Pioneer SX 590 Tuner/Amplifier

[Half a Mullard]

I wonder if anyone can help with this please?

Apart from a low background hum there is no signal from the amplifier. I have checked all the fuses and they appear to be intact. Although I am OK with valves I’m a bit ashamed to say that my transistor knowledge is practically zero. Therefore could I ask:

a) Is the amplifier a viable repair?
b) If so, is this something I could attempt myself, albeit with some ‘hand
holding’ or
c) Should I post a request in the Repair Services Wanted Section?

I would be very grateful for any advice!

[Nickthedentist]

Hi Howard,

Do you have a circuit diagram?

Power amps are notoriously hard to fix though, and those hybrid modules are all but impossible to get hold of nowadays, but it's worth having a quick look just in case it's something daft.

Nick.

[vinrads]

When you say low background hum is this on all functions , also check for dc voltage across the speaker terminals . Mick.

[John_BS]

1. Can we assume the hum is coming from the speakers, not the amp!

2. A lot of switches, so give them all 20 or so operations

3. As Mick says, is there no ouput for any of the input sources ?

John

[vidjoman]

On the photo of the front - you have the tape monitor button pushed in - that'll stop any input unless it's from a tape recorder plugged into the tape playback sockets.

[Half a Mullard]

Many thanks for your replies folks, much appreciated. Nick, I don’t have a circuit I’m afraid. I did wonder if it might be a difficult fix. Shame as it’s a very solid amp.

Yes, the hum is from the speakers. I’ll check for DC voltage across the terminals as Mick suggests. Can I do any harm leaving the amplifier switched on? Assuming that’s OK I’ll check all the switches and controls as suggested and report back. Thanks again!

[John_BS]

Circuit if it helps.
John

[vinrads]

I see there are two output modules ,it would be unlikely that both were faulty not impossible though, the next step is to check some voltages start with the STK's Mick.

[Half a Mullard]

Thanks folks and thanks for the circuit John

I fired up the amplifier this morning, faint hum from the speakers then silence. I released the monitor button (well spotted) but no avail. All inputs are dead. I checked for dc voltage at the output, zero on each channel. The speaker select buttons were set correctly. All knobs and switchs twiddled. Sadly all to no avail. The facia lights up and all fuses still intact. So I am a bit stumped.

Out of curiosity do the same precautions apply to transistor equipment as valves when applying mains for the first time?

Your help is appreciated.

[Half a Mullard]

I've now printed off John's circuit and can see the STK's that Mick refers to.

On removing the bottom panel I've exposed the PCB. I'm sure that I've seen circuit boards before with the components printed on but can't see any sign on this one so I'm just wondering the best way to take voltage readings?

[vidjoman]

The STK's are clearly marked with pin numbers on the front. Just remember to reverse your thoughts when looking from underside.
Also on the underside there is what appears to be some charring around 2 pins, which I think are the heatsink on a small regulator situated next to the 3 blue caps under the edge of the fuse board. That could have burnt out and needs checking for volts in and out.

[Half a Mullard]

Originally Posted by vidjoman

The STK's are clearly marked with pin numbers on the front. Just remember to reverse your thoughts when looking from underside.

I've looked at the board and I think I've identified the pins. It looks as if the STK's are plugged into a DIL (?) socket but only 6 pins are connected underneath which I assume are 1,2,3 and 8,9,0
So the next step is to check the voltage on each pin?

[Half a Mullard]

Originally Posted by vidjoman

Also on the underside there is what appears to be some charring around 2 pins, which I think are the heatsink on a small regulator situated next to the 3 blue caps under the edge of the fuse board. That could have burnt out and needs checking for volts in and out.

Well spotted, on closer inspection that does look to be the case. I've taken another photo where the support pins are clearly visible. I'm guessing that the regulator has 3 connections, possibly the ones I've indicated? If so, which pins need checking for voltage? Also if the regulator has failed will I need to find the cause first before replacing it and could this mean there will be no voltage on the STK's?

Apologies for this, as I mentioned earlier my experience in this area is limited to say the least!

[vidjoman]

That could be Q401 or Q402 and the voltage is shown on the circuit diagram. The pin pointing down could be a dry joint looking at the picture. It looks like the voltages are marked on the circuit so it shouldn’t be too difficult to find where they are wrong.

[Radio Wrangler]

Transistor audio power amplifiers have a stinker of a reputation as far as repair goes, but they're not actually anythink like as bad as they are reputed to be.

The reputed difficulty lies in the repairer more than the amplifier. They have a nasty eternal circle of destruction trick which catches people out. People are inclined to assume that a dead amplifier has only one fault. Find it, replace the part and all will come good at once they assume.

Not with these things. One transistor sets the bias conditions of others downstream of it, and worse, the correct bias of the input devices is set by the feedback from the output!

Transistors are fragile little things. You can subject them to too much current or too much power and they get hot, until they are destroyed by the temperature. It takes a fair bit of power to over-cook them. You can over-voltage them but it takes a fair amount more than running voltage to dispatch them so it's not easy. So far they may be looking to be quite tough, but they have an achilles' heel. We've been talking about forwards voltage.

reverse bias the base-emitter junction and you'll find it breaks down at around only 7 volts (5-12 sort of region) Pass a little current, sub milliamp and it looks like your transistor has survived.... or maybe? Try measuring its gain its leakage currents and most subtly of all, its noise factor and you'll find it degraded. Do it again and you'll find it degrades a bit further. run it above a few mA in breakdown and you'll kill it properly.

Ideally, a circuit design ought to have reverse Vbe protection, but with everything OK, you can ensure it doesn't happen unless something goes wrong. Usually the transistors tend to protect each other while all is working OK. But apply a transient to the output by say disconnecting a loudspeaker while it is playing and you can kill the thing. Some amplifiers have speaker selection switches to help you blow them up without having to even touch the speaker wiring!

Guess what this one has... they're on that diagram.

Marketing want these switches. Designers cottoned on to the risk, and fitted catching diodes at the output of their amplifier design clamping the voltage if it tried to go outside the power supply rails. Pretty much all of the time they sat there doing nothing until they were called upon to save the ship.

Accountants spotted these diodes, got the designers to admit they sat there reverse biased normally, and so the designers had to remove them. 4 diodes, 5-10p each, hey that's pure extra profit!

Guess which diodes aren't there in your schematic?

Another corner cut is that the bass/treble Baxendall-style tone controls aren't wrapped around their own preamp stage, but are blended into the power amp feedback loop. This robs the power amp of feedback when you apply treble or bass boost and it complicates fault finding.

Back to the power amp:

Say a transient or just a part failure crashes a power amp. one transistor dies and that screws up the bias on a neighbouring transistor, that dies and maybe passes the favour on.

You come along to fix the thing and start testing transistors. The first dead one you find, you replace then fire the set up to see if it fixed it. Nope. The remaining dead one just instantly ans silently killed your new one. You move on and test the next one. A-ha! so you fit a new one and try it out. Nope. Your recently deaded one now kills your most recently replaced one. You never thing the previous one you replaced could be duff. It's new, innit?

This can leave you circling in an eternal game of whackamole. Eventually you see you're replacing the same transistors for a second or even third time and you realise something big is wrong.

The way out of it is to switch the damned thing off and leave it off until you've tested everything cold and replaced all the dead ones. Then you dare try it. It might even work now.

To make life a bit more difficult, we now have lots of little device testers. Pocket sized, battery powered, dead convenient. Absolutely perfect for leading you astray. They test devices at voltages and currents far below those which they'll see in the amplifier. They can make devices which will break down and propagate further destruction look OK. You can imagine how confusing this becomes in the circumstances of a transistor power amp.

One way out is to choose a good curse word, say it, and then just fit a full new set of transistors.

So there is a route through the troubles. It looks like the long way round. There are seductive looking shortcut opportunities, but they almost all lead to disaster.

Few people can resist a short cut. Few people have the self control to slug it out the long way round. So, relatively few people can fix transistor amplifiers, and the amplifiers acquire a terrible reputation.

The good news is that most repairers CAN do a good job on transistor amplifiers once they know the nasty trap they pose, and once they muster the determination. Along the way they find out about those catching diodes, the triply-damned speaker switching and treat the feedback arrangement as an indispensable part of the bias system.

The output Darlingtons and the Vbe multiplier transistor used to bias them is integrated into theose output modules. The modules used to be a cost saving for the manufacturer and might even allow the management and accountants to shift to a posher golf club. But now, they're pretty much unobtainable or faked.

Unless that amplifier has special significance for you, if it belonged to a relative or if it's one you've lusted after back in the day and you really really want to fix it, then it's not a good example of decent performance. There are corners cut like the tone controls in the power amp loop and those STK modules. You could take the view that maybe the effort of fixing might be better spent on something without those modules and without the cut corners. Or you could take the view that you've got the thing so you might as well try in the process of learning about solid state amplifiers. Those of us who handle them without (much) drama weren't born knowing how. We all had to learn somewhere along the way and it was usually from our own mistakes with no-one to act as guide.

In audio amps, STK modules or obsolete power FETs can be show stoppers. Choosing an am to fix and live with means dodging these things.

The hifi amp I designed for myself has run 44 years without bother. It was done carefully and no corners cut. It would give an accountant fits. However, I used power MOSFETs and comparable ones haven't been made for the last 35 years. But I stashed a set of spares. AND a set of loaded boards.

Welcome to the wacky world of transistor audio power amps. Once you've fixed a couple, you immediately qualify as a mage of the seventh level in the view of everyone else, and then you can get your own back by pontificating at this sort of length...

David

[Nickthedentist]

What a brilliant synopsis, David, well done!

[Half a Mullard]

Indeed! Thank you David I’m going to print off your post and keep it for future reference!

In answer to your question, no, the amplifier does not have any particular attachment. It actually belongs to my wife, she brought it from her old house and it has been in dry storage for quite a while. I just thought it would be interesting to see if I could get it working again.

Given your very helpful advice and my general lack of expertise in this area I will put the amplifier to one side for the time being and try to gain a little more experience! However can I thank you and all the members who were kind enough to take up their time to offer advice.

Can I ask one final question though? Are there steps to be taken before applying mains to solid state equipment which has been out of use for some time? I’m sure that I read somewhere that a variac cannot be used?? so are there other precautions which need to be taken?

[Radio Wrangler]

I put a deliberate Arthur Mullardism in the second line, 'anythink' instead of 'anything' but I fear it got lost amongst my real typos...

Variacs are a yes-and-no sort of thing. Running on reduced mains voltage can have the amplifier run with less than full stress on various parts, but you have to understand that a working amplifier is a bias control arrangement in the shape of a ring each sub section setting the bias for the next around the ring.

But it has to start somewhere? Right?

It's hidden in plain view. The input transistor of those power amps is a dual transistor. Two matched ones on the same bit of silicon, packaged together. They are used in a circuit traditionally called a long-tailled pair. Their emitters are joined together and a relatively high value resistor to the negative supply approximates a fixed current source, pulling an almost constant current out of the pair of emitters. It's slight differences in the two base voltages that apportion this current between the two collectors. In posh-speak, it's a differential amplifier.

The left side base is driven by the ac-coupled signal. The right base is driven by the feedback network, which works from DC to above the audio range.

So, what sets the DC level on that left side base?

Easily overlooked, there is a 150k resistor to ground. This is a crucial part. With the amplifier stretched between symmetrical positive and negative supplies, ground represents the half-way voltage. So the left-hand base is set at zero volts DC and wiggled around it with the signal input voltage.

The right-hand base monitors the voltage from the feedback network. The gain of this little amplifier is quite high, and it gets multiplied by the gains of the later stages and the differential amplifier uses the later stages to steer the output voltage being sent to the speakers to force the voltage coming to the right base to follow what is on the left base.

There is so much gain that this following action is pretty damned good. Transistors aren't terribly linear devices and so a good audio amplifier needs to have plenty of forwards gain, so that it can throw away a lot of gain by having strong feedback. It all comes together nicely.

Looking at the feedback network in the schematic attached by John, it gets unnecessarily complicated by the tone control. But notice that the DC value of the amplifier gets to the right base via series resistors and shunt capacitors WITHOUT any attenuation. To make zero volts DC on the right base in order to match the zero volts DC on the left base, the amplifie has to hold the DC level on its output pretty accurately at zero volts. Altgether, this sets the bias voltages throughout the amplifier circuit.

Without this loop working, there is no control of the operating conditions in the amplifier at all.

A second thing to notice is that for transistor amplifiers looking at the DC level at their output is a very very good indicator of things being wrong.

(Older single-supply power amps need a pair of splitter resistors to make a half-supply reference voltage, and the amp tries to plant its DC output to follow it (and the big DC blocking capacitor on the output passes AC signal currents to the speaker.)

If you bring the power supplies up slowly with a variac there comes a point where amplifier stages in the power amp start to get enough supply to start working enough to turn the next stage on and so on. But things won't be right until all the stages in the loop are working reasonably well. Before that point, the bias of some stages may well be out of control. This could be damaging or cause oscillation, or smoke.

You must have seen those engineering drawings of impossible things where you could machine all the parts, and they could all exist in the assembled state, but which are deliberately impossible to assemble, held together by bolts with impossible to access heads, etc.

There are electronic equivalent jokes, circuits which could run just fine, but which could not be started without destroying themselves. Some audio amps come close. Some rely on going through a silly condition pretty quickly and come up working before anything bad happens. Designers work on a mental model of their circuit working nicely, of unicorns gambolling outside their windows. Rarely do they get worried over how it starts or shuts down without damage. Good designers do worry and also concoct hellish scenarios to throw at them in siulation and in reality to verify that they behave themselves evenif things outside themselves goes crazy.

So you can run into trouble and make smoke trying to bring them up gently on a variac. These are the amps which go up in smoke when the mains comes up slowly with odd reversals etc as the power network tries to get going again after a power cut.

You've probably formed the opinion that there are subtleties in doing a desent audio amp design. There are. You've already had the thought that it's difficult. There is plenty of evidence that plenty of commercial designers, designing amplifiers for well-known brands have put a fair collection of goofs into production. If it isn't immediately obvious, they get sold and the firm moves on to other models. Just because it's shiny, expensive and carries a famous badge doesn't mean it cannot be crap.

The counter to this trouble isn't even difficult. It just amounts to thoroughness, of noting errors in other people's designs and noting your own errors. Of not taking short cuts. There are places where short cuts are safe, but they aren't in transistor audio amplifiers.

So there you have it, warts and all.

On the forum, we have a lot of people who would have a go at building a valve amp, but would never think of doing anything with a transistor one. It's a bit of a shame. Transistors have limitations and problems, but ways of getting round this have evolved and are successful. We're in an age where transistor amplifiers offer freedoms that can get around the limitations found in valve amps and also the transistor's quirks can be successfully evaded.

It can be fun. I used to give an occasional lecture to the engineering school of Dundee University on the process of design and how to survive as a designer. I used a Quad 303 power amp as an example and showed how you couldn't possibly repair one because you could no-longer get transistors BAD enough. Seeing their faces as realisation struck!

David

[Radio Wrangler]

Oh, rather than a variac, a pair of lab type regulated power spplies with current limiting CAN be used to gently test a power amp. This can save money in fuses and transistors.

You can also run some amps without their power transistors. No loads connected, but using thre drivers to drive nothing more than that essential feedback network.

With care and a few temporary lash-ups it's possible to bring up a new design in stages if you have to.

David

[Techman]

Originally Posted by Nickthedentist

What a brilliant synopsis, David, well done!

I read it earlier and I can't help but agree It has to be said, absolutely brilliant!

Probably should bookmark it and point people to it when the question comes up in the future.