The AM Forum

Ok, I'm working on this KWS-1.
Got it making some signal through the low level stages.
It's making signal @ the VFO frequency.

The way it is supposed to work is the VFO +/- 250Khz = Dial Frequency.
The "unwanted" lower frequency is rejected by a tunable "trap".

I'm not getting much if any mixing action.

The attached pdf file shows the circuit and description.

They call the tunable section that sits between pins 1 and 6 is referred to as a feedback
filter. Since pins 1 and 6 SHOULD be out of phase with each other, whatever is output on
pin one and then tuned to the lower ( -250kHz) frequency will cancel if it is also on pin 1.
Good enough idea.

I'm seeing ONLY the VFO frequency at the output, as far as I can see, so far.

No sign of the 250kHz component.

What puzzles me the most is that the text refers to the ratio of the VFO to the 250kHz signal
as a bout 1.5v to 0.1v respectively. I'm failing to understand how one mixes two signals
of such disparate amplitude and gets a usable sum and difference signal??

I find those two voltages, approximately... seem reasonable.

Maybe this works just fine, and I have some bad components in or around the mixer/filter.
Have yet to find that. It's also not the easiest place to work in. (nice design work there
Collins! Yeah... right.)

Anyone got insight/ideas??

                   _-_-bear

Hi Bear
The signal from FL101 should be the SSB signal I suppose, so the amplitude has a wide dynamic range.
If the signals are present at the grids (VFO and 250 kHz SSB) AND if you get a DC of 1+ volts at the cathode (meaning that the tube pulls current) than I think that the only reason can be that the tank circuit is not good or one half of the tube is dead. But that you can verify at the cathode, there you should see both signals though the VFO will be much stronger
Can you couple an RF signal into pin 6 ( the tank circuit) via a few (2,2 - 4.7) pF and measure the resulting signal there in order to find the resonance frequency of the tank circuit?

Look closely - the "carrier level" bypasses the FL101 to insert 250kHz "carrier" for AM... or CW.
OH! Yeah, that may not be in the pdf... but it's in the full schematic...

So you can ignore the balanced modulator & mechanical filter for the purposes of looking at the mixer...

One can simply use the 250kc INJ (BNC) as an injection jack... remove the oscillator tube and
there's no signal there! :D Or alternately use the VFO INJ jack the same way!

I'll have to do just what you suggested to see what that "filter" is doing.
I'm still completely puzzled buy the difference in input signal levels and how they mix properly.

Thanks for the suggestions!

               _-_-

After applying some inductive reasoning, I've come up with the idea that this circuit is mostly "trap"
for the VFO frequency... leaving the pizzweak sum + difference signals... or perhaps working like
a high pass, or trap for both the VFO and the lower sum...

                      _-_-

I know the full diagram, I once had a KWS1, by when moving had to sell it  :'(
The signal from the FL101 filter has a max of 1.5 Volts when whistling in the microphone. for CW you need to inject approx the same level 250 kHz, so not really piss weak. Low voice SSB from the filter is, a few tens of a volt.
The mixer does have gain, so if the tank is tuned well, you can expect a few volts upto 10 Volts RF at VFO + 250 kHz.
Indeed this kind of mixer does not balance the input away, so there is a trap needed for the VFO frequency to attenuate that frequency at the output. That  is the left part of the tank circuit.
When you don't find that signal, the tank is not tuned or the tube is bad.

Yep... it's not working as it should.

I'll be investigating more fully, and disabling the 250kHz osc so that it is easy to tune the trap, then
turn it back on and try to peak the output...

It calls for 0.1v from the 250kHz side in the book... I'm seeing about that amount.
As I mentioned, I'd have expected closer to a 1:1 ratio, which is part of why I'm puzzled as to what it
is supposed to do (the mixer).

BUT not more than that with the CW INJ turned all the way up.
Not seeing anything obviously wrong though...

It may yet turn out that one or more caps has moved or otherwise gone south.