On Oct 22, 2023, at 10:35, Gordon Gibby via groups.io <docvacuumtubes@...> wrote:

?Thanks Evan. Yes we need confirmation. One way or the other. Either I’m doing something wrong, or my unit is a complete fluke, or there’s a problem!

I’m not sure why anyone would bother developing software for a radio that might not be legal to operate…..

What if ?only 25% of the units have a problem? But that still way way too high!! ? This is a groundbreaking transceiver and it should not be troubled by these kinds of problems. ?We certainly fixed them on the developer unit. ?And my measurements there may have been clouded by the Heathkit Cantana so it may be even better than I know! ?

The bypassing issue I think is important for the 470 ohm resistors. ?Only c204 protects at that end —-Adding several more may help. Also between the 470 and the indconductors may help

It’s dicey to slow down the switchover with capacitors on the 470 K. ?Theoretically, into a 50 ohm system, the 470 K alone provides better than 60 DB isolation. ?So I’m not sure a lot needs to be done there. Maybe a few picofarads? ?The output of the RF voltage doubler has a few hundred picofarads already. ?so it provides some significant isolation. ?

I think the real issues are the +12 line, where RF energy can get back into earlier stages and wreck havoc as far as spurious responses there, !!!!! and maybe just maybe on the ground line itself which I hope is solid. ?

The QDX seem to show that it’s possible to fix all of this. But we’re running a lot more RF current through those diodes than it did. ?I don’t know if we need additional diode’s in parallel at the output, I just don’t know. ?? A text jig fixture and a good clean commercial transceiver might be able to prove this— run the RF through a turned-on 1N4007 and measure the spurious response afterwards; this would also be a way to select preferred diodes ?for this service

I’ll be back to Gainesville in about 10 days and then I may be able to start measurements again. ?

I don’t know how many other people have quality test set up such as the Siglent or Rigol spectrum analyzer‘s but we could sure use some data; as my finances allow I’m going to buy a higher power RF pad so I can easily tolerate the full output for some time. ??

Thanks for your thoughts!

Gordon?

On Oct 22, 2023, at 07:15, Evan Hand <elhandjr@...> wrote:

?Gordon,

If it is passed through on the bias circuit, would adding some capacitors help?? Every connection of the bias to the LPF circuit may need to be bypassed.



This might mess up the filters instead of helping.

So far, you are the only one that is reporting an issue.? It would help if others verified that there is an issue with other V2 sbitx or for HFSignals to respond with measurements from the production units that have been tested.

73
Evan
AC9TU