lab gruppen 1600 faulty channel B

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello guys,

I had fail on channel 2 of my lab gruppen amplifier. When open it i found a fuse burned and one of the amplifier power transistors MJ21193 in short circuit. i have replace the transistor but now when power on the amplifier on channel 2 there is a sound distortion and after a few seconds channel 2 enter in protection mode. I have try and replace all the 5 MJ21193 and also 5x MJ21194 but i still have the same problem. Any idea what may cause this distortion and protection mode?
Thanks in advance
 
Administrator
Joined 2007
Paid Member
If one of the output transistors was shorted then it is likely that the driver transistor/s could also be damaged, and possibly also some low value resistors around that area.

Ideally we would need to see circuit details to advise better.

Entering protection mode could mean that there is either a DC offset present at the output, or that the output stage is passing to much current.
 
If one of the output transistors was shorted then it is likely that the driver transistor/s could also be damaged, and possibly also some low value resistors around that area.

Ideally we would need to see circuit details to advise better.

Entering protection mode could mean that there is either a DC offset present at the output, or that the output stage is passing to much current.

Thanks for your reply.
Driver transistors i assume that is the MJ21194. i have also exchange all 5 of those.

Attached you can see the circuit diagram in case you can advice further. faulty one was the Q43

Thanks again for your time
 

Attachments

  • lab1600.jpg
    lab1600.jpg
    646.3 KB · Views: 180
Administrator
Joined 2007
Paid Member
The driver transistors are the ones before the output transistors, so that is Q31 and 32 and Q29 and 30.

All the emitter resistors on the output transistors should be checked, these will typically be in the 0.1 to 0.22 range.

Also check the resistors that connect to the driver transistors directly.

The DC offset at the amplifier output should be close to zero volts.

I would also recommend that you use a 'bulb tester' to limit damaging currents while working on this and that you also temporarily reduce the bias to zero by either turning preset TP1 to give minimum bias or you could temporarily unsolder one end of R25. That would force the bias current to zero. The amp would still operate normally under this condition.
 
Thanks once again for your valuable information.
I am an electronics engineer but i have never work on an amplifiers.
I will try to find some instruction about how to adjust bias.
By unsolder R25 what should i have? do you mean that the driver transistors maybe good and the problem can be solved by adjusting bias from TP1?
Thanks
 
Administrator
Joined 2007
Paid Member
Unsoldering R25 will allow Q28 to turn fully on. That will remove bias from the output stage under no signal conditions. Zero bias will increase distortion a little but it should not be enough to be very noticeable except at extremely low volume.

The final bias value is set by monitoring the voltage across any of the low value emitter resistors and altering the preset to give the specified current (work out the voltage expected from ohms law). A PA amp like this is probably deliberately under biased to run cool, perhaps with as little as 15 milliamps flowing per pair of output transistors.

The full service manual should detail the procedure and value but 15ma would be a very safe value (and you can always compare with the good channel).
 
So, in case that the amplifier did not goes into protection mode by unsolder R25 that means that problem can be solved by adjusting the bias, correct?
If i understand correctly, i have to measure in one of the emitters resistor in parallel the voltage while the amplifier is working to adjust the bias?
Thanks
 
Administrator
Joined 2007
Paid Member
I doubt your problem is simply bias related, however forcing the bias to zero is a good starting point and safeguard.

It wont be a bias adjustment problem tripping the protection.

Measuring the voltage across an emitter resistor is the standard method of deducing the current flowing.

Your first checks have to be:

1/ Force the bias to zero.
2/ Measure the DC voltage at the output. Should be zero.
3/ Check that there really is close to zero current flowing in those emitter resistors.

And don't forget to check and confirm that the supply voltages are correct. That should really be the first thing to check with any fault-finding.
 
once again thanks for your answer.
Regarding supply voltages as i saw the supply generates channel A & B in parallel, so if it was a supply problem then both outputs were getting into protection. Channel A is working perfectly. I will measure across R83 resistor the voltage on channel B and i will compare it with the same resistor on channel A.
 
I doubt your problem is simply bias related, however forcing the bias to zero is a good starting point and safeguard.

It wont be a bias adjustment problem tripping the protection.

Measuring the voltage across an emitter resistor is the standard method of deducing the current flowing.

Your first checks have to be:

1/ Force the bias to zero.
2/ Measure the DC voltage at the output. Should be zero.
3/ Check that there really is close to zero current flowing in those emitter resistors.

And don't forget to check and confirm that the supply voltages are correct. That should really be the first thing to check with any fault-finding.

Finally i found one of the drivers transistors that was already replaced and was new to be the problem. It was new and measured ok, it was the same exactly type/model as the original one but for some reason it was causing the problem. now the amplifier is working but i have to find a way to adjust the bias. By connecting my multimeter to the emiter resistor in parallel i measured 0volts whatever i change the TP1. i must did something wrong i assume.
 
If it stil isn’t working you haven’t found all the problems yet. When push pull stages go haywire, it often takes out other parts with no visible evidence. Measure the vbe of the outputs, drivers and pre drivers and look for something that isn’t around 0.5 to 0.6 volts. Measure the whole vbe bias stack - for this type of output stage it would be near 3.45 volts total. Make sure this voltage adjusts as you turn TP1. You really need a light bulb if you haven’t been using one.

When servicing amps or building new ones I usually test the bias stack in situ before even powering up the amp. It involves injecting a 10mA current source across the vbe multiplier (powered by a 9 volt battery), checking for functionality, and pre-setting for about 3.2 volts for an EF3 stage like this one.

As for the brand new faulty driver transistor - it was probably a fake. Is it an MJE15033 or 5, a 1302, or another 21194? Double check you component sources before proceeding.
 
If it stil isn’t working you haven’t found all the problems yet. When push pull stages go haywire, it often takes out other parts with no visible evidence. Measure the vbe of the outputs, drivers and pre drivers and look for something that isn’t around 0.5 to 0.6 volts. Measure the whole vbe bias stack - for this type of output stage it would be near 3.45 volts total. Make sure this voltage adjusts as you turn TP1. You really need a light bulb if you haven’t been using one.

When servicing amps or building new ones I usually test the bias stack in situ before even powering up the amp. It involves injecting a 10mA current source across the vbe multiplier (powered by a 9 volt battery), checking for functionality, and pre-setting for about 3.2 volts for an EF3 stage like this one.

As for the brand new faulty driver transistor - it was probably a fake. Is it an MJE15033 or 5, a 1302, or another 21194? Double check you component sources before proceeding.

Thanks for your reply.
Amplifier is working now but before i found the problem i have change TP1 position so i assume that i have to set the bias correctly now. By ear i don't see any difference between the 2 channels but is better to align it again. The problem is that is the first time that i have open a power amplifier so i am newbie and don't know what exactly to do. I am an electronics engineer but i am on satellite fields and don't have any experience with amplifiers.

what i did i have connect my multi meter in parallel of the resistor (R83 in my case) that is connected to the emitter in one of the MJ21193 transistors. I did not manage to measure any voltage. my questions are:
1. the amplifier must be in lower gain control or do i have to increase the gain knobs?
2. do i have to connect a source to the input of the amplifier to give input signal?
3. do i have to connect he speakers while measuring?

thanks
 
Bias should be adjusted with no speaker and no signal. If there is a small DC offset, current will flow thru the voice coil. A few milliamperes and it won’t hurt anything, but the bias will be off by that much if you set it with that current flowing in the speaker. Gain setting shouldn’t matter, but turned all the way down will minimize noise.
 
5 millivolts across the emitter resistor. Light bulb goes in series with the AC mains power. Must be an old incandescent. It lights up if there is a short, goes dim if the current draw is low like it should be. If you were not using one thank your lucky stars you haven’t burnt your house down yet. Or gotten flying hot pieces of power transistor embedded in your forehead.
 
i like your comment about the light bulb. this is a method that i was using long long time ago while repairing old tvs :) i didn't thought that is still a method that may be in use. So you mean that by adjusting the bias there is a chance to burn the power transistors?
actually there are fuses in each channel so if there is a short i power transistors then the fuse will be burned (after the transistor dead i of course)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.