[Elecraft] K3 sidetone level

Jim Brown jim at audiosystemsgroup.com
Sat Nov 9 02:39:56 EST 2024


On 11/8/2024 8:36 PM, Chris wrote:
> For some unknown reason, the sidetone level in my K3 doesn't sound right to
> me. As some of you already know, I have two K3. The CW sidetone on one at
> 60 generates the same volume as the other at 25.

I've been running K3s since 2007, and I don't know the answer. It MIGHT 
be related to the speakers or phones you're listening on, or it could be 
that you're hearing a difference between the AF and RF gain settings and 
whether the preamp is on or off, and on rigs with the newer preamp 
that's in the XVTR module. If on SSB, could be difference in mics. Mic 
gain settings could be different.

BUT -- the numeric values you see change with most controls are 
meaningless. The numeric value of compression, of mic gain, line gain, 
of VOX, Anti-VOX, the many ways in which AGC can be varied, are all 
examples of numeric values that are meaningless.

Numeric values that DO directly reflect what we're adjusting are TXEQ, 
RXEQ (dB boost or cut), POWER (in watts) that are telling the radio to 
transmit (to the extent that it's properly calibrated), anything with 
frequency, voltage, current, the relative dB voltmeter tied to received 
signal level. The S-meter IS calibrated to voltage at the antenna terminal.

I've many times described how to set up TXEQ and COMP for best sounding 
audio for communications and "talk power." I'll repeat it. First, set up 
TXEQ: turn the first three bands (50, 100, and 200 Hz) all the way down. 
Turn the fourth band (400 Hz) down by 6 dB. set the top two bands for 
3-6 dB boost. Turn COMP all the way down. Next, set mic gain per the 
manual. Now, start turning COMP up until the on-screen meter shows about 
10 dB of gain reduction on voice peaks as you transmit.

We've accomplished two important things here. First, we've killed all of 
the voice frequencies below about 400 Hz, which do NOTHING useful for 
speech intelligibility, but burn about 3dB of transmitter power. Second, 
the COMP (compression) function turns down the gain on voice peaks, 
making the quieter parts of our speech 10dB louder. 3dB is double the 
transmitter power, 10 dB is 10X the transmitter power. So we have made 
turned our 100W radio into the equivalent of a 2kW radio! or our 5W 
radio into a 100W radio.

There's one other step in this. Get another experienced ham with good 
radio ears to listen to you with the filters in his radio set fairly 
wide (like the normal 2.7 kHz bandwidth that is pretty much the standard 
of ham SSB transmission). The settings I've given above are the starting 
point for most good mics (and I don't mean expensive) for ham rigs, but 
some sound a bit different from others. And not all voices are the same. 
Some voices have lots more lows, and with these, we may need to turn 400 
Hz down more than 6dB. Remember -- we're going for talk power, whether 
we're DXing, contesting, or ragchewing with QSB.

What I'm saying about speech intelligibility was learned in the earliest 
days of telephony by engineers at AT&T Bell Labs, which until about 
1980, when the company was broken up because it was a monopoly, was the 
most important engineering organization on the planet for the prior 100 
years! The transistor was one of their later creations! They discovered, 
published, and patented most of what we know about electronics, audio, 
radio, transmission lines, and how we understand speech. They invented 
and patented the best of the two systems for stereo in 1936.

73, Jim K9YC



More information about the Elecraft mailing list