Hi,

I've just successfully troubleshooted (touch wood) my first power amp,
I've always avoided them in the past.

I need:

1x 2sc2565 (or 2sc5200)
1x 2sa1095 (or 2sa1295)
1X 2sc2238

Donberg do all three, but cost loads.
RS do the last one only.
Rapid do the first two only.
Telepart do the last one slightly cheaper than RS.

Looking for any alternatives. Idealy somewhere with reasonable
handling charges which does all three.

Thanks,
Alex

"Alex Bird" wrote in message
oups.com...
Hi,

I've just successfully troubleshooted (touch wood) my first power amp,
I've always avoided them in the past.

I need:

1x 2sc2565 (or 2sc5200)
1x 2sa1095 (or 2sa1295)
1X 2sc2238

Donberg do all three, but cost loads.
RS do the last one only.
Rapid do the first two only.
Telepart do the last one slightly cheaper than RS.

Looking for any alternatives. Idealy somewhere with reasonable
handling charges which does all three.

Thanks,
Alex


www.grandata.co.uk have all three. Good company. Reasonably priced. Have
used them for years. Take phone orders by credit card, and despatch same
day. Have trade counter in Wembley where they're based.

If your output TRs have failed, make sure that the low value emitter
resistors have not gone open circuit also

Luck. Arfa

Arfa Daily wrote:

www.grandata.co.uk have all three. Good company.

Great, cheap too!

I've just been reading a few threads about counterfeit transistors. Is
this a problem in the UK?

If your output TRs have failed, make sure that the low value emitter
resistors have not gone open circuit also

I've tidied it away for now, but would that be one 'designed' to blow,
about 1cm off the board ? I can't remember the actual circuit.
Runs fine with transistors from the other channel. I've a feeling I'll
need to adjust the bias, any FAQs on that ? They didn't seem to get
warm when idling, does this mean bias is 'ok' ?

Thanks!
Alex

"Alex Bird" wrote in message
oups.com...
Arfa Daily wrote:

www.grandata.co.uk have all three. Good company.

Great, cheap too!

I've just been reading a few threads about counterfeit transistors. Is
this a problem in the UK?

If your output TRs have failed, make sure that the low value emitter
resistors have not gone open circuit also

I've tidied it away for now, but would that be one 'designed' to blow,
about 1cm off the board ? I can't remember the actual circuit.
Runs fine with transistors from the other channel. I've a feeling I'll
need to adjust the bias, any FAQs on that ? They didn't seem to get
warm when idling, does this mean bias is 'ok' ?

Thanks!
Alex


I personally, have never had a problem here with counterfeits, as far as I
know, although I did have a problem just this week, with some 7915's. which
came from a very large reputable supplier that I deal with regularly. I had
ordered in 5 to replenish my stock, and one of them was needed for a repair
in hand. When I fitted the new one, it didn't work, nor did the next, or the
next ... In fact none of the 5 had output, despite a good voltage level
going in. I finally fitted one from another source, and that was fine. All
faulty ? Marked up wrong ? Counterfeits ? I don't know. I wouldn't have
thought that this particular company would have bought from anyone but the
manufacturers, or their agents, but you just never know.

The emitter resistors that I referred to, are usually white ceramic square
bodied types, rated about 4 watts, and may have a value anywhere from about
0.01 ohms to 0.47 ohms. It's very common for them to fail when the output
transistors do, but if the amp works with transistors from the other
channel, they are probably ok. I say probably, because even though they are
usually very stable wirewound types, I have known them to go up in value, as
a result of the overload imposed on them, by short circuit output
transistors.

As far as the bias goes, I would not expect there to be much difference
between the original failed transistors, and the ones that you have cross
substituted from the other channel, as they will probably have originally
all come from the same production batch. However, when you fit the new ones,
they may have significantly different characteristics, and require the bias
to be set. In general, if the heatsink temperatures remain about the same to
the touch between the channel with original transistors in, and the new
transistors, after the amp has been left to idle for 10 mins, then assume
the bias is ok, and leave it alone. If the renewed channel is significantly
warmer, then the bias needs turning down, and if cooler, turning up.

Rule of thumb value for the quiescent current in an ' average ' domestic
stereo amp, is about 20mA, and this is a good starting point, if you don't
know exactly what it should be. Some amps are equipped with linked
testpoints, which can be used to break an ammeter into the circuit to
measure this current directly. Others, however, use an indirect measurement
method, where the voltage across a resistor is measured, usually with a
millivolt meter, or 'scope set to DC input. You can always compare with the
good channel, for a valid setting.

A word of caution before altering the bias. Because the bias presets are
generally not altered during their lifetime, they can get ' touchy '. Before
adjusting, mark the current position, then add a few drops of switch cleaner
to the pot. Making sure that the amp is off, give the pot a vigourous scrub
back and forth, then reset back to the original position. If you don't do
this, and the pot goes open circuit as you move it, the output stage can go
into thermal runaway, and ultimate meltdown. Also, only adjust the pots in
VERY small increments, as some are really quite violent in their operation,
and the difference between a virtually cut off stage, and one pulling 2
amps, might only be a few degrees of rotation.

Hope all this is helpful to you

Arfa

Arfa Daily wrote:

I personally, have never had a problem here with counterfeits, as far as I
know, although I did have a problem just this week, with some 7915's.

One guy had opened the cases of some and found the silicon die to be a
quarter of the size of the proper part, they didn't last very long!


The emitter resistors that I referred to, are usually white ceramic square
bodied types, rated about 4 watts,

Ah, I think I knew at the back of my mind I should have checked them.
The night was wearing on, I got the amp for nothing, and I skipped
ahead to the substitution.

I'll check it out.

As far as the bias goes, I would not expect there to be much difference
between the original failed transistors,
snip However, when you fit the new ones,
they may have significantly different characteristics, and require the bias
to be set.

Especially as I cannot get the original part, I'm still not sure what
the best replacement pair is. The part numbers in
my original post turn out not even to be a pair!
If I could find a data sheet for the originals...
I've emailed grandata anyway to see if they can suggest.

In general, if the heatsink temperatures remain about the same to
the touch between the channel with original transistors in,
snip
Hope all this is helpful to you

Yes, much appreciated. I know I'm not the first person to open this
amp, so I'll probably be checking out both channels.

Could I feed a tone in and tune it looking at a spectrum display? The
one I have is basic but fine for audio.

Alex

"Alex Bird" wrote in message
oups.com...
Arfa Daily wrote:

I personally, have never had a problem here with counterfeits, as far as
I
know, although I did have a problem just this week, with some 7915's.

One guy had opened the cases of some and found the silicon die to be a
quarter of the size of the proper part, they didn't last very long!


The emitter resistors that I referred to, are usually white ceramic
square
bodied types, rated about 4 watts,

Ah, I think I knew at the back of my mind I should have checked them.
The night was wearing on, I got the amp for nothing, and I skipped
ahead to the substitution.

I'll check it out.

As far as the bias goes, I would not expect there to be much difference
between the original failed transistors,
snip However, when you fit the new ones,
they may have significantly different characteristics, and require the
bias
to be set.

Especially as I cannot get the original part, I'm still not sure what
the best replacement pair is. The part numbers in
my original post turn out not even to be a pair!
If I could find a data sheet for the originals...
I've emailed grandata anyway to see if they can suggest.

In general, if the heatsink temperatures remain about the same to
the touch between the channel with original transistors in,
snip
Hope all this is helpful to you

Yes, much appreciated. I know I'm not the first person to open this
amp, so I'll probably be checking out both channels.

Could I feed a tone in and tune it looking at a spectrum display? The
one I have is basic but fine for audio.

Alex


Yes, you could do that. Any serious distortion would soon show up as
harmonics. Providing the generator produces a reasonable sine wave, you
could also look at the output on a 'scope, paying particular attention to
the zero crossing, where you will see the bias setting having most effect.

Arfa

Arfa Daily wrote:
However, when you fit the new ones,
they may have significantly different characteristics, and require the bias
to be set.

Yes indeed. I ordered 2SC3264 and 2SA1295 as replacements.

With the amp on full mains voltage the quiescent current is heading for
300mA on that channel with no sign of stopping, I turn it off at this
point.

It turns out the adjustment I thought was bias only affects the DC
offset, there does not appear to be any way to adjust the bias.

It's a TEAC amp of a few years ago, no info at all online, not a hit.
A-919AV.

Briefly: Some sort of +/- drive comes from the input board, where the
DC offset pot is, this comes to the output board, to a circuit of 6
small transistors with a smattering of diodes and one transistor
attached to the heatsink, these feed the driver transistors, which then
feed the output transistors.

I think I should be looking for some diodes and resistors setting the
bias but I'm not sure where to look in all that!

I suspect the thermally coupled transistor is biasing it all, there are
3 parallel resistors connecting one of it's legs to base, obviously
some production trimming. I'm too tired to trace the circuit properly
just now, but does this sound reasonable?

I've a feeling this isn't going to be as simple as I thought...

Thanks again for all your help,
Alex

Alex Bird wrote:
Arfa Daily wrote:
However, when you fit the new ones,
they may have significantly different characteristics, and require the bias
to be set.

Yes indeed. I ordered 2SC3264 and 2SA1295 as replacements.

Oh s**t, I've just realised Grandata have the originals, no idea why I
didn't find them before.

Still, any ideas on making the replacements work and saving a few quid?
This is not for real hi-fi use so channel matching doesn't matter.

ALex

"Alex Bird" wrote in message
oups.com...

Alex Bird wrote:
Arfa Daily wrote:
However, when you fit the new ones,
they may have significantly different characteristics, and require the
bias
to be set.

Yes indeed. I ordered 2SC3264 and 2SA1295 as replacements.

Oh s**t, I've just realised Grandata have the originals, no idea why I
didn't find them before.

Still, any ideas on making the replacements work and saving a few quid?
This is not for real hi-fi use so channel matching doesn't matter.

ALex


I'm assuming that this amp has a standard linear supply, and a proper on /
off switch ? If it does, you can bring it up on a variac, if you have one,
or if not, you can try a light bulb in series with the mains. If it is
electronically switched, then neither of these will do you any good. If you
can get it up with much reduced rails, you should be able to fault find it
safely with a meter. As a matter of interest, where are you measuring the
quiescent current, and how does it compare to the other channel ?

Another thing you can do, is to take comparitive resistance checks between
the good channel, and the bad one. Only use an analogue meter for this, as a
digital one will just mislead you. Clip the negative probe to chassis, then
take readings from the same point on each channel. You will usually quite
quickly come up with some significant difference.

If output transistors was the original problem, check carefully for leaky
drivers, open circuit or ' gone high ' coupling resistors, driver emitter
resistors etc.

I will be away for a couple of weeks from Monday

Arfa

Arfa Daily wrote:

I'm assuming that this amp has a standard linear supply, and a proper on /
off switch ? If it does, you can bring it up on a variac, if you have one,
or if not, you can try a light bulb in series with the mains.

I've been using a light bulb.

As a matter of interest, where are you measuring the
quiescent current, and how does it compare to the other channel ?

Across the emitter resistors, there are test points there.

Another thing you can do, is to take comparitive resistance checks between
the good channel, and the bad one. Only use an analogue meter for this, as a
digital one will just mislead you. Clip the negative probe to chassis, then
take readings from the same point on each channel. You will usually quite
quickly come up with some significant difference.

If output transistors was the original problem, check carefully for leaky
drivers, open circuit or ' gone high ' coupling resistors, driver emitter
resistors etc.

The thing is, the transistors from the working channel are still
swapped, I'm testing the replacement transistors in the undamaged
channel.

Sadly the replacements have just popped!

I put a couple of pots in where there were wire links across resistors
in the driver circuit, this forming a voltage divider for a 'variable
diode' biasing arrangement - I think. One wire link was cut on each
channel, a very basic trimming arrangement.

This let me get the current down to 60mAish. Then I checked the other
channel, this was at about 50mV, so I figured that was the normal
operating point for the amp, and changed it, it settled down, I heard a
PHUT, and smoke was issuing, isolated it. One transistor shorted, the
other has a hole in the top!

Did you see what I did there? I got mV and mA mixed up. Curses.
Looks like I'll be ordering the original parts anyway. And checking
for further damage, though I think I stopped it in time, emitter
voltage was going up but not very high when I cut the power.

I'll put this down to experience then,

Alex

"Alex Bird" wrote in message
oups.com...
Arfa Daily wrote:

I'm assuming that this amp has a standard linear supply, and a proper on
/
off switch ? If it does, you can bring it up on a variac, if you have
one,
or if not, you can try a light bulb in series with the mains.

I've been using a light bulb.

As a matter of interest, where are you measuring the
quiescent current, and how does it compare to the other channel ?

Across the emitter resistors, there are test points there.

Another thing you can do, is to take comparitive resistance checks
between
the good channel, and the bad one. Only use an analogue meter for this,
as a
digital one will just mislead you. Clip the negative probe to chassis,
then
take readings from the same point on each channel. You will usually quite
quickly come up with some significant difference.

If output transistors was the original problem, check carefully for leaky
drivers, open circuit or ' gone high ' coupling resistors, driver emitter
resistors etc.

The thing is, the transistors from the working channel are still
swapped, I'm testing the replacement transistors in the undamaged
channel.

Sadly the replacements have just popped!

I put a couple of pots in where there were wire links across resistors
in the driver circuit, this forming a voltage divider for a 'variable
diode' biasing arrangement - I think. One wire link was cut on each
channel, a very basic trimming arrangement.

This let me get the current down to 60mAish. Then I checked the other
channel, this was at about 50mV, so I figured that was the normal
operating point for the amp, and changed it, it settled down, I heard a
PHUT, and smoke was issuing, isolated it. One transistor shorted, the
other has a hole in the top!

Did you see what I did there? I got mV and mA mixed up. Curses.
Looks like I'll be ordering the original parts anyway. And checking
for further damage, though I think I stopped it in time, emitter
voltage was going up but not very high when I cut the power.

I'll put this down to experience then,

Alex


Happens to the best of us. Keep me posted. Not around now though for a
couple of weeks.

Arfa