Hacker RP33 Autocrat

[Joe]

Thanks, Al. The circuit diagram is in post 29 above.

I'm still making use of the decent weather so will investigate later.

Cheers

Joe

[Joe]

Pictures, though I'm not sure how much help they'll be.

This is with 1kc/s going in at the volume control with the volume set low. Top trace input, bottom one across the speaker.

   

Here's it turned up a bit but still SOUNDING OK

   

This is with it turned up further and distorting

   

This is Absolute Radio at low volume

   

and turned up

   

They don't show on the photo but there were fine spikes reaching to the top of the screen.

- Joe

[AlanBeckett]

As transmitters go that seems to be working quite nicely :-)
Put 1,000p between Collector and Base on the Silicons and try again.
Alan

[Joe]

I thought it looked a bit like that!

- Joe

[Skywave]

The cct. diag.: got it, thanks.

So I see that is doesn't use diodes to bias the O/P transistor pair: it uses a transistor: O.K. Now reading through what you've reported so far, the following seems likely. On the ext. PSU, things seem reasonably O.K. - and there is no motor-boating. But when powered from the batteries, the trouble begins. You can adjust the quiescent to a certain value, and then that current jumps. All that, to me says this: Your ext. PSU is delivering a PSU voltage that is about 3v less than what the batteries are delivering - and that that 3v difference is critical (it shouldn't be: it's a symptom of the fault condition). When you adjust the bias pot., the current through the O/P pair correctly increases up to a point - then trouble begins. The link between that and the two sources of d.c. power is that either change (PSU source or bias adjustment) is increasing the gain of the amp. but to a point where it is going unstable - hence the motor-boating (low-freq. instability). I don't think you have any faulty semi-conductors now. So, on the basis of all that, I'd change all electrolytics in the feed-back path, one at a time, testing after each replacement, starting with C2 - then C3, C4, C6 and C9

Amendment!
I've just seen your recent post showing waveforms - I was about to suggest hooking up a 'scope to do just that! And now we have a much better idea of what is going on: it's gone unstable - at R.F.: high-frequency instability !  I thought that might have been a possibility: I was on the brink of suggesting that. So forget the above comment on electrolytics for now - change C9 first. If that doesn't fix it, change the following, testing after each replacement: C6, C4, C3, and C2. If none of those replacements fixes it, go back to my earlier comment about replacing the electrolytics.

Al.

[Skywave]

Put 1,000p between Collector and Base on the Silicons and try again.
Alan

Yes, I understand the thinking behind that, but silicon transistors? The only silicon I see is the BC 108 at the front. O.K. perhaps I've missed a bit of this whole story. If some of the germaniums have been replaced with silicons, that explains a lot - since silicons will have a much higher current gain - especially at R.F. - than germaniums! I'd replace any of those substituted silicons with the correct germaniums first, before proceeding further - as per my immediate post above.

Al.

[Mark Hennessy]

This amplifier uses dominant pole stabilisation that includes the output stage (C9). This means that output stage distortion at HF is reduced, but including the output stage is normally risky and can risk oscillation. If replacing C9 fails to help, adding 100pF or so between collector and base of T2 might help.

But it would help to know the oscillation frequency. If it's a couple of MHz or less, then you have Nyquist oscillation - i.e. the whole damn lot is oscillating. But if it's much higher, then it could be local oscillation in one of the output transistors - very common with emitter followers. In which case, putting 100 ohms in series with each of the bases should stop it.

I suspect it's the latter...

Remember the old expression: Amplifiers oscillate; oscillators don't

Good luck,

Mark

[Joe]

Sorry Al, I should have posted an amended diagram. T3, T4 and T5 are now silicon. I haven't got any germaniums and part of my reason for doing this has been to find out how (if?) this circuit can be made to operate with the easier to come by silicon variety. I'll certainly change the capacitor you suggest next.

I tried Alan's idea with mixed results. A cap between the collector and emiter of T4 and/or T5 sent it utterly Deolali but one across T3 seemed to help. I could set the quiescent to the correct 4mA and the traces are better:-

1kc/s at low volume
   
Turned up to about a quarter
   
Absolute low
   
and turned up
   

It's a long way short of right but is getting there.

- Joe

[Skywave]

Joe:
I hear what you say about not having the correct transistors: germaniums types.
Please excuse me if the next lines sound blunt: I am trying to help. I honestly believe that until you fit the correct types, you are really wasting your time and effort pursuing this further. As I said previously, silicon transistors generally have a much higher current gain at high freqs. than germanium types - hence, doing what you have done runs the risk of the whole amplifier going unstable at some high frequency. Of course, if you're happy to redesign the amplifier . . . .

Al.

[Joe]

Thanks for your honesty, Al.

- Joe