf-rigs/73806-elecraft-k3-review-2.html#post228882.
Personally, I think it would be interesting to see someone do the same tests
using some better test bench gear. I've sold most of mine...
Tim
David Gilbert
I also use a cheap computer headset (I've only found one single electret
headset out of several I've tried that ever sounded badly) with mic gain
set to 20 and compression set to 20. A compression setting of 25 for
my voice still gives decent audio, but I can definitely hear the onset
of noticeable distortion at that setting. I'm sure most people would
find it perfectly acceptable, though.
By the way, individual voice characteristics and assertiveness have a
LOT to do with how much "punch" a person has in their signal. I do a
lot of contesting and I hear lots of folks with strong signals who don't
speak strongly enough to do their signal justice. It has nothing to do
with how much gain or compression they are using ... they just sound
weak period. Listen to the DX multi-multi's who keep their mic gain low
enough that they don't distort but practically yell into the mic and
you'll see what I mean. The ones I'm referring to totally punch
through the noise and QRM without distortion even when their signal
isn't at killer strength. To be clear here, I'm not referring to
those several contest stations that have their gain and compression set
so high they splatter +/- 8 or more KHz.
I'm no audio expert, but I suspect that speaking more strongly adds more
spectrum to someone's voice, probably shifting the bulk of the energy
upwards toward higher frequencies. Maybe I'll do some spectrum tests
some time with my own voice to investigate it. I already have one
recent example where I was trying to help a guy address an issue he had
with weak and marginally choppy audio. He was transmitting while a
friend of his was in the shack making various adjustments (I think they
were even using a K3), but almost nothing they did helped very much.
The friend took over the mic to make a comment and the audio immediately
improved dramatically in both clarity and punch. They switched back to
the first operator and when he tried to speak more forcefully he sounded
much better, but for him it just wasn't "natural" and he couldn't
maintain it very long.
Whatever it is, it makes a difference and I'm not sure comparing one rig
to another using different operators is a good test.
73,
Dave AB7E
> On Mon, Oct 4, 2010 at 11:13 AM, Rob May<roblit...@hotmail.com> wrote:
>
>> The K3 has sufficient equalization to tailor the sound of any microphone,
>> the external equalizer is completely unneeded in my opinion. My cheap-o
>> computer gaming headset sounds really good (through the monitor) and gets
>> great reports on the air. When chasing DX or during a contest I'll turn the
>> compression up to 25. I've heard my signal through a couple of different
>> radios that have their output online. The audio is surprisingly good.
>> Setting the compression that high really makes a huge difference in average
>> output and it makes a big difference on the other end.
>> Rob
>> NV5E
David Gilbert
By the way, having more "punch" does not necessarily add intelligibility
.... it all depends upon the voice and where the additional energy goes
in the spectrum.
73,
Dave AB7E
Joe Subich, W4TV
> I need some advice in how to get the same punch my fellow contester
> gets on his FT-1000MP. He claims the K-3 audio will never be as good
> as Yaesu, Kenwood, or ICOM.
That should probably be restated as will never be as dirty as Yaesu,
Kenwood or Icom since the ALC circuits all three radios actually work
by overdriving the final amplifiers BEFORE they start to produce any
ALC action. In effect, their final amplifiers act like an unfiltered
RF clipper.
That said, the K3 is certainly capable of providing significant "punch"
if adjusted properly. With nearly any mic adjust the lower two bands
in the K3 equalizer to the minimum and adjust the top three for the
amount of high frequency boost desired (EQ is before the clipping
function, if I remember correctly so it sets the "balance" of highs
and lows). The HC-4 will need less high frequency boost than an HC-5
which needs less boost than the new HC-6.
Adjust the mic gain until you see 5 to 6 bars of ALC action and then
add compression until you see 10 to 15 dB indicated on the COMP
display (setting between 20 and 30). Control power output with the
power control and adjust TXG VCE for +5. to +1.5 dB to bring the
peak power on SSB (using a pulser) to the same level as peaks on
CW (using CW dots).
For a better understanding of the K3 speech processor see the study
by Jack, W8OZA at:
<http://cliftonlaboratories.com/elecraft_k3_speech_processing.htm>
73,
... Joe, W4TV
On 10/4/2010 1:41 PM, Dick, WN3R wrote:
Grant Youngman
On Oct 4, 2010, at 2:00 PM, Tim Tucker wrote:
> "Setting the compression that high really makes a huge difference in average
> output "
>
> On my Icom 746 Pro, I see 65 watts AVG with
> the same mic and outboard compression. I spent hours doing the tests into
> a dummy load, even comparing adding a Yaesu 897D into the mix for comparison
> purposes.
Just guessing here, but 65 watts of "average" power out with 100 watts peak must sound pretty awful ..
Grant/NQ5T
Tim Tucker
of what it looks like on a P3 around here somewhere. Yep, you're "just
guessing" :) :)
Don Wilhelm
takes an inordinate length of time to make the call and report intelligible.
73,
Don W3FPR
Nate Bargmann
their diaphram into the act and speak with confidence, not just voice
volume. I know some folk who will speak at a good volume until they
grab the mic and then seem to drop 10 dB or so by speaking in almost a
hushed tone of voice. As amateur rigs generally don't use compression
amps in the mic circuits, a strong, confident voice level is essential
to modulating the radio correctly. I tell new ops to "belly up" to the
microphone as an SSB rig's power output is directly related to how well
they speak in the mic. Some seem to get it and others don't.
Even on FM I have tried to tell ops to speak up which they will for a
bit and then lapse back into a whisper which is next to impossible to
copy in a noisy vehicle.
73, de Nate >>
--
"The optimist proclaims that we live in the best of all
possible worlds. The pessimist fears this is true."
Ham radio, Linux, bikes, and more: http://n0nb.us/index.html
Jim Miller KG0KP
contests that are very strong and they are almost impossible to understand
because they have their compression WAY to high. Turning it down would make
them much better.
I just made 1752 contacts in a special event and got many unsolicited
comments on my audio being very good, excellent, etc. I was running the HC4
and the mic gain at 25 and the compression at 25. My voice is relatively
low.
73, de Jim KG0KP
Gary Gregory
Could this be alleviated when using a Turner +3 microphone.
in the case of a soft spoken operator?
After reading your post I went and plugged the Turner in and immediately
noticed a more "punchy" audio..
Does this sound right to you?..I am definitely guilty of speaking too
softly..:-(
73's
Gary
--
Gary
VK4FD - Motorhome Mobile
http://www.qsl.net/vk4fd/
K3 #679
For everything else there's Mastercard!!!
Joe Subich, W4TV
> Could this be alleviated when using a Turner +3 microphone.
> in the case of a soft spoken operator?
Only of the "soft spoken operator" is in a very quiet shack.
The typical problem is that the operator's voice is barely
louder than the background noise ... blowers, fans, other
radios, etc ... and the amplified microphones bring up that
background noise as much as the operator's voice.
73,
... Joe, W4TV
David Gilbert
Totally true. I've probably previously cited the example from last
winter's NAQP contest where I heard a Midwest guy endlessly calling CQ
with very few callers even though he had a booming signal here in
Arizona. The problem was he had so much mic gain and compression nobody
could understand his callsign. I gave him a call and told him that, so
he kept cranking it down until he was fully intelligible. He was more
than 12 db weaker than when he started (and it wasn't due to QSB) but he
immediately started getting replies.
I came across another example in the CQP this past weekend. I was
running and the caller had so much compression that I couldn't make out
his suffix even after a few tries. I told him to back down the mic
gain, he did, and he sounded way better.
People who set their mic gain and compression levels to maximize their
output power are a scourge on our hobby.
73,
Dave AB7E
juergen
Oh no, we dont any more LIDS with amplified CB microphones on the bands . The ESSB garbage produces enough splatter as it is.
We dont need to encourage the ""all knobs to the right"" brigade with amplified microphones.
The K3 has HI and LOW mic gain settings, and thats more than enough with any decent microphone.
73
John
--- On Mon, 10/4/10, Gary Gregory <gary...@gmail.com> wrote:
> From: Gary Gregory <gary...@gmail.com>
> Subject: Re: [Elecraft] SSB transmit audio - Where's the punch?
James Sarte
increasing compression and using an amplifier. Too much, and the amp goes
out of linearity resulting in "flat topping" as can be easily seen on a
bandscope.
Just a thought...
--
73 de James K2QI
President UNARC/4U1UN
Eric Swartz - WA6HHQ, Elecraft
modes (like ESSB) from this thread. Please keep it friendly here.
Its OK to discuss the technical merits of various settings and operating
methods, but it is inappropriate to call anyone a lid or to characterize
anyone as a scurge or a mode like ESSB as garbage. Please take comments
and opinions like those to other forums.
73, Eric
Elecraft list moderator
----
On 10/5/2010 1:42 AM, juergen wrote:
> Hi Gary
>
> Oh no, we dont any more LIDS with amplified CB microphones on the bands . The ESSB garbage produces enough splatter as it is.
Jim Brown
> > Could this be alleviated when using a Turner +3 microphone.
> > in the case of a soft spoken operator?
>
> Only of the "soft spoken operator" is in a very quiet shack.
> The typical problem is that the operator's voice is barely
> louder than the background noise ... blowers, fans, other
> radios, etc ... and the amplified microphones bring up that
> background noise as much as the operator's voice.
There are several good solutions for background noise, and they can be
used in combination.
1) ALWAYS work VERY close to the mic, and turn down the mic gain to
compensate. Sound levels drop by 6dB for each doubling of distance, so
your voice gets louder while the background noise remains the same.
2) ALWAYS roll off the low frequencies (below about 350 Hz). They
provide NO useful intelligibility, but contain a lot of the noise (and
they waste TX power).
3) Use a directional (cardioid) microphone, and speak straight into it.
A directional mic rejects roughly 6dB of the noise.
4) Use only enough compression so that the compression meter on your rig
indicates 10dB compression on peaks. Do NOT use your SWR meter to gauge
compression. Compression "turns up the gain" during the quieter part of
your speech, so it makes background noise louder.
Dave, AB7E, said:
>I was running and the caller had so much compression that I couldn't
>make out his suffix even after a few tries. I told him to back down the
>mic gain, he did, and he sounded way better.
I have the same problem in every SSB contest. I find that at least one
third of all callers are so badly overdriven that they are hard to copy.
And I make a point of telling each of them to turn down their audio to
make it easier to copy.
73, Jim K9YC
Jim Brown
> I'll turn the compression up to 25.
That is the WRONG WAY to set compression. Rather, compression should be
set using the COMPRESSION section of the K3's meter (the one that's to
the left of the ALC indicator.
Grant said:
>Just guessing here, but 65 watts of "average" power out with 100 watts
>peak must sound pretty awful ..
You're absolutely right, Grant. Any good audio (or broadcast) engineer
knows that speech has a peak-to-average ratio on the order of 24dB. That
ratio can be reduced (using peak limiting and compression) by 10-14dB
before it starts sounding bad. Broadcasters use VERY sophisticated signal
processing systems to reduce that ratio by another 3-4 dB. Many years ago,
I sold some of this gear to broadcasters. Cost was in the $5K range in the
early80s. One of the common techniques is to split the spectruminto
multiple frequency bands and process each band separately, then combine
them. This only works with full bandwidth audio (that is, 20-20,000 Hz);
the bandwidth we transmit would fit into one of those individually
processed bands.
73, Jim Brown K9YC
Guy Olinger K2AV
TX for communications audio IS IN THE TX components and cannot be
changed or "optioned" out. The K3, to satisfy all the vocal desirers
of options, lets the user set everything, so that every body can have
it his own way, from "I want my beautiful deep bass voice to be heard
on ESSB", to contesters' "I just want my highs out there on the power
peaks, I ONLY care about maximum QSO's, fidelity be damned".
In the old analog rigs, all that stuff was wired in, one resistor and
capacitor at a time, and the choice implied by the discrete components
used was THE choice, PERIOD. I note that some of the rigs quoted
earlier are in that collection. What was done in those was to favor a
highs-emphasized TX audio, with as little distortion as possible.
SOME clipping helped with average power. ESSB advocates should note
that NONE of those emphasized the bass, unless someone went in and
monkeyed with the discrete components, or put a banded preamp between
the mic and the rig.
What that state of affairs did was ENFORCE a defacto communications
audio default, very soft on bass and hard on highs.
Now you have K3 users with no clue about how to set the TX eq and
clipping level OPTIONS to get the IDENTICAL shape to their voices on a
K3 as the other rigs. There actually is a clarity advantage to the
K3's clipping method IF you know where to set all the options.
The problem is that in the bright new digital world, with options to
satisfy every conceivable preference, ONE HAS TO KNOW HOW TO SET THE
OPTIONS AND LEVELS TO GET WHAT HE WANTS. Add that to NOBODY EVER
WANTS TO READ THE MANUAL. (I, personally am no better than anyone else
here, I HATE reading manuals.) With the combo you get complaints that
"My K3 is broken" because the user doesn't know that's an option and
he has to set it his way in that menu. Likely perceived awful and
confusing, because to understand the menu you have to read the manual.
And there are so many optional behaviors that keeping up the manual
is a real piece of work, and requires the most talented of technical
writers to explain it in a straight-forward effective manner.
This is not a peculiarity with a K3. N1MM has that problem because of
the huge number of options, as does all the MicroHam stuff, which
serve a very wide audience IF the users understand the options. I
pretty much suffered brain damage learning MM logger. Microham was
better because I had W4TV. K3 was easier yet because of the reflector.
Flexibility generates confused digital options newbies all over the
place in all kinds of pursuits. The universal curse of the age of
digital options freedom.
These days RTFM is really the only way out. Unless someone who has
the time, inclination, and the sharp knowledge of all the options,
sets up a utility which sets a spectrum of options based upon older
rigs.
Since the version D DSP board, this actually seems possible. Once you
learn all the diddles, the K3 is marvelous. I have its RX sounding
like my 75A3, IF I use a good speaker. ALL my computer speakers turn
out to be crap beside my old-time Acoustic Research bookshelf speaker.
If I run my 75A3 to the computer speakers, the 75A3 sounds like crap.
Of course the sonorous old AR 8 ohm needs a 10 watt audio stage or an
amp to drive it because it is so brutally inefficient.
Toby Deinhardt
> get their diaphram into the act and speak with confidence, not just
Full ack.
Go to a news studio or most other TV studios, and listen to with which
volume and dynamic style is spoken. I think a lot of Hams would be very
surprised.
vy 73 de toby
Joe Subich, W4TV
> Now you have K3 users with no clue about how to set the TX eq and
> clipping level OPTIONS to get the IDENTICAL shape to their voices on a
> K3 as the other rigs. There actually is a clarity advantage to the
> K3's clipping method IF you know where to set all the options.
How very true ...
As K9YC (and others) has often advised, start by setting the first two
(LF) bands of the TX Eq to -16 and set the third (200 Hz) to - 9dB or
lower. Set the top three bands for at least 3 dB/octave (+3, +5, +6
dB) boost ... that works well with a mic that has some natural high
frequency boost like the HC-4. For flatter mics (like the new HC-6
or a CM-500) use more high frequency boost (between +6. +10, +12 dB
and +9, +16, +16 dB) to provide clarity. Since the human voice has
little energy in the 600 - 1200 Hz band, I like to add a bit of a
"notch" in the middle (-6dB at 800 Hz) ... cutting that band helps
to reduce background noise without impacting voice quality.
With reasonable adjustments to enhance the frequencies important for
communications (vs. some "golden ear" belief in a bandwidth more
appropriate to classical music), reducing the power wasted in the
lower octaves that do not contribute to enunciation, and 10 to 15 dB
of clipping, the K3 can more than hold it own while remaining very
clean compared to the Yaecomwood rigs that drive the PA into clipping
in order to generate ALC!
73,
... Joe, W4TV
Richard Siegmund Lindzen
Dick, WO1I, K3# 911
Joe Subich, W4TV
Jim,
> This only works with full bandwidth audio (that is, 20-20,000 Hz);
> the bandwidth we transmit would fit into one of those individually
> processed bands.
Wes Stewart, N7WS had a very successful split band approach for amateur
use years ago ... it was butchered by Alpha (the Vomax) but three bands
(300 - 600 Hz, 1200 - 2000 Hz and 2000 - 3500 Hz) is a very good choice
for "communications/talk." By keeping each band less than one octave
in width, IMD and harmonic products from clipping can be minimized even
with analog techniques.
With modern DSP technology one can add phase scrambling (rotation),
per band noise gating, pre- and post-clipping equalization and use
half octave bands with minimal hardware cost to achieve very clean
compression/clipping with very tight dynamic range even in noisy
environments.
73,
... Joe, W4TV
Lu Romero
I have no issues with lack of "punch" with the K3, and I am
considered an audio "dweeb" by some of my contesting
contemporaries.
On my former TS-850, I had "lotsa punch". My audio chain
consisted of the following:
Heil DX-4 headset
Behringer EuroRack 802 mixer with three band "classic" EQ
Radio Design Labs ST-GCA2 limiter/leveler/AGC
Behringer MDX-2600 Compressor/Gate
Home made precision T-Pad line to mic level interface
microHAM microKEYER
It made nice, crisp "razor sharp" audio that cut through
piles, if a little "dirty" at the edges. "Thin" sounding
but penetrating.
Now, I have the following audio chain with my K3:
Yamaha CM-500 headset
microHAM microKEYER 2
And I feel I have the same punch, but with a little
difference... I how have some minor bass "presence" at
around 300-400hz that I couldnt get with the DX4, thanks to
the very nice EQ built in to the K3 and the very broad
CM-500 capsule. And I have none of the "grunge" around the
edges that I had with the 850. The K3 is much cleaner.
I have the same peak to average ratios, as witnessed by my
scope waveforms, the same "penetration" and the same
"punch", but with a mellower, cleaner, less "thin" sound.
It has taken me six months of tweaking and the Yamaha
headset to get to to where I am 98% happy with K3 transmit
audio.
I have a very noisy shack environment, so there are some
caveats:
First, the DX4 is a "close talking" microphone. You have to
almost eat it for it to work right. The CM-500'ds Electret
capsule has a pronounced proximity effect. You HAVE to keep
it AT LEAST an inch from your mouth or it sounds mushy and
bassy. This, combined with its omnidirectional response and
the need to run mic gain relatively high to command good
peak to average ratios from the K3 processor, cause "much"
room noise (not a lot really, but Im picky) to find its way
to the transmitter.
Second, I could fix this with the AF GATE function, but the
gate has and issue (to me) that it does not have enough
hysteresis, that is, it drops RIGHT NOW, and when it drops,
it drops IMMEDIATELY off the cliff with a straight as a
ruler response. WHAM! Audio's GONE!
I have worked with the IHY product at the Multi-Multi I work
with (we have standardized on it for all positions now using
TenTec Orions) and I find both my old Behringer and Julius'
Gate to have a logarithmic ramping (the gate ation starts
slow and speeds up over time) plus there is some hysterisis
in the gate's response to instantaneous noise spikes, so
these two are not as "nervous sounding" as the K3 gate is.
As it is, the K3 gate responds so very quickly that it adds
a "crackle" to the audio because it partially triggers with
small noise spikes (lacks hysteresis and immediately turns
on and off). The effect is rather annoying and spoils the
gate's effectiveness. But the processor works so well now
that I will put up with the gate till Lyle can get around to
it.
Other than those niggling comments, Im a happy camper with a
much simpler audio chain. Simple is better!
The key to remember is that we are transmitting into a very
noisy medium (the HF bands during a contest). You need to
do two major things: Sacrifice wide "dynamic range" in your
transmit waveform for effective communications and
concentrate your available transmit power into the frequency
bands that will do you the most good for inteligibility.
Bass below 300hz is your enemy! If you do this while
maintaining the highest possible peak to average ratio with
the lowest distortion figures your transmitter can muster,
you will be heard and, most importantly, you will be
UNDERSTOOD. And it helps to have proper enounciation skills
(ask any ATC or pilot!)
This is not as simple to do as it sounds. It takes
discipline with mic technique and voice projection BEFORE
you apply any digital processor "magic". Garbage in -
Garbage out!
Then there's VooDoo... I always put a 3x5 card on the
monitor that says "I AM LOUD". If you think you are, you
are!
Lu Romero
W4LT - #3192
Message: 10
Date: Mon, 4 Oct 2010 20:52:41 -0500
From: "Jim Miller KG0KP" <JimM...@STL-OnLine.Net>
Subject: Re: [Elecraft] SSB transmit audio - Where's the
punch?
To: <elec...@mailman.qth.net>
Message-ID: <D6C7D2F59E51421A9944095DE3F9D62D@HMJM1800>
Content-Type: text/plain; format=flowed;
charset="iso-8859-1";
reply-type=original
Don't get it too high. Higher is NOT necessarily better. I
hear station in
contests that are very strong and they are almost impossible
to understand
because they have their compression WAY to high. Turning it
down would make
them much better.
I just made 1752 contacts in a special event and got many
unsolicited
comments on my audio being very good, excellent, etc. I was
running the HC4
and the mic gain at 25 and the compression at 25. My voice
is relatively
low.
73, de Jim KG0KP
----- Original Message -----
From: <w5...@w5ov.com>
To: "Dick, WN3R" <wn...@verizon.net>
Cc: <elec...@mailman.qth.net>
Sent: Monday, October 04, 2010 1:11 PM
Subject: Re: [Elecraft] SSB transmit audio - Where's the
punch?
______________________________________________________________
Jim Brown
> First, the DX4 is a "close talking" microphone. You have to
> almost eat it for it to work right. The CM-500'ds Electret
> capsule has a pronounced proximity effect. You HAVE to keep
> it AT LEAST an inch from your mouth or it sounds mushy and
> bassy. This, combined with its omnidirectional response and
> the need to run mic gain relatively high to command good
> peak to average ratios from the K3 processor, cause "much"
> room noise (not a lot really, but Im picky) to find its way
> to the transmitter.
The CM500 is NOT an omni-directional mic -- you've noted that it has
proximity effect, and that's BECAUSE it is a cardioid (directional) mic.
As to the proximity effect -- you can compensate that the same way that
virtually ALL vocal mics are compensated within the mic -- with
additional low end rolloff.
73, Jim K9YC
Ignacy
I used to have TS-850 and FT-1000MKF. The processor in TS-850 was very good
and in MKF not as strong but audio was nice. If you read some reports, ALC
in 1000MP and probably other radios in the series is too fast and produces
lots of IMD at high processing. ALC in K3 is very slow and thus avoids
distortion but also the extra (useless) power that goes with distortion.
K2 IMHO beats both radios. I am amazed that even cranking COMP to 30 does
not produce distortion but brings extra contacts.
As mentioned before, set equalization first, MIC gain second, and crank up
COMP. I use COMP of 15 in casual conditions, 20 as a norm, 25 for normal
DXing or contesting, and 30 when things go tough. Never a bad report.
Ignacy
--
View this message in context: http://elecraft.365791.n2.nabble.com/SSB-transmit-audio-Where-s-the-punch-tp5600109p5605356.html
Sent from the [K3] mailing list archive at Nabble.com.
Wes Stewart
Actually, credit for the original concept belongs to Walt Schreuer, K1YZW. He developed the "SPB-3" sold by Maximilian Associates in about 1976. No schematics were provided, of course, so I and my late friend Jim Metzer, W7TKR, produced our own way to do it.
See: http://k6mhe.com/n7ws/S_PROC.pdf
Our simplified method for accommodating the phase shift through filters does lead to more passband ripple than more elegant phase correction, but operationally it's never been an issue.
IMHO, the split-band approach is better than rf clipping. Split-band clipping was the choice of the engineers of one of the arguably best sounding radios around, the Kenwood TS-870.
I personally know of several hundred of these that were built around the world, not counting however many kits were sold after the magazine editor founded a kit company on my design, the RADIOKIT DX-1.
Wes N7WS
--- On Tue, 10/5/10, Joe Subich, W4TV <li...@subich.com> wrote:
Jim,
> This only works with full bandwidth audio (that is, 20-20,000 Hz);
> the bandwidth we transmit would fit into one of those individually
> processed bands.
Wes Stewart, N7WS had a very successful split band approach for amateur
use years ago ... it was butchered by Alpha (the Vomax) but three bands
(300 - 600 Hz, 1200 - 2000 Hz and 2000 - 3500 Hz) is a very good choice
for "communications/talk." By keeping each band less than one octave
in width, IMD and harmonic products from clipping can be minimized even
with analog techniques.
With modern DSP technology one can add phase scrambling (rotation),
per band noise gating, pre- and post-clipping equalization and use
half octave bands with minimal hardware cost to achieve very clean
compression/clipping with very tight dynamic range even in noisy
environments.
73,
... Joe, W4TV
Luis V. Romero
Its very mildly cardioid then! :) This aint no shotgun, Jim!
I choose not to compensate for the boominess with EQ and not eat this mic
like an HC4 as it adds to the distortion I hear in the monitor and see on
the scope. I would rather run the mic gain a little hotter and put up with
the room tone than deal with the boominess electronically. I can always
gate the room tone. I have always tried not to "fix stuff in the mix" but
I'm severely crippled with a very poor operating environment for sound.
Then again, this is Ham Radio, not work!
And with 12 volt finals, I need all the signal cleanliness I can get :)
Of course, you are right, but I'm walking a fine line between distortion and
punch. I find the fatter foam windscreen from an Audio Technica Pro8HEx mic
helps me place the CM-500 capsule at just the right distance from my mouth
for me.
-lu-
Message: 27
Date: Tue, 05 Oct 2010 14:25:59 -0700
From: Jim Brown <j...@audiosystemsgroup.com>
Subject: Re: [Elecraft] SSB transmit audio - Where's the punch?
To: elec...@mailman.qth.net
Message-ID: <4CAB97E7...@audiosystemsgroup.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
On 10/5/2010 2:09 PM, Lu Romero wrote:
> First, the DX4 is a "close talking" microphone. You have to
> almost eat it for it to work right. The CM-500'ds Electret
> capsule has a pronounced proximity effect. You HAVE to keep
> it AT LEAST an inch from your mouth or it sounds mushy and
> bassy. This, combined with its omnidirectional response and
> the need to run mic gain relatively high to command good
> peak to average ratios from the K3 processor, cause "much"
> room noise (not a lot really, but Im picky) to find its way
> to the transmitter.
The CM500 is NOT an omni-directional mic -- you've noted that it has
proximity effect, and that's BECAUSE it is a cardioid (directional) mic.
As to the proximity effect -- you can compensate that the same way that
virtually ALL vocal mics are compensated within the mic -- with
additional low end rolloff.
73, Jim K9YC
No virus found in this outgoing message
Checked by PC Tools AntiVirus (6.1.0.25 - 6.14880).
http://www.pctools.com/free-antivirus/
Tim Tucker
earlier in this thread because I spent some more time testing. Several
private emails confirmed my suspicion that others were able to achieve
better results than I was seeing.
Tonight after playing around with the rig some more, I was actually able to
get the AVG power output real close to 60 watts on on specific syllables or
words by playing with the audio response, audio gain input levels, etc.
Typical voice response produced about 40-45 watts AVG. This is considerably
better than I reported earlier, all things considered. What's different
about the K3 is that it's AVG power output seems very sensitive to a lot of
variables, especially your EQ settings. The K3 CMP setting quickly reached
a point of diminishing returns for me, unlike what others are reporting, but
you probably have to with what works for your voice/mic combination. I just
can't run that much compression before the audio sounds like trash to me.
This definitely isn't a plug and play rig. It is probably easier to achieve
certain results with other rigs, but what is nice about the K3 is the
ability to tweak almost everything to your heart's content.
Jim Brown
> Typical voice response produced about 40-45 watts AVG. This is considerably
> better than I reported earlier, all things considered.
I would be very cautious about saying it's BETTER until you get some
critical audio reports on the air. You could easily have great meter
readings and lousy sounding audio.
73, Jim K9YC
Don Wilhelm
As far as the response of the K3 is concerned, I refer you to Guy
Olinger's (K2AV) recent posts. The K3 is a highly configurable
transceiver, and "Read The Fine Manual" is an important step. If after
reading, you have problems understanding the implications, definitely
ask here on the reflector, and you will get an informed response.
When the questions are more like "tell me how to do ----", I often feel
like just replying RTFM, because it is obvious that the poster has not
even cracked the manual and just tried to do special things without
trying to understand how things work.
The K3 with its default settings is good for the average types of
operation, but for those who want to "do something special", it is
important to RTFM first and then ask questions - failure to do that will
likely lead to not understanding the responses to the questions, and
further confusion when what is suggested is not what the poster expects.
In other words, vary from the default setting only with a valid reason
to make that change. Many changes from the defaults are good for
certain situations, but will not produce the desired results for other
situations.
The K3 is highly configurable, but modify the default configuration with
an informed eye, and see if the results do what you desire rather than
accepting someone else's changes without question. I repeat, RTFM is
important for any modern transceiver with configuration menus - the K3
is no exception.
73,
Don W3FPR
On 10/6/2010 1:39 AM, Tim Tucker wrote:
>
> This definitely isn't a plug and play rig. It is probably easier to achieve
> certain results with other rigs, but what is nice about the K3 is the
> ability to tweak almost everything to your heart's content.
>
David Woolley (E.L)
do. In their case, the aim is to get the dynamic range into the gap
between the threshold of hearing and threshold of pain, both of which
tend to converge towards each other in hearing loss.
One figure of merit for such aids is the number of compression channels.
Beyond this, modern hearing aids are essentially sophisticated DSP NR
devices, in ham radio terms.
Top quote through list policy only.
Jim Brown wrote:
> early80s. One of the common techniques is to split the spectruminto
> multiple frequency bands and process each band separately, then combine
> them. This only works with full bandwidth audio (that is, 20-20,000 Hz);
> the bandwidth we transmit would fit into one of those individually
> processed bands.
--
David Woolley
"we do not overly restrict the subject matter on the list, and we
encourage postings on a wide range of amateur radio related topics"
List Guidelines <http://www.elecraft.com/elecraft_list_guidelines.htm>
Terry Schieler
sitting down. More diaphragm punch when standing (especially after lunch!)
Terry, W0FM
-----Original Message-----
From: Toby Deinhardt [mailto:dj7...@muenchen-mail.de]
Sent: Tuesday, October 05, 2010 1:31 PM
To: elec...@mailman.qth.net
Subject: Re: [Elecraft] SSB transmit audio - Where's the punch?
Guy Olinger K2AV
>
....snip... For flatter mics (like the new HC-6
> or a CM-500) use more high frequency boost (between +6. +10, +12 dB
> and +9, +16, +16 dB) to provide clarity. Since the human voice has
> little energy in the 600 - 1200 Hz band,
Except for the lady folks who seem to START in that range. What
happens in so many female voices I hear (including broadcast stations)
is that high central tonality dominates to the point of overwhelming
the silibant range and 600-1200 NEEDS to be seriously suppressed to
let the silibants emerge. I'm not sure 6 dB is enough for them.
73, Guy.
> I like to add a bit of a
> "notch" in the middle (-6dB at 800 Hz) ... cutting that band helps
> to reduce background noise without impacting voice quality.
>
> With reasonable adjustments to enhance the frequencies important for
> communications (vs. some "golden ear" belief in a bandwidth more
> appropriate to classical music), reducing the power wasted in the
> lower octaves that do not contribute to enunciation, and 10 to 15 dB
> of clipping, the K3 can more than hold it own while remaining very
> clean compared to the Yaecomwood rigs that drive the PA into clipping
> in order to generate ALC!
>
> 73,
>
> ... Joe, W4TV7
Guy Olinger K2AV
> Jim:
> Of course, you are right, but I'm walking a fine line between distortion and
> punch. I find the fatter foam windscreen from an Audio Technica Pro8HEx mic
> helps me place the CM-500 capsule at just the right distance from my mouth
> for me.
And drool or dripping spittle in the keyboard is not an issue? :>)
73, Guy.
>
> -lu-
Eric Swartz - WA6HHQ, Elecraft
returns ;-)
73, Eric
Elecraft List modulator
---
On 10/6/2010 8:45 AM, Guy Olinger K2AV wrote:
> On Tue, Oct 5, 2010 at 10:36 PM, Luis V. Romero<lro...@ij.net> wrote:
>> Of course, you are right, but I'm walking a fine line between distortion and
>> punch. I find the fatter foam windscreen from an Audio Technica Pro8HEx mic
>> helps me place the CM-500 capsule at just the right distance from my mouth
>> for me.
> And drool or dripping spittle in the keyboard is not an issue? :>)
> 73, Guy.
>
Lu Romero
That's another reason for the bigger foam windscreen, Guy,
its more absorbent! :)
-lu-
----- Original Message Follows -----
From: Guy Olinger K2AV <oli...@bellsouth.net>
To: lro...@ij.net
Cc: elec...@mailman.qth.net
Subject: Re: [Elecraft] SSB transmit audio - Where's the
punch?
Date: Wed, 6 Oct 2010 11:45:38 -0400
>On Tue, Oct 5, 2010 at 10:36 PM, Luis V. Romero
>> <lro...@ij.net> wrote: Jim:
>
>> Of course, you are right, but I'm walking a fine line
>> between distortion and punch. I find the fatter foam
>> windscreen from an Audio Technica Pro8HEx mic helps me
>place the CM-500 capsule at just the right distance from my
>> mouth for me.
>
>And drool or dripping spittle in the keyboard is not an
>issue? :>)
>
>73, Guy.
>
>>
>> -lu-
Joe Subich, W4TV
> Except for the lady folks who seem to START in that range. What
> happens in so many female voices I hear (including broadcast stations)
> is that high central tonality dominates to the point of overwhelming
> the silibant range and 600-1200 NEEDS to be seriously suppressed to
> let the silibants emerge. I'm not sure 6 dB is enough for them.
I've found that my general EQ practice of cutting the 800 Hz band
(1000 Hz and one to two bands either side in a professional 1/3
octave EQ) also helps the "thin voiced" females by relatively
enhancing the sibilant range.
Note that the fundamental of most male voices is between 100-200 Hz
(it is rare to find a voice below 90 Hz with the lowest recorded
measurement around 80Hz) with most female voice in the 150-300 Hz
range. What most lay persons consider the "fundamental" of a voice
is really the dominant overtones (harmonics) ... specifically those
harmonics that are most pronounced due to the physical structure of
the throat, mouth and nasal passages as well as the specific way in
which a given language is produced (voiced).
73,
... Joe, W4TV