I believe I fed too much voltage
to the final - it did thermal
runaway and shorted. At least
that's what the variable power
supply seems to have indicated,
by the readings of voltage and
current.
The oscillator is fine. The buffer
is amplifying. The output network
looks OK. The modulator - no problem
there. It's got tons of voltage coming
out of the modulation transformer.
So much, in fact - it was powering
the transmitter somehow, through some
leakage, or something. (It's on a different
supply.) I shut the transmitter supply off,
and there was still a carrier. I looked
at the variable supply voltage meter for the
transmitter, and the voltage meter was swinging
with the audio! (And the supply was off!)
Somehow, it all was sort of working
backwards! I know why, but it was
quite amusing.
So I think the final WAS running correctly.
As much as I try, I cannot recreate the
original correct operating conditions.
But, that's OK. There's always tomorrow.
Bruce, W 60 HZ
MICRO1700 predicted it, and it's here.
This is tomorrow.
We have inspected it, measured it, checked it for authenticity, and it's the real thing. Another day has arrived.
The panel is now meeting to decide the exact moment when tomorrow becomes today. There is some disagreement that a box of donuts could resolve.
Keep inspecting that final stage, Bruce. Within every mystery lies a hidden truth.
"The panel is now meeting to decide the exact moment when tomorrow becomes today."
Tomorrow Is Yesterday. STTOS Season One.
"We make tomorrow look like yesterday" Deuce Ex
"The future is the past is the future". ST Voyager
RFB
"I believe I fed too much voltage to the final - it did thermal runaway and shorted."
Voltage regulators come in handy at that point.
" I shut the transmitter supply off, and there was still a carrier. I looked at the variable supply voltage meter for the transmitter, and the voltage meter was swinging with the audio!"
Choke! Too much leaking RF back into the supply rail through the modulator, thus will also be the case on the front end of that modulator circuit and bleed into buffers, oscillators and PLL.
This may be indication that the design is flawed and has insufficient isolation between the stages and/or incorrect coupling values between them, causing internal VSWR between those stages. Basically what happens there is the signal is just being forced from one stage to the next. Not good!
It could also be that your circuits on individual boards spread apart like they are is the reason, shifting inter-stage coupling values by the interconnecting wires.
Need some spare chalk for the drawing board? 😉
RFB
There are various capacitors involved with the B+ voltage sections that can easily be left out, which can cause some of the problems being described.
Looking here at Version 1.0.3 of Big Talker I'll list a few...
C4, C5, C6.
The missing IC1 regulator would contain C10 & C11, more B+ filtering.
The power supply section has another regulator and more filter capacitors.
Everything is posted at
http://kdxradio.com/bigtalker.html
Wasn't going to throw this out here till the right time, perhaps now is the time.
After constructing both the Big Talker TX and the LW TX all based on one design which was for the SW Big Talker, I found I had to make several modifications to the basic design to eliminate several problems, most of which was inter-stage coupling and isolation.
The base design is good, however there is too much bleed back in the opposite direction to the signal path between each stage.
Even built onto a single PC board.
What will help is putting the buffer stage between the oscillator and modulator/final. The buffer stage can do two things, actually three but two are of most concern. Either provide isolation and no gain, or isolation and gain. In this case with this design, it needs pure isolation, thus the buffer stage should have ZERO gain.
However sometimes a buffer stage needs to tone down the incoming signal a bit to prevent over-driving the next stage. This is what is happening in the base design. Like I pointed out, the signal is being forced through the stages and thus the excess has no where to go except back in the other direction and into other circuits such as your power source, and even shove it back into your audio lines.
Effective inter-stage coupling and isolation is the key to building a good transmitter. Each succeeding stage is like a load, so the previous stages MUST be properly matched to one another, just like how a final stage must be matched to a load like an antenna. And just like a mis-matched antenna load on a final, mis-matched inter-stage coupling will produce internal VSWR, and just like a mis-matched antenna, that excess returned signal goes where...right back into the transmitter output. Same effect with the coupling of each section of the transmitter.
RFB
I'll say a few things, and then go
back and read what you both had to
say again. This seems to work best
for me. By the way, I am finding
this to be REALLY fun and educational,
too.
I do have a choke in there that should
keep the RF from the supply. I made
it myself, maybe it's not quite right.
I'll have to check that.
The different perf boards are next to
each other, so there shouldn't be a
problem there.
I noticed bleed back through the stages,
too. You can put a radio tuned to
13.560 next to the different stages
and observe that. You can also cut
power to different stages and see
what that does. This is all very
interesting.
Tonight, I'm going to attempt to get
the final running again.
We'll see.
I wanted to respond to your comments
quickly, but now I'm going to go
back and look at what you guys wrote
again, just in case I missed something.
Best Wishes,
Bruce, W 60 HZ
That's good info.
Zero gain. Got it.
One thing was nice:
Even though it wasn't
running correctly the
other night, it wasn't
FMing. So the buffer
stage was helping there.
Extra chalk sounds good.
Maybe a new chalk board, too.
Bruce, W 60 HZ
Hi RFB:
I don't recall knowing that you built a Big Talker and modified it.
If you provide a schematic I will post it with the Big Talker Short Wave Transmitter Project.
As to the Deep Voice Long Wave Project, it is a total redesign except for the original Radio8Z buffer, and contains the oscillator/PLL circuit submitted by yourself.
You also submitted a buffer suggestion which I will be posting on the Deep Voice page.
What we are talking about here is having two buffers rather than one; double buffering.
There is no need to double buffer. The problem isn't that. The problem is the design itself with the inter-stage coupling and matching.
Though the BT and LW units are different, they are using the same output circuit and modulator circuit. thus identical. The only differences is in the resonance circuit..ie LC tank, filters and inter-stage coupling.
Placing buffer stages will help isolate the stages, but those buffer stages will also have to be matched on the input side and output side if we want the whole to function properly with no internal inter-stage foldback..or VSWR.
There is no need to post a schematic. The schematic did not change, only the component values between the stages, with a few added upgrades with the use of toroid coils to do the inter-stage coupling.
When I find time I will post a list of those particular components and values.
RFB
I'll be working on that, too.
Bruce, W 60 HZ
"So I had to make my own choke and coils. It remains to be seen how well they are actually working. Tests will show results for that. The three final RF filter coils look pretty good. I believe they are close in the needed values. For the 20 uH choke, to isolate the power supply, I took a Radio Shack 100 uH choke and modified it. The above are all compromises on the original design, but so far so good."
Obviously 20uH isn't good enough for the frequency of operation. I would have left the 100uH choke intact. Might grab another and leave it the way it is. Ain't nothing wrong with over-choking the RF to prevent it from entering the B+ path. As long as the choke does not choke off your voltage and current being delivered to the circuit, who cares if the choke is over-valued.
Problem now is that it is UNDER-VALUED, thus POOF! And that's exactly what happened.
No amount of final LC tank tuning is going to prevent that RF from traveling through an insufficient valued choke. That RF will travel through that under-valued choke even if the output tank is perfectly tuned to resonance. In fact, your loosing some RF level by it "leaking" through that under-valued choke, and you did see that effect on the power supply voltage meter dancing with the audio signal passing through the modulator, so there again is a problem to be looked at...meaning too much RF from the previous stage prior to the modulator and it's forcing itself through that dead final circuit and into the power supply metering.
When you try the 2222A, place a fresh 100uH choke inline with the B+ PRIOR to the modulator and finals. It should be in front of both of those circuits...ie at the B+ point that feeds the modulator which feeds the finals with varying voltage/current from the audio signal.
One single 100uH choke should do it and will keep that RF away from where it is not supposed to be.
What type of inter-connecting wire are you using for these separated PC boards with the sections of the transmitter? Are you using mini-RF coax where RF paths are, or just plain wire? Is each PC board containing each section of the transmitter well filtered and choked in case RF is being picked up by that inter-connecting wiring? Bypass caps only do so much and unless is a tuned shunt bypass, those caps may only be there for kicks and giggles and not doing anything in the way of shunting unwanted RF from those other circuit paths.
I think you will find that using the stock RS 100uH choke on that B+ rail will solve a ton of problems.
In fact, it would be a good idea to feed EACH section of the transmitter's stages with their own choke. In other words, don't use just one choke at the power supply output to feed the whole entire transmitter..wont work because although your preventing RF from going back into the power supply, there is nothing stopping that RF from bleeding over into the stages themselves past the choke on the B+ rail from the power supply. So if you have 3 separate boards, or even if the thing was on one board, each section should have it's own choke and filtering of it's power source regardless if there is a voltage regulator there or not. Voltage regulators are not RF chokes and will pass RF just as easily as if there was no choke at all on the supply rail.
RFB
Yup. I can grab a few more of those
100 uH chokes. One for each stage.
But (and it's a long story) Radio
Shack is pretty far away. I have to
walk there. So this may have to go
on hold for a bit.
I'll look at the bypass problem again.
And, as you know, because I just got that
550 kHz rock:
In a few days, it will be back to the
RC-6A. There is one thing I forgot to
tell you. The tuning coil is stripped
where you would put the screw driver in.
It is really easier to tune the transmitter
coil and caps inside where the HV circuitry is.
Of course, I have to be very careful. And I am.
I also got some thin copper wire to make a
coil that will resonate on 550 kHz with a
365 pf variable cap. I am going to build
a small RF coupling transformer, which will
be able to fit in a shielded box.
I guess that belongs on the other thread.
RADIO!
Bruce, W 60 HZ
RFB, I had another Radio Shack
100 uH choke and put it in.
That helped a lot.
I rebuilt the final stage.
And Carl, I will be using your
bypass cap info.
I have to go to sleep, but I
will give a better report in
the morning.
Best Wishes,
Bruce, W 60 HZ
Modulation sounds good.
My power supply voltage and
current meters are analog,
as well as my QRP watt meter,
so it was a little hard to
read them, but:
With about 2 volts going to
the final stage, I got about
7 milliwatts out. With about
6 volts going to the final
stage, I got about 65 milliwatts
out.
The final is running. There is still
a lot more decoupling to do, but
I will be getting a few more chokes
and caps for that.
My money for this project has run out
for now. But I will have this transmitter
on the air - assuming it doesn't get unstable
when I connect it to an antenna. The transmitter
and the antenna will comply with the regs.
Also, I don't have the money for running the
transmitter off of a supply yet, so it will
probably run off of rechargeable batteries
for a while.
The "BT #002" will be on the air next week,
I hope.
My only listener will be my friend who isn't
too far away, who has a shortwave radio.
This is great stuff!
Bruce, W 60 HZ - - 1020 AM CC and
13.560MHz Experimental