[kiwi_steve]

Well, I'm at a standstill. I've ordered 2 NOS 78's and a NOS 6A7 from the US - so thats probably going to be a couple of weeks away. I'm staring at it but seeing nothing helpful any more and getting a little annoyed at it, and myself. It for being such a PITA and me for getting annoyed at it for being a PITA. Its simple, yet I can't see the wood for the trees so I'm going to put it away and work on something else for a while.

One other very annoying 'feature' (we use this word in software engineering - undocumented feature - to justify bugs in our code) it has is that it always sounds like its tuned just slightly off-station. No matter how I tune or align it, speech has that hissy distortion that makes it sound very much like someone turned the tuning knob a little bit off frequency. Interestingly this is easier to hear if I am not in the room than if I'm right beside the set on the bench.

One thing I did sort, with the help of Marc from that other forum, was the SG voltage... which was way high. One of the two resistors generating that voltage had drifted way low since I first measured it (Should have been 16k, was around 14.8k when I first measured it from memory... was 12 when I measured it the other day (and yet no warning flag went off in my mind... sigh) and just over 9 sitting on the bench now. Replaced with 2 x 33k 1W in parallel. That knocked about 25V off the screens. It also seems to have knocked about 1V off the AGC line when I have a strong local station tuned in - but its still almost 3V which is too high for the shadow meter.

Maybe a 37-640 will be easier to work on... how hard can an RF sub-chassis be?

[Mondial]

Steve, where are you measuring the AVC voltage? Is it the actual AVC voltage across the detector diode load (the volume control), or are you referring to the voltage across the shadowmeter coil? The AVC voltage at the detector should be on the order of negative 10 to 12 volts with a strong signal.

The screen voltage is not really significant, in that the AVC action will compensate for the increased gain caused by the high screen voltage. The negative bias generated by the AVC circuit will lower the tube amplification in opposition to the increased amplification of the higher screen voltage. In the end, the IF tube plate current which activates the shadowmeter will remain about the same with a given received signal strength.

You might check those two 110 pF caps at the detector output. They filter the IF freq signal from the detected audio. If they are not attenuating the IF signal, it can cause the raspy audio you describe.

You could also have an IF oscillation, which may cause the poor audio. Does the radio tune in a station symmetrically from above and below the center carrier freq? That is, as you tune to the station from below its frequency, does it sound the same as when you tune down to it from a higher freq on the dial?

[kiwi_steve]

Steve, where are you measuring the AVC voltage? Is it the actual AVC voltage across the detector diode load (the volume control), or are you referring to the voltage across the shadowmeter coil? The AVC voltage at the detector should be on the order of negative 10 to 12 volts with a strong signal.

Sorry, yep I'm measuring across the shadowmeter... another casualty of my brain having been staring at this too long... I need some fresh air and sunlight

The screen voltage is not really significant, in that the AVC action will compensate for the increased gain caused by the high screen voltage. The negative bias generated by the AVC circuit will lower the tube amplification in opposition to the increased amplification of the higher screen voltage. In the end, the IF tube plate current which activates the shadowmeter will remain about the same with a given received signal strength.

Good point - although my concern with the high screen voltage was more that it was about 25% over the max rated running point of the tubes. This all happened in about a week on the bench... it wasn't that high when I started, but the big power dogbone that made up half of the voltage divider decided a change was as good as a holiday... and it needed a holiday

You might check those two 110 pF caps at the detector output. They filter the IF freq signal from the detected audio. If they are not attenuating the IF signal, it can cause the raspy audio you describe.

I replaced these a few days back - although not with micas, just high-reading poly film 1nF caps. How critical are the values here? I'm assuming this is a PI filter arrangement. If I calculate the filter frequency of a 110pF and a 20k resistor it comes out to around 72kHz... I suspect I'm looking at this wrong though? I do have some micas coming by the way - just have none on the shelf at present... Also, I used a 22k resistor - if that value is critical I can throw a 220k resistor across it...

You could also have an IF oscillation, which may cause the poor audio. Does the radio tune in a station symmetrically from above and below the center carrier freq? That is, as you tune to the station from below its frequency, does it sound the same as when you tune down to it from a higher freq on the dial?

It seems to, but I will check this later today...

Thanks for all the help - I'll leave it on the bench for now and see what I can find.

[Mondial]

Steve, regarding the problem with getting the RF circuit to tune, have you properly located the dial scale on the tuning cap?

From your description, it seems you have too much RF stage tuning capacity for a given receive freq, which is why your trimmer is all the way open and still wont peak. How does the dial freq calibration compare to the actual freqs of received stations?

 It seems you need less tuning capacitance in the RF stage and since the trimmer is all the way open, the only other thing to do is to tune the tuning cap higher (less capacitance) and compensate with the oscillator trimmer so you receive the signal at the higher dial location. Sometimes you have to play with the position of the dial scale on the tuning cap shaft to compromise dial accuracy and tracking of the RF and Osc circuits.

The other thing to check is that the IF is accurately set to 460 kHz. You mentioned that you had to retune the IF as all circuits were off. Are you using a freq counter or a synthesized generator to accurately generate the IF signal? I would not trust a usual service grade generator to be accurate enough unless you are setting the freq with a counter.

[kiwi_steve]

Great suggestion... I will check the historical screw mark on the shaft against where I have it sitting now and have a bit of a play with that and report back.

Currently the dial is accurate - set initially using the sig gen and then tweaked using known stations at the top and bottom of the dial. Its pretty bang on through the range at the stations I listen to.

The IF frequency is set pretty accurately I think... I use my digital scope to check the frequency of the sig gen (a leader LSG-200). I generally leave this connected throughout the alignment to check the frequency. Its a bit dodgy on higher ranges (I think thats the sig gens fault, not the scopes) but seems fairly accurate and stable at 460kHz.

Cheers

Steve

[kiwi_steve]

Had a quick play with it last night - didn't have a lot of time... but I wasn't able to get it to peak without losing the ANT trimmer peak (tightened down fully before the peak), and same with the OSC... but I didn't really have time to fully investigate that.

I also noticed that the volume output increased significantlywhen I accidentally hit the ANT coil can, then settled back to normal after a second or so. Same thing happened when I tapped the OSC can... instant boost lasting a second or so then cut back to normal (not a gradual decline)... dry joint somewhere? I'll pull those two cans apart and check them properly when I have some time, hopefully tomorrow - open night tonight, I'll be showing parents what we do to their kids during the day

Cheers

Steve

[kiwi_steve]

Steve, regarding the problem with getting the RF circuit to tune, have you properly located the dial scale on the tuning cap?

From your description, it seems you have too much RF stage tuning capacity for a given receive freq, which is why your trimmer is all the way open and still wont peak. How does the dial freq calibration compare to the actual freqs of received stations?

Had some free time so shot home for a tinker (advantages of living 3 minutes from work) - Nope - no position of the tuning cap that I can get the OSC trimmer to peak on will allow me to peak the RF trimmer. I'm about to pull all the cans apart and go over them with a fine tooth comb.

Oh - and the volume thing was a loose joint on a mica cap I had temporarily soldered across another one (the 110pF cap off the BC RF TX)... removed it now because it didn't affect the tuning in any way.

Steve

[HOTWIRE]

Hi Steve.
Just wondering. How did you get on with those cans, Did you have a closer look at them solder joints etc.

Or. Is it Parked. LIKE MINE. for now.

Regards
Will.

[kiwi_steve]

Parked, but I'm working on it in the back of my mind...