I succeeded to repair my old Traynor 25B bass amp (which I asked and
received help for in this group a few months ago), however I now have
another problem with it.

Background: amp was producing a loud hum, originating in power stage. I
was recommended to replace some of the transistors, so I changed them
all. Problem solved. Here's the schematic btw:
http://img258.imageshack.us/img258/4232/ampkh8.png

Now at least it sounds as it should, but it quickly becomes devilishly
hot. It features two power transistors (BD911/912, formerly TIP100/105)
screwed on to an aluminium plate which is screwed on to the steel
chassis (applied thermal paste as well). I attached insulating silicone
washers to the transistors, and use nylon bushes to electrically
isolate the screws and nuts from both the chassis and the transistors,
and the ohm meter assures me that the plates of the transistors are not
in contact with the chassis (which is grounded) and thus not leaking.

After only 1½ minutes of operation, the chassis around the BD's are
hot enough to burn yourself on. Ten minutes later, the whole back and
bottom of the steel chassis is scorching hot, as well as the
transformer core! Something is clearly consuming a lot of power, but
the fuses do not blow. There is one 800mA slow blow fuse before the
transformer, and two 1.6A slow blow fuses after, in parallel. Just as
it's always been, as well as according to the schematic.

Voltage levels are fine, at least before and after the rectifier
diodes. The amp did not get near as hot as this when it was in working
condition last time.
What do I do?

On 15 Oct 2006 07:29:07 -0700, "
wrote:

After only 1½ minutes of operation, the chassis around the BD's are
hot enough to burn yourself on. Ten minutes later, the whole back and
bottom of the steel chassis is scorching hot, as well as the
transformer core!
Does "operation" mean just switched on with no audio?
I would check with no audio signal and possibly with the speaker disconnected
(or replaced with a larger resistor say 100R to 1K). The output voltage at the
speaker terminal should be zero.
What happens in this state? - still heat produced? If so then the bias for the
output pair could be wrong. Measure the voltage across one of the 0R5 resistors.
I don't remember the power supply circuit (if you gave it before) but you
mentioned rectifer diodes. Check that /all/ of them are ok since a failure here
could cause the transformere heating.

Geo

Geo wrote:
On 15 Oct 2006 07:29:07 -0700, "
wrote:

After only 1½ minutes of operation, the chassis around the BD's are
hot enough to burn yourself on. Ten minutes later, the whole back and
bottom of the steel chassis is scorching hot, as well as the
transformer core!
Does "operation" mean just switched on with no audio?
I would check with no audio signal and possibly with the speaker disconnected
(or replaced with a larger resistor say 100R to 1K). The output voltage at the
speaker terminal should be zero.
What happens in this state? - still heat produced? If so then the bias for the
output pair could be wrong. Measure the voltage across one of the 0R5 resistors.
I don't remember the power supply circuit (if you gave it before) but you
mentioned rectifer diodes. Check that /all/ of them are ok since a failure here
could cause the transformere heating.

Geo

0.5 volt across each of the two 0R5 resistors. There is no obvious
difference in heat generation if I don't feed it a signal, or if I turn
down the volume and gain dials (which only adjust the signal levels at
the preamp).
I could only a detect a very low DC voltage across the speaker (without
signal/volume). The battery of my digital multimeter ran down, so at
the moment I only have an analogue one at my disposal, and this sensed
a voltage in the neighbourhood of 0.1V. I'm no expert, but that seems
fine. Not feeling comfortable with disconnecting the speaker
completely, and I don't have a spare resistor on me at the moment (I
assume a 1/4W one would not suffice).

I will try to look for a schematic for the power supply, and I'll pull
out the diodes to measure them as well.

For the BD's to heat up like this with literally no input signal,
shouldn't they have to conduct a large DC current instead?

wrote in message
ps.com...

After only 1½ minutes of operation, the chassis around the BD's are
hot enough to burn yourself on. Ten minutes later, the whole back and
bottom of the steel chassis is scorching hot, as well as the
transformer core!

What is the idle current? You replaced Darlingtons with transistors. How did
you adjust the circuit? Do you have a scope or DMM?

On 15 Oct 2006 09:24:45 -0700, "
wrote:

0.5 volt across each of the two 0R5 resistors. There is no obvious
difference in heat generation if I don't feed it a signal, or if I turn
down the volume and gain dials (which only adjust the signal levels at
the preamp).
I could only a detect a very low DC voltage across the speaker (without
signal/volume). The battery of my digital multimeter ran down, so at
the moment I only have an analogue one at my disposal, and this sensed
a voltage in the neighbourhood of 0.1V. I'm no expert, but that seems
fine. Not feeling comfortable with disconnecting the speaker
completely, and I don't have a spare resistor on me at the moment (I
assume a 1/4W one would not suffice).
No problem.


For the BD's to heat up like this with literally no input signal,
shouldn't they have to conduct a large DC current instead?
Yes - if you have 0.5 volt across the resistor then there is 1 amp flowing all
the time and with 50 volts available you are dissipating 50 watts of heat.
I would guess (not a design engineer) that there should only be a small
quiescent current - just enough to get the transistors out of their non-linear
region. The schematic does not show any adjustment for the bias which seems to
be fixed by the voltage across the 3 1N4148 diodes. You could check this - maybe
about 2.1 volts?


Geo

On Sun, 15 Oct 2006 16:35:54 GMT, "Homer J Simpson" wrote:


You replaced Darlingtons with transistors.

Oops - I completely missed that - it would account for the high current since
the bias voltage would have been designed for the darlingtons.

I think the output pair will need to be changed back again..


Geo

wrote in message
ps.com...
I succeeded to repair my old Traynor 25B bass amp (which I asked and
received help for in this group a few months ago), however I now have
another problem with it.

Background: amp was producing a loud hum, originating in power stage. I
was recommended to replace some of the transistors, so I changed them
all. Problem solved. Here's the schematic btw:
http://img258.imageshack.us/img258/4232/ampkh8.png

Now at least it sounds as it should, but it quickly becomes devilishly
hot. It features two power transistors (BD911/912, formerly TIP100/105)
screwed on to an aluminium plate which is screwed on to the steel
chassis (applied thermal paste as well). I attached insulating silicone
washers to the transistors, and use nylon bushes to electrically
isolate the screws and nuts from both the chassis and the transistors,
and the ohm meter assures me that the plates of the transistors are not
in contact with the chassis (which is grounded) and thus not leaking.

After only 1½ minutes of operation, the chassis around the BD's are
hot enough to burn yourself on. Ten minutes later, the whole back and
bottom of the steel chassis is scorching hot, as well as the
transformer core! Something is clearly consuming a lot of power, but
the fuses do not blow. There is one 800mA slow blow fuse before the
transformer, and two 1.6A slow blow fuses after, in parallel. Just as
it's always been, as well as according to the schematic.

Voltage levels are fine, at least before and after the rectifier
diodes. The amp did not get near as hot as this when it was in working
condition last time.
What do I do?


When you say that you changed all the transistors, do you mean that all 8
transistors were replaced, or only those that were recommended to be changed?
The 0.5V across each of the 0.5ohm resistors means that the output transistors
are idling at 1 amp, which is WAAAYY too high. The idle current should be less
than 0.1 amp.
The BD911/912 transistors are not good substitutes for the TIP100/105
darlingtons. The BD911/912 units are not darlingtons. You need to get the
correct transistors for the amp. Since the DC voltage at the output is low, the
rest of the circuitry is probably OK.
Get the correct transistors and put them in. That will probably solve the
problem. Be sure to get a matched pair!!!! If you use unmatched transistors,
they will likely be unbalanced enough to give substantial DC at the speaker.
You don't want that to happen. The DC voltage at the speaker should be within a
few millivolts of 0V.
If you can't find a matched pair, you might buy a few of each type and match
them yourself. If you have to do that, ask here again and I can give you
directions or a link that will show you how. It only takes a DMM, a small power
supply or a few batteries, and a few resistors to do the matching.

Cheers!!!
--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in the
address)

Some days you're the dog, some days the hydrant.

Geo spake thus:

On 15 Oct 2006 09:24:45 -0700, "
wrote:

For the BD's to heat up like this with literally no input signal,
shouldn't they have to conduct a large DC current instead?

Yes - if you have 0.5 volt across the resistor then there is 1 amp flowing all
the time and with 50 volts available you are dissipating 50 watts of heat.

Say what? My arithmetic gives me 1/2 watt (0.5 V @ 1 A).


--
"In 1964 Barry Goldwater declared: 'Elect me president, and I
will bomb the cities of Vietnam, defoliate the jungles, herd the
population into concentration camps and turn the country into a
wasteland.' But Lyndon Johnson said: 'No! No! No! Don't you dare do
that. Let ME do it.'"

- Characterization (paraphrased) of the 1964 Goldwater/Johnson
presidential race by Professor Irwin Corey, "The World's Foremost
Authority."

David Nebenzahl wrote:
Geo spake thus:

On 15 Oct 2006 09:24:45 -0700, "

wrote:

For the BD's to heat up like this with literally no input signal,
shouldn't they have to conduct a large DC current instead?

Yes - if you have 0.5 volt across the resistor then there is 1 amp
flowing all
the time and with 50 volts available you are dissipating 50 watts
of heat.

Say what? My arithmetic gives me 1/2 watt (0.5 V @ 1 A).

Your arithmetic is correct for the resistors but that same 1 amp is
flowing throught the power transistors as well. Each of those has 25
volts (though I suspect it droops with that much load) so each
transistor drops 24.5 volts at 1 amp. 49 watts needs a pretty
significant heatsink. I would target the idle current at 20-50mA.

GG