[Elecraft] First pass... sidetone output tests

[Tom Hammond NØSS]

GM George:

>Tom, while my brief examination of the audio on #489 is limited to
>measuring speaker output only, perhaps the data will be of interest.

And of use as well.. thanks pal.

>Equipment used:
>
>K2 #489
>
>Goldstar 9020 20-MHz dual-trace scope
>
>Beckman 2020 True RMS DMM and Radio Shack 22-168 DMM
>
>Procedure:
>
>1. The scope and 10X probe were calibrated against the internal scope
>calibrator signal to 500 millivolts per centimeter. A resistive load
>measuring 4.03 ohms was connected to the Speaker jack and voltage
>measurements taken across that load.
>
>2. With ST L set at 050, I measured a peak-to-peak 702 Hz tone signal of
>240 millivolts. This corresponds to 84.8 millivolts rms. The BI DMM read 80
>millivolts rms while the RS DMM read 83.1 millivolts. While the waveform
>was a true "eyeball" sinewave, the BI meter reads true rms while the RS
>meter does not.
>
>3. With ST-L set at 255, the scope signal computed to be 424.2 millivolts
>rms. I did not make DMM measurements since the first test produced similar
>readings.
>
>My K2 does have the AF filter installed but it was not used for these
>measurements. The sidetone waveform mod has been installed. In general
>usage, I do not consider the K2 to be deficient on audio except on very
>weak or marginal signals. On any signal that is S5 or above, the audio
>level is adequate.
>
>Tom, I note that your sidetone level was set for 052 which might account
>for the somewhat higher readings you observed.

Possibly. I set the ST L to produce a specific p-p output from the TONE 
generator and then went from there. I'd initially considered merely setting 
ST L to 50 and going with that, but then stopped to think that possibly the 
level might not be the same from all of the DACs (or from the CPU in cases 
where a KIO2 had not been installed), so I picked a fixed output level, 
rather than an arbitrary ST L setting.

In actuality, I suspect the difference is negligible, for these tests. I 
suspect that differences which can be really 'noticed' will be off by an 
order of magnitude one way or the other.

>Incidently, I have to set CAL ST P at 0.69 in order to get the 702 Hz tone.
>The sidetone pitch is about 10 Hz high if I set it at 0.70. C22 has been
>adjusted to place the MCU clock at 4000.000 MHz per my Icom 756 PRO which
>is calibrated to within less than two-Hz error against WWV.

That should be more than adequate accuracy.

>I calibrated C22 at the fourth harmonic of 16 MHz using One-Hz resolution
>on the PRO and measuring the beatnote both by scope and by HP 400 AC VTVM
>to find minimum AC for the zero-beat state. Final adjustment had the
>oscillator at less than one Hz beatnote against the PRO. Even after all
>that, ST P still has to be set at 0.69 to produce a 702 Hz sidetone pitch.
>Any ideas?

No clue on that one... I've never gone to that length to check the 
frequency of my sidetone. I just knew that at 600Hz, it matched (dead-on) 
the 600Hz sidetone from my CMOS Super Keyer III, so I chose not to go any 
further.

>I have recently completed a 95 dB attenuator for the PRO to provide a
>sufficiently low output signal to run tests with the K2. The PRO has a
>minimum r-f output level of 1.55 watts which is reduced by 95 dB to produce
>about an S5 signal on the K2. So, until I can find something like an HP 606
>signal generator, I will have to make do with the PRO.

I'm looking for a good sig gen as well. Yesterday, I borrowed an IFR 
Service Monitor from a local ham who is a BC station engineer... problem is 
that I have to give it BACK to him... <G>

73,

Tom