Hacker RP33 Autocrat

[tiv48]

ah direct coupled transistor amps dont you just love them (all together NO!) a fault anyware in the circuit can cause mass destruction ,in this case even a fault around T1 (upsets mid point voltage) can kill t3,t4,t5.
the give away is inability to set quiescent current (4ma) on this one
ps yours uses the newer lfk4 package so line up is tr1 bc108 .tr2 ac128. tr3 ac127 .tr4 ac127 . tr5 ac128.
best of luck rob t

[Joe]

I was using PP9s when I blew those transistors. I wasn't sure if my PSU may be the cause of the trouble and it can only manage an output of about 15V.

I've re-examined the corpses and found T3 (which is an AC127) and T5 are short circuit and T4 open.

I can get the quiescent current down to 12mA and it work at low to moderate volume with things cool. Turning it (the volume) up higher caused a machine-gun type noise (motor-boating?) which the ammeter needle spikes up in time with it.

I was dong this with it connected to the PSU which was delivering 15.45V. When I switched to batteries it didn't work.

So it looks like progress is being made but there's a way to go yet. It seems like a stability thing and I am wondering if there'd be any point in changing T2 so that all four would be silicon ones.

As to what was initially wrong, it seemed blooming loud. To listen to it the volume knob only needed to be just clicked on which is what lead me to think that there was something set wrong which was causing over-loading.

- Joe

[pwdrive]

Hi Joe, you need to replace with the original types (or near subs.) if you can, mixing silicon with germanium can upset the DC conditions, it is interesting that you say it can be made to work (sought of) when using your PSU but not on batteries, are the batteries ok? Even though your PSU is only putting out 15 volts the radio should still work reasonably well if there are no faulty components in the set. The motor boating you describe sounds bad, does it motor boat with the volume control turned down to minimum? It might be a B+ A/C impedance problem either within the receiver or the PSU or excessive feedback somewhere in the receiver caused by a particular component/break/short etc. I realise it's easier said than done but you need to diagnose from a known good base line, it's all to easy to go round in circles if you don't, good luck

Lawrence.

[Mark Hennessy]

Personally, I like direct coupled transistor amplifiers.

It's funny - I had an uncle who was largely responsible for starting me off on electronics - not least by giving me a soldering iron for my 5th birthday! He was an RF person, and worked for Thorn (latterly Thompson) in Enfield, and I believe he had a hand in the designing the tuner in the TX10 and others. He has a callsign, and built all sorts of weird and wonderful RF things. But despite all this, he really didn't understand anything to do with audio! Well, that's what he said - I reckon he was being modest.

I could never get my head around this, because the stuff he did was well beyond my comprehension - still is, really. Audio, to me at least, is much, much easier. Like you Rob, he cited the DC-coupled nature of everything as the problem - yet DC-coupled amplifiers are used in many walks of life beyond audio - video circuits and oscilloscopes, for example... And I know that he designed and built video gear. Still, I guess some of us are just "wired" differently...

Sorry to ramble!

I don't think that a fault in T1 could damage anything. After all, under conditions of overdrive, T1 could be driven into saturation or total cut-off, and if being overdriven resulted in damage, it would be a pretty poor design! Actually, a common fault-finding strategy is to short-circuit base and emitter (to turn it off) or s/c collector and emitter - I do that with a pair of tweezers while observing the midpoint voltage on a 'scope.

The biggest cause of problems is normally a faulty output transistor. The initial failure could be caused by random component failure (maybe tin whiskers!), overheating or over-current. What happens next depends on the design and a degree of luck, but let's assume that one transistor has a collector-emitter short. The midpoint will jump to the rail or ground, depending on whether the upper or lower transistor when s/c - but the rest of the amplifier will see this, and via negative feedback will attempt to bring the midpoint back down, and it does this by driving current into the base of the working output transistor. What happens next depends - the other transistor might pass enough current to bring down the power supply and no further harm will result - apart from perhaps flattening the batteries! On the other hand, if the power supply is a "beefy" one, it will "win", and the other output transistor will fail through excess collector current or overheating...

In terms of further consequential damage, if this was a big hi-fi amplifier with no output capacitor and +/-50 volt rails, your loudspeakers will be somewhat unwell after all this! Decent amplifiers have a protect circuit that disconnects the loudspeakers via a relay should DC offsets appear on the output.

Problems can propagate back down the amplifier too. What tends to happen is that the junctions in the output devices become a great big blob of metal, meaning that all 3 terminals get connected together. So the rails appear at the bases of the transistors - and thinking about it, this is probably how T3 bought it. Oh, and have a look at R14/R15, because these probably cooked during the "event". T2 should have survived, but you never know...

Again, sorry to ramble, and sorry if I'm stating the obvious here - but hopefully all of this reinforces my advice to use a current-limited PSU

I reckon that Joe needs a break from this set, so adding a current limiter to his PSU would be a nice little diversion

Oh, and regarding the transistors, I know that the manual mentions LFK4 - something that I hadn't heard of before, but the good folk at UKVRRR came up trumps: http://www.vintage-radio.net/forum/showt...hp?t=57721

But...

One thing to check; with at least 2 examples that I've looked at, the biasing arrangements around T3 were different to the manual. Rather than the NPN AC127, there was a PNP something or other... As a result, R13 was connected to a different part of the circuit.

I've looked some photos of mine, and they use an OC71 for the bias generator, not an AC127. R13 connects to the junction R11/R12 from memory (can't see in the photos)

All good fun

Mark

[Mark Hennessy]

I was using PP9s when I blew those transistors. I wasn't sure if my PSU may be the cause of the trouble and it can only manage an output of about 15V.

OK, so good batteries are obviously "beefy" enough. Actually, modern PP9s seem to have an impressive capacity and internal impedance...

I've re-examined the corpses and found T3 (which is an AC127) and T5 are short circuit and T4 open.

I've just checked your serial number (27684), and see that it's quite a bit earlier than my 2 (43867 and 45834). So that's obviously a change that was made in the lifetime of the model - I see another data collecting exercise coming on

I can get the quiescent current down to 12mA and it work at low to moderate volume with things cool. Turning it (the volume) up higher caused a machine-gun type noise (motor-boating?) which the ammeter needle spikes up in time with it.

Check the 5-pin connector. Also, I think I'd change the remaining electrolytic capacitors now. If only to return the set to "as new" condition so we can start to apply logic.

I am wondering if there'd be any point in changing T2 so that all four would be silicon ones.

No, none whatsoever...

Silicon vs germanium questions need to be put into the context of the circuit you find them in. Often, no changes are needed when you swap them - sometime it's really important. Here, you've changed the output pair, and that requires a change in bias voltage, but as T3 provides the bias voltage, and T3 is a "Vbe Multiplier", it's automatically already in the right ballpark.

As to what was initially wrong, it seemed blooming loud. To listen to it the volume knob only needed to be just clicked on which is what lead me to think that there was something set wrong which was causing over-loading.

This is a common fault with old volume potentiometers - basically, the connection between the end of the track and the external connection goes high-resistance.

Don't worry about "overloading" - if you drive the amp into clipping it won't sound nice, but nothing will break. Once built into a well-designed circuit, transistors are not as fragile as valve aficionados would have you believe. They get a bit grumpy if you overload them, but wouldn't you?

[pwdrive]

Hello Mark that was an interesting post regarding your Uncle and DC coupled amps, I used to do a lot of DC coupled stuff years ago...many years.
I worked for Sony of Canada as head tech. at their main sevice dept. in Vancouver, Sony made a lot of stuff in them days (wide product range) TC & IC this and TA that, to be honest as a radio/tv sevice engineer I have to say they were the best company I ever worked for (no tight fisted spares dept) any spares, any bench test gear was no problem, I had six techs under me there, If they could not fix it then it ended up on my bench, same went for most of the other Sony service depts in Cananda if they could not fix it the offending unit ended up with me also, used to get stuff sent from all over..Halifax to Yellowknife, those were the days.
As you say, if you are familiar with DC coupled audio then servicing that stuff is not usually a problem in a service dept. context but at home it can be more frustrating as transistors usually cost us money. Good luck to Joe anyways, I'm sure he will get it working ok.

Lawrence.

[Skywave]

Hi Joe,
What you've learnt there is that fixing d.c.-coupled amplifiers require a different approach to the classic 'switch it on and wander about with a voltmeter measuring various voltages'. That's maybe O.K. for valve stuff - but it's a real no-no here.

When you have a fault like that, the first thing to do is to switch it off - quick! Then remove all the transistors and diodes, one by one, and check every one for internal shorts or open circuits; replace any that are faulty. Check and if necessary, replace, any low-ohmic value resistors - especially emitter Rs. in any O/P pair. Then remove and check all the capacitors (there probably won't be many): replace any that are faulty. If there is a quiescent bias pot, check it; replace if necessary; set it for minimum current in the O/P transistors. (Don't be tempted to do this by connecting the amp. up to the PSU yet!). Check for mechanical open or short circuits on printed tracks. When you're really sure that you've got all of the 'nasties' out, make sure that the amp. has no signal input, hook up a load on the O/P ( I use a resistor and hang a 'scope across it) & connect the amp. up to a current-limited PSU. Switch on and look at the current meter in the PSU. If the current looks way too high, switch off immediately . Now go back and re-check every component - just like you did before. ( I know, I know! ) Until you can eventually escape that loop, there's no point in trying anything more adventurous - like trying to pass a signal through it. Next step is to set the quiescent current and check the temp. of the O/P transistors and check a few critical d.c. voltages.
It's only when you've got to that point is it worthwhile testing the amp. with a signal - and even then, something may go pop!

So there we are: d.c - coupled amps. can be a real pain. As anyone who has repaired Tek. transistorised oscilloscopes can tell you! (But that's another story for another time )

And finally, I now have a Leak Stereo 70 (needs mending) sitting in the top of my in-tray: writing all the above has reminded me of that necessary approach.

Al.

[Joe]

I can't sort the PSU out yet as the extra control won't fit in the current (no pun intended!) case and it's too wet to go out and cut the wood to make a bigger one. It's certainly no chore as that's the sort of thing I enjoy doing.

I was thinking about changing T2 as the data says T2 and T3 and T4 and T5 are matching pairs.

Interestingly it also says that the EARLY version has OC81s

As you say, it's probably time to have a fiddle with something else for a bit. At least I'm learning more this way.

Thanks

Joe

[Skywave]

At least I'm learning more this way.
Joe

Quite so, Joe, quite so. You may sometimes wonder (as indeed I do) as to how some people manage to make some repair jobs (and not necessarily electronic, either) appear so easy for themselves. The answer of course is hard-won experience! And such experience does not come cheap! Yet despite having an almost unlimited range of experience, we all lose a few now and again: it's what is known as Living in the Real World.

Al.

[Mark Hennessy]

Hi Lawrence,

That sounded like a really interesting job. I must admit that some of my happiest memories are from working in test or service departments. I would have like to have done more design, but I'm much happier to do that as a hobby with only occasional professional jobs. Fault-finding does take a certain mind-set, but it's something that most people can pick up with help and practice.

The biggest problem is that most circuits have negative feedback, and it's not always practical to "break the loop". Everything becomes a chicken and egg problem. Earlier I mentioned a set that is causing me grief, and I'll write that repair up separately when it's done, but I must admit that I'm running out of logical ideas and am about to change every single component until it's fixed! Even the best of us get caught out sometimes