[hpsdr] Penny PTT unkeying spike - T/R sequencing?

[Tom Cathey]

With a lot of email help from Joe/K5SD and Frank/WA1GFZ,  I have the new
HPSDR rig running!  Thanks much guys!

In the past week I've done a lot of bench testing and find Penny to be a
very clean exciter. I especially like the IMD and opposite sideband
suppression figures. With Mercury, what a combination.

Now, I've run across something that may need more refinement. I am out of
solutions. I am seeing (and hearing) the effects of a spike during PTT
UNKEY. One symptom is a small spike 2khz above the main carrier frequency.
To see and hear it, PTT key Penny on AM or FM while watching the Mercury
panadapter or other scopes - then UNKEY and you will see this pulse ramp up
as the main carrier drop down. It will last maybe 50 ms and be down about
-30db from main carrier. (At least on my rig)  On the Merc "scope" feature
you may see the tone oscillation as it cuts off, and faintly hear it in the
headphones.  My S-meter bangs up to about 40-60+ over and I hear a rushing
pop in the headphones for maybe 500 ms.  (T/R sequencing amiss?)

On a separate receiver I can hear the after-spike as a clean carrier pulse,
weak, but still there.  I could be wrong, but I think my particular spike
level is somewhat higher than others because I compared it to Frank's Penny
over the telephone and found his pop, S-meter and spike level down lower,
but still there.

** This 2kz removed spike may NOT be related to the sequencing timing
S-meter and headphone pop however.  I mentioned it for reference.

I changed the firmware version and changed the PSDR to different versions
and saw no difference. I checked supplies for stability and ran many other
tests but had no luck in smoothing out this T/R sequencing.  If I disconnect
all cables except for the USB and power supply - and key Penny using the MOX
mouse, the spike is still there. Same with external keying.

The problem is a number of times it seemed this T/R sequencing noise caused
Penny to loose its programming and lock into a full dead carrier of 500 mW.
I had to reboot and repower the hardware to recover.

The reason I'm bringing it up is that I've had a lot of experience with T/R
sequencing homebrew transmitters and receivers and see the symptoms here. Is
it my particular lash up or do the rigs vary in this area? Maybe there is
something that can be done in firmware or software to lengthen or better
sequence  the T/R transition to smooth things out so that Penny  drops out
more quietly and the pulse is lower in the headphones, Smeter bang lower and
on the panadapter.  I checked with two others - one said he had the s-meter
bang and the other did not.

Other than that, I dearly love the overall performance of the transceiver
and look to add on an MRI (-50db IMD) linear amplifier and go from there.

Thanks again for all who have helped. 

BTW, Phil is working on that .bat file to permit audio line input to Penny.
Will post it when tested.

73,

Tom Cathey, K1JJ




 1296614697.0