Well.. I thought I found the problem. One was a stupid mistake of not
putting in a thick jumper in the drain supply of the second transistor in
the pair. I noticed the output of the second transistor was a lot lower (as
it's drain was being powered through a higher resistance path). I thought
this was because the transistor was shoddy (a used unit with one gate
threshold of 4.0V and the other 4.1V) so I replaced with a new one that has
exact 3.5v thresholds on both gates. Imagine my surprise when I found still
one drain produces a higher output. It is then I fitted the jumper and the
power jumped to almost double the before (40W out of 1W input) I didnt dare
increase the drive just yet because there was still ~20% discrepancy. And I
actually traced it to the input signal not being fully symmetric. One side
is definitely higher. I messed with the trimmer but all I could achieve was
getting both sides lower of higher...
Anyone has any ideas how to make the output of a quarter wave symmetrisator
more symmetric? Maybe I need to tune the length more precisely. I wonder
will it become asymmetric away from the quarter wave frequency ? One side
being dominant above and the other side below? It sems logical, but I've no
idea if that is true.
73, Lukasz
On Sat, 29 Mar 2025, 20:27 Lukasz, <sp4it.mail@gmail.com> wrote:
> Thanks Steve. I'm not sure how a watt meter is more useful than a vna in
> this instance. If the vna is showing me a good match the watt meter will
> agree. If the watt meter shows no power going in, it will not tell me
> anything above it. With a vna at least I know the direction to go to, how
> far I am and do on.
>
> Having said that I do have a Bird watt meter, but it has so239 plugs so I
> didn't really consider it aporopriate when I'm trying to get best swr match
> possible (not so much the plugs themselves but the short leads with even
> more pl259/so239 plugs. I'd have to use to connect to sma). It'd be: a sma
> to so239 adapter -> pl259 to pl259 cable -> so239 input of bird wattmeter -
> so239 output -> pl259 to pl259 cable and finally so239 to sma adapter. I
> though this would be completely pointless at 70cm. That entire stack would
> probably have swr of 3 by itself.
>
> As for other meters I have pretty good hp power meters with probes from
> -20dBm down to -75dBm (nothing higher though). Whenever I need to measure
> high power I just put few high power attenuators together and use my hp
> meter with -20dBm to 15dBm probe. I also have a directional splitter and an
> rf probe so I guess could hook up both meters and see both power going in
> and out.... But I'm not seeing the point at the moment .
>
> I spent entire day messing with components. Trying different coax lengst
> including too long ones and too short. Moving this or that cap half an inch
> left or right. Adding 1.2pF here or there. I can get great swr at 460mhz. I
> can get very good swr at 390Mhz. But for some bloody reason it is hard as
> hell to make it work I between on 435mhz. And when I finally did manage to
> get 1.9swr it was so narrow it was 4 by the time I went to 430mhz.
>
> So, yes, a bit more difficult than expected. I though "hey, as I'm waiting
> for replacement transistors let's build this kit that's supposed to work
> immediately (and go back to my 1w amps tomorrow). 3 days later I'm still
> troubleshooting the kit amp.... I guess I underestimated the challenge. No
> surprise there are a total of three people on the Internet that can
> actually show one working.
>
> 73, Lukasz
>
>
>>
>>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|