[Elecraft] K2 SSB Mic Input Frequency response?

[Jim Brown]

On Sun, 28 Aug 2005 13:18:35 -0400, Francis Belliveau wrote:

>I am now confused by the existence of two different mod's with the same
>goal.

>It makes perfect sense to me that I should not be wasting Tx energy on audio
>frequencies as low as 20Hz.  Beyond that I am somewhat in the dark.

>Jim Brown and Stewart Baker have each suggested a modification to improve
>this situation.

>I am at a loss to know which, if either, I should implement.  I must admit
>that I an very tempted to incorporate the indicated improvements.  However,
>I would have no means of evaluating the results and therefore look to the
>remainder of those on this list for further opinion.

>Summary of the mod's are:

>from Jim:
>    Change R14 from 1K to 2K.
>    Change C34 from 2.2hF to 0.22 uF.
>    Add 1uF in parallel with R15

>from Stewart:
>    Change C34 from 2.2uF to 0.47uF
>    Change C31 from 2.2uF to 1uF
>    Change C20 from 0.33uF to 0.047uF

Differences:  I'm doing more low cut with C34 than Stewart is, but Stewart is 
doing more low cut with C31, which I haven't changed. Both mods are good, and are 
additive. At some point we go too far and the audio gets too thin. 

Since I don't know anything about the details of U3 (the comp/limiter chip), I 
haven't messed with C31. If it is reducing the LF compression, it is probably a 
good thing. Perhaps Stewart can comment. 

>Is the situation severe enough to warrant an official mod for this?

Severe is the wrong word. What Stewart and I and KI6WX have done is tweak a 
design that is lacking in gain and has too much LF response. Yes, these mods 
SHOULD be integrated into the K2 standard production. 

On Sun, 28 Aug 2005 11:18:20 -0700, Ron D'Eau Claire wrote:

>I'm wondering if the difference is just a
>difference in impedances, or whether something else is going on. 

Both. The impedance of the mic input circuit is lower than it needs to be, and 
the GAIN of the entire audio chain is a bit low. One of KI6WX's mods addresses 
the gain issue AFTER the comp/limiter. The ones that Stewart and I are talking 
about are BEFORE the comp/limiter. Both are good sets of mods. 

In general, the Z of the input stage should NOT load the mic -- that is, Z in 
should be 10X the Z of the mic. Pro mics are typically 150-250 ohms, and have 
relatively low voltage output. The K2 input Z is roughly 1K, which is not bad for 
pro mics, and the loading is minimal. BUT -- since the gain is low, you need a 
transformer to step up the mic to get more voltage to the input stage. When you 
do that, the turns ratio increases the voltage, BUT it also makes the mic look 
like a much higher Z -- the mic Z is multiplied by the SQUARE of the turns ratio. 
So a 1:3 step up gets you 3x the voltage (9.4 dB) but the 150 ohm mic now looks 
like 1.35K, which forms a voltage divider with the input stage and causes much of 
that 9.4 dB to be lost. Changing R14 to 2K gets most of that gain back. Going to 
3.3K or 4.7K would give you another couple of dB. 

The reason that good pro mics sound dull with ham transmitters is that for some 
stupid reason, an international standards body long ago established an HF rolloff 
for communications transmitters that is both too low and too severe -- -6dB at 
2.6 kHz. A mic with a smooth, flat response will sound dull and muffled with that 
kind of transmitter response. Rather than change the dumb standard, mic 
manufacturers began building all their mics with a 6 dB peak between 3 and 4 kHz 
to compensate. You can see this is their data sheets. The resulting response 
approaches what it should have been in the first place -- flat to about 3 or 3.5 
kHz, then rolled off sharply. 

So when you use a communications mic, like a Shure 444 or the wildly overpriced 
Heil mics, you're simply compensating for that standard that was wrong in the 
first place!  All of these mics have some variation of that peaked response in 
the 2-4 kHz region. Plug them into a good sound system and they sound awful. 
Should that standard be changed?  Of course. But that's like trying to change the 
course of a nation's foreign policy -- no one wants to admit they were wrong, and 
too many people/companies have a vested interest in things remaining the same. 

Yes, mis-alignment of the crystal filters and IF can screw up the overall 
response too, but that is in addition to these purely audio issues. 

Jim Brown  K9YC