Beomaster 3000

[Michael Maurice]

I'm having a bit of trouble repairing this Beomaster.

The LH channel is distorted, I can't set the bias current to anywhere near what it should be 100mA and the collector of TR44 which should be at 0.5V is actually 0V.

I've replaced a couple of electrolytics and the trimpot itself.

All resistors read OK, the mid point is ½VCC as it should be.

The voltage at the emitter of TR48 is correct at 0.2V

The base/emitter voltage on TR47 is 0.2V instead of 0.6V

The only thing I can think of is to replace TR43, 44, 45, 46, 47 and 48. Unless of course someone else can think of a likely cause.

The PDF is the same as the JPG but should be easier to read.

[Radio Wrangler]

Not enough bias current? or too much?

I might have mis-read the diagram, but the collector of Q44 looks like it goes to the supply rail, so 0.5v or 0v would be wrong.

You might want to remove TR46, TR43 temporarily. They are current limiters and steal base drive from the quasi-complementary output pairs. If these transistors have gone leaky they can lead you a merry chase.

What is the DC output voltage at the midpoint of the two resistors, 491 and 49? is it close to half the supply rail?

Whenever you get mysterious voltages or currents in a transistor amp, it's worth a poke with the scope just to make sure things are stable.

If the midpoint voltage is reasonable, it may be that the quasi complementary pair are OK and it's just bad bias to be chased.

David

[flyingtech55]

Hi Micheal

I worked for a B&O dealer in the past....

We used to get these older units in from time to time and I've had that type of fault before. What I used to do was renew all the small signal transistors in one go. Something weird seemed to happen to one of the junctions in one of the transistors. Obviously working for a dealer meant you couldn't spend time narrowing it down and the small signal transistors were pennies compared with charging out time. I know that's not too helpful but that what we used to do. All you had to then was set the pots, bias and midpoint, soak test it and send it on it's way. I don't think there was anything too demanding in the small signal transistors although, of course, I got the set direct from B&O back in the day.

HTH

TimR

[Michael Maurice]

Thanks David and Tim, I'll replace the small signal transistors and report back. Cricklewood have them in stock. I'll report back later.

[Michael Maurice]

I replaced all those transistors and I even replaced TR45 but there is no difference. I will be doing more tests and update you

[Ambientnoise]

Michael, as David said earlier, it’s worth scoping the various points as a parasitic oscillation is possible. I had this on a B+O unit due to two dried up caps.

Ken

[Radio Wrangler]

The DC output voltage (upstream side of the big DC blocker electrolytic driving the speaker) is always one of the first checks on a transistor amplifier.

It's one of the main indications of health.

If it's rammed over towards the supply rail, or towards ground, you'll get silly numbers trying to set the quiescent and plenty of distortion.

David

[Michael Maurice]

David, the mid point, that is the output prior to the DC blocker is around 29V same on both channels.

Ken: I've scoped all around, no parasitic oscillation, it looks as quiet as a mouse.

I changed a couple of resistors that were slightly out of spec, but to no avail.

[Radio Wrangler]

Hmm, that doesn't sound silly if the supply voltage is around 60V

This would suggest that most of the amplifier is OK and is running with feedback control working.

Maybe if the Vbe multiplier transistor is leaky, but you've replaced all the small transistors.

Maybe something in the network controlling that transistor.

Have a look at the collector-emitter voltage on TR45. It should be 4xVbe + the voltage expected across the pair of resistors (493, 497) in the output totem pole. say 2.6v plus the expected drop in those resistors. If you move the bias current pot around, the voltage should change.

Something odd seems to be happening, so it's time to engage low gear, to trust nothing, and to slowly check and double check everything. When I get stuck with this sort of thing, it's usually an assumption I've made that blinds me to soma clues. Looking at everything in at least two ways breaks the log-jam,

DAvid

[cathoderay57]

All of the trim pots suffer from dirt and poor contact between wiper and track. Make sure you have checked all of them - I seem to remember there was more than one. The amp also suffered dreadfully from tarnished contacts on the many switches which caused distortion and loss of sound on one or both channels. When it worked, it was lovely! Best of luck. Jerry

[Max Ripple]

It must be a typo Michael,but TR44 collector would be 60v,do you mean TR47?
I would replace the "bootstrap" cap (475) and the other ROE caps too once you have sussed the main problem !
Check 484 ref zener,and the PTC for resistance(compare to the other channel)
The PTC lives on the heatsink and can become disconnected or dud.
496&496a protection diodes can be troublesome but measure ok.I usually replace them anyway.
Lovely old amps these!
The "achilles heel" as other's have mentioned is the complex function select switch,which can cause all sorts of "funnies",including missing channels etc,due to the speaker selection routing.
There is eg, provision for connecting another amplifier,to allow sharing of speakers.A throwback to the tape recording day's of course,where a tapedeck could be connected,and monitoring possible.
Nick

[Michael Maurice]

Originally Posted by Max Ripple

It must be a typo Michael,but TR44 collector would be 60v,do you mean TR47?

I do mean TR47 sorry about that

Originally Posted by Max Ripple

I would replace the "bootstrap" cap (475) and the other ROE caps too once you have sussed the main problem !

I've done that

Originally Posted by Max Ripple

Check 484 ref zener,and the PTC for resistance(compare to the other channel)
The PTC lives on the heatsink and can become disconnected or dud.

I've checked that, it measures the same as the good channel both resistance wise and voltage wise.

Originally Posted by Max Ripple

496&496a protection diodes can be troublesome but measure ok.I usually replace them anyway.

All replaced.

There is no change, what I do find puzzling is that the voltage across the trimpot is zero as is the voltage C - E of TR45

I haven't replaced the zener diode as I cant get 1.5V, besides the voltage on the base of TR48 is roughly the same on both channels.

However the BE voltage on the good channel is around 630mV whereas on the duff channel it's around 590mV.

I'm going to try and get some 2.0V zeners and see what happens.

[Michael Maurice]

I've replaced the zener diode with a 2V one but its made no difference, VBE is slightly up at 0.6V, still no volts on TR47 collector.

[Radio Wrangler]

Let's break it up. We can fault find half of the amp with out the other half.

Remove RR43 and TR46 so the overcurrent protection is out of action.

remove TR48 (The VAS stage and the hardest-worked transistor in the amp)

Now R478 is 1.5k and so is R477. I think, so put 3k (two 1.5k from where TR48 collector had been to ground.

Now we have the Vbe multiplier TR45 foating between a pair of 3k (total) resistors, one to ground, one to 60v. This should sit about midway. If not, something is taking or adding current. The output (midpoint of those two equal 1W resistors and the + end of the 3000uF electrolytic) should be near 29v

You should have a few volts c to e on TR45.

You can play around with the bias current pot and the output bias current should be OK.

You now have a lot less amplifier to debug at once. Just the vbe multiplier, one darlingtom NPN emitter follower and a quasi complementary faked PNP darlington emitter follower.

Diode 496 is an attempt to balance up the Vbe drops of the simulated complementary pairs.

David

[Max Ripple]

Just a couple of long shots here,but what about the coupling cap between the amp and previous stage?It looks like 461(1uf)in the drawing.
It's a 'tant from memory,and these were troublemakers too.Might be leaking dc?
Otherwise,check the connections to the main output transistors on the heatsink.
The originals were Motorola types,with aluminium cases,and covered with a n isolating plastic cap.The collector(case) is connected via a countersunk screw so that it pierces the plastic cap,and connects through to the sub pcb.
The screw can become loosened or oxidised.
Might explain the odd dc behaviour in the stage?
Nick

[Michael Maurice]

I've done some more work on this and got absolutely no further.

Max: I checked the output transistors and indeed there was a problem in that a screw had had sheared, so I replaced it making absolutely sure it was contacting the body of the transistor. I also removed the 1uF tant capacitor on the input, it made no difference

David: I carried out your suggestions as follows: I removed TR43 and TR46, I replaced TR48 with a 2.7K resistor but despite this the voltage across the pot remained at around 30mV. I then lifted thebase of TR45 the voltage across the pot then increased, so I replaced TR45 with a BC548 but the voltage across the pot dropped back to 30mV.

In addition, I removed the base of TR49 and 50 to protect the output transistors. I also removed the diode 496. This means that the collector of TR47 is only connected to R496. There is just over 0.6V between the emitter and base, but the collector remains at around 0V

I am totally flummoxed by this amplifier.

[ms660]

What are the voltages on the pot connections WRT ground?

Lawrence.

[HECTOR63]

https://forum.beoworld.org/forums/p/...50.aspx#297950

Try this or maybe post a help wanted advice as it seems there are a lot of issues with these amps and there's a lot of informed threads there

[HECTOR63]

https://www.beoparts-shop.com/
This might be worth a visit for possible service kit and or further advices specific to the troublesome 3000.

[Radio Wrangler]

There is something seriously weird going on.

With the base lifted of TR45, and no current in TR45 R478 and R477 pulling up with your new 2.7k pulling down ought to try to put a lot of volts across the pot, limited by the Vbes of TR 44 TR49, TR47 (aided by TR50 boosting its current)

So the current should turn TR49 and TR50 hard on, pulling a fuse.

Now, 30mV across the pot is in the ball park of the c-e voltage of a saturated transistor... turned hard on.

BUT in this circuit, the drop across Vce is what powered the pot and it's associated resistors, and that powers Vbe of TR45 Now Vbe needs to be over 0.6v to turn the transistor on, so there needs to be more than 30mV across the pot.

Something stinks.

Let's strip things back to absolute basics. Pull TR48 and fit that 2.7k resistor. Pull TR43, tR46, TR47 and TR44.

You should just have a resistor network from 60v to ground and TR45.

Measure the voltages down the chain R47*(smudged 8 ?) R477 R480 R481 R482 and your new 2.7k

Check them again with TR45 base lifted.


I'm afraid there's a lot of circling around hoping to hit clues, but with lots of the amp disabled, there is less by default to have to worry about.

Solder splashes? R482 open or bad joint?

Don't put anything back until you've found why the minimal circuit doesn't do what it should.

David