I'd like to see at some point: (final tests)

1. Temperature feedback transistor attached to heatsink with thermal glue. Would you order from Farnell uk?

2. Quiescent current vs time. I'm not sure you got 0V or X mV across the emitter resister? Alternatively, set the scope to look at a low signal, at the +- crossover point and look for crossover distortion. Flat at the zero crossing when the amp is cold.

3. 1/2 to 3/4 power 20 min test with temperature of the output transistors.

If you have quiescent current, I think your done unless this thing removes it at idle like Les sort of suspects.

The output transistors SHOULD BE slightly warm idling for 20min to an hour.

 

I think i will have to wait for my resister to turn up to complete these tests

 

If you want to start on the left channel I suggest taking the following reading.
With no signal input and nothing connected to the output.
Test the voltages across these resistors.
R6 =
R7 =
R8 =
R15 =
R16 =
R24 =
R25 =
R26 =
R20 =
R21 =
R22 =
R23 =
R19 =

And.
The voltage on D3 anode with respect to T4
The voltage with respect to T4 on Q1 base =
The voltage with respect to T4 on Q22 base =

I think that you said that D4 was short circuit. If so then I think it is likely that the zener diode D5 is also faulty.
I think you component tester will give a good enough indication if it is OK. (I tried a 3.3 volt zener diode in my component tester and it showed about 3.04 volts. (The low reading is probably due to the low test current.)
I think you should replace D4 and/or D5 if they are faulty before doing the above tests. If you don't have these diodes you could do the above tests bot you may have to repeat them after replacing D4 and/or D5.

Les.

 

Yeah i am prety sure D4, D5, D6 and D7 are shot, i was going to replace them and remove the final outputs and leave the resister fuses in befor powering up and starting testing on the left side,

D4 and D6 are 1N4148
D5 and 7 are 400mW 3v3, i think i should be able to find replacements but on Farnell website the 400mW are no longer available, thay have a500mW instead, would these be ok?,

At present i have 2 of the old final outputs installed on the now working right side, should i leave them in or replace them for new to match the new ones i have for the left side ?

 

1. Temperature feedback transistor attached to heatsink with thermal glue. Would you order from Farnell uk?

I think there is some thermal tape that would work. It wasn't clear if the tape was double sided. Probably a better choice than a hard glue and definitely better than thermal grease.

 

I would approach things a bit differently, but to each his own.

In circuit measurement of resistors in the output stage. You now have another side to compare to now.
and the B-E diode tests.

Before applying power.

I also like the dim bulb methodif you don't have a variac. Signal applied, no load. Scope on output.

 

@KeepItSimpleStupid ,
When you say dim bulb method, are you refering to the 230v AC bulb in the input from the consumer outlet?
If you are i dont think this is needed as we have successfully managed to repair the right chanel without this, i think the resistor fuses work well enough, also by removing the final outputs there is less chance of shorting on i initial power up, once i replace the damaged diodes i will do the resister voltage check advised by @LesJones , then i will check for bias between Q7 collector and emitter, then follow both base emitter junctions to the final outputs, unless something looks wrong before i get to that point ,

Pete

 

I think @LesJones may be right with the output stage, untill the final outputs have enought bias to become active they stay dormant, this explains the lack of current across R64 and R73 with no input,
Or the bias regulator is down by 1.2v to forward bias the final outputs in which case should be 5.2vdc roughly

 

@KeepItSimpleStupid ,
When you say dim bulb method, are you refering to the 230v AC bulb in the input from the consumer outlet?
If you are i dont think this is needed as we have successfully managed to repair the right chanel without this, i think the resistor fuses work well enough, also by removing the final outputs there is less chance of shorting on i initial power up, once i replace the damaged diodes i will do the resister voltage check advised by @LesJones , then i will check for bias between Q7 collector and emitter, then follow both base emitter junctions to the final outputs, unless something looks wrong before i get to that point ,

Dim bulb, resistor one in the same, almost. Close enough: Hand grenades and horseshoes in when it counts.

if it's really really bad, your just doing a quick test. Is it supposed to be 100 ohms and it reads 500. it looks like you never replaced any resistors which is good, then why not B-E junctions only and/or check the same parts that failed in the other channel? It has been my EXPERIENCE, that the low valued resistors need to be checked.

You can't always remove the outputs.

You can't let the magic smoke out when there is no power, right?


Thermal tape suggestion: https://uk.farnell.com/multicomp-pr...e-300mm-x/dp/3267515?st=thermal adhesive tape

at least it says double sided.

 

Friends had an electronics business and one day I was there, there was a shopper whose only question was:

How may Watts is this? I thought it was funny.

 

Thermal tape suggestion: https://uk.farnell.com/multicomp-pr...e-300mm-x/dp/3267515?st=thermal adhesive tape

at least it says double sided

That looks to be a fairly good solution,
Thanks i may use that

 

I would approach things a bit differently, but to each his own

@KeepItSimpleStupid

How would you approach this ?


Also i have concerns about the cooling fan speed, it seems realy fast given the outputs are at idle, could this be down to the left chanel not being operational?

I did look at the fan control circuit, it looks like its controed by 2 ic's with 2 op amps and 2 transistors, ( forget what thet are , i dont have the schematic in front of me), maybe it will sort itself out when the left chanel is running ok,
Voltage on pins 2 and 5 (i think it was) relative to 0v apeer to be the same so i cant work out which chanel is powering the cooling fan ,

Pete

 

It appears to use Q42 and Q6 as temperature sensors. They may connect at T15 and T16.

Are Thermal L or Thermal R LED's lit?

IC3-A and IC3-B are comparitors

D39 and D4 act as OR Gates.

VERF(R) is probably VREF(R) etc.

CONTROL-L and CONTROL-R might actually reduce the signal levels on a high temperature.

The first few steps of troublestooting usually are:

Symptoms reported by customer
Verify symptoms
Visual inspection
Examples: Rattles, cat hair, roaches, sandwhich in VCR, wrong voltage selected
Me: The high voltage regulator (tube) fell out of socket in transport.
Check power supply voltage and ripple. (+15, -15, +24, -24)
The -15V ripple can cause all sorts of issues.

The test points (T11, T12, T15, T16) are useful and TH-L and TH-R
IC3a-1 and IC3b-7
Base of Q50

I think the fan is either on or off.

But anyway:
Visual inspection.
Are any of the hi temp LEDs on?
Ripple and magnitude of the supply voltages?

You have both bisecting and comparison.
e.g. Which side is causing the fan to come on?

Aside:
At work we had 6 or so of a certain model vacuum guage and about every 5 years, I would have to replace some capacitors. That was OK with me. I looked at the power supply circuit recently and it could have used some re-design. It was always the negative supply ripple.

Also i have concerns about the cooling fan speed, it seems realy fast given the outputs are at idle, could this be down to the left chanel not being operational?

I don't think they are related. Are the high temperature LED's lit?

 

I agree with kiss's post # 373. Measure the following voltages with respect to T26 (0VD). +15 volt rail, -15 volt rail, T11, T12, D39 cathode, D40 cathode. My understanding of the schematic suggests that at normal temperature the fan will be running with about 9 volts (24 -15 volts) across it's terminals. At high temperatures it will have about 39 volts (24 + 15 volts)across it's terminals. These two voltages will be less than my estimated values due to the voltage drop across R131 which I can't calculate as I don't know how much current the fan takes. When the voltage at the cathode of D39 or D40 is negative it will be requesting high fan speed. D39 and D40 form a negative OR gate so either one can request high fan speed. (D39 for left channel, D40 for right channel).
Decide which fault you are going to work on first and fix that before starting on the other fault.

Les.

 

@KeepItSimpleStupid

No i have no high temp or warning LED'S on at all,

Edit, while i have been sat here troubleshooting i now have the L thermal LED lit, this has never happened before, i have no load or signal on at present,
The only thing i noticed that is realy hot is R131 and R 123,

Here are the results you requested @LesJones
I took the + and - 15v readings from ICB:3 pins 8 and 4,
Relative to T26ovd
+ rail = 14.773vdc -- rail = 15.064vdc
T11= 0.563vdc T12 = 0.558vdc
D39 cathode = 14.78vdc
D40 cathode = --13.69vdc

 

That being the case, and if you did do the inspection/ripple measurements, then
IC3a-1 to ground
IC3b-7 to ground

What side is causing the issue.

The voltage across the fan (for Les) - indication of speed,
Voltage across R123 (for Les) (fan current draw at this speed)

Does the fan have any useful info on it, particularly voltage? It's probably a 48V DC fan.

That should give a quick idea of what's going on.

IC numbering:
There is usually some sort of mark at the lower left hand corner which marks pin #1.
Other pin #'s increment counter-clockwise.

 

Measure the voltage between the fan terminals.

Les.

 

Could this be because i currently dont have Q7 installed on the left chanel ?

 

Measure the voltage between the fan terminals.

Les.

Voltage at the fan connector is 39.25vdc

 

@KeepItSimpleStupid
Here is a photo of the fan , and sorry no i did not do the ac ripple tests, not 100% sure i know how to