Help with blowing transistor NAD T752 receiver

[rddube]

Originally Posted by eccerr0r

H Is the relative position of the R522 potentiometer close to the position of the R545 potentiometer (don't touch either for now, just look at position). Perhaps R522 is turned all the way down for 1KΩ and hence the voltages are so close.

Forgot to answer this question...yes the position of the the potentiometers is close as seen in the photo (the one on the right is R522).

[eccerr0r]

Okay that is a clue that something's still wrong with the R522 channel unlike the R545 channel. Spread is too low.

Next I'd be curious what the voltages are on R513, hopefully both are way negative... If they are, Q504 is probably bad. If they aren't, R513 is bad.

... but still not quite sure why this would cause final roasting... more data needed...

[rddube]

Ok, so here are my measurements with 1K resistors from Base to Emitter on Q509 Q510 and Q521 and Q522:

Q506 C to gnd +1.51V
E to gnd -1.451V

Q518 C to gnd +-4.41V
E to gnd -7.92V

Q530 C to gnd -3.97V
E to gnd -7.44V

Looking at board such as the photo I uploaded:
R5143 from left to gnd -6.5V
from right to gnd -5.82V

R5146 from left to gnd -4.4V
from right to gnd -3.2V

R5143 from left to gnd -3.2V
from right to gnd -3.1V

Q504 B to gnd -6.2V

I double checked again every measurement.

[eccerr0r]

Check your component designators and signs again, you have R5143 listed twice and you have +-4.41V for Q518-C. Assuming you meant -4.41V, the C-E drops all look reasonable now though the "second" R5143 spread looks too close. Again something doesn't look right, when the spread is close the transistors should be safe from destruction.

Now Q504-B is way too high, it should be -65+1.4 = about -63V, but the voltages downstream do not seem to correspond to Q504 being turned on...

[rddube]

Originally Posted by eccerr0r

Check your component designators and signs again, you have R5143 listed twice and you have +-4.41V for Q518-C. Assuming you meant -4.41V, the C-E drops all look reasonable now though the "second" R5143 spread looks too close. Again something doesn't look right, when the spread is close the transistors should be safe from destruction.

Now Q504-B is way too high, it should be -65+1.4 = about -63V, but the voltages downstream do not seem to correspond to Q504 being turned on...

Hello Eccerr0r,

Sorry was trying to go too fast, here are the corrections to my voltages that I checked again:


Q506 C to gnd +1.51V
E to gnd -1.451V

Q518 C to gnd -4.41V
E to gnd -7.92V

Q530 C to gnd -3.97V
E to gnd -7.44V

Looking at board such as the photo I uploaded:
R5143 from left to gnd -6.5V
from right to gnd -5.82V

R5146 from left to gnd -4.4V
from right to gnd -3.2V

R5149 from left to gnd -3.2V
from right to gnd -3.1V

Q504 B to gnd -6.2V

Q504B is -6.2V not -62V.

[rddube]

Hold on gentlemen, I removed the board and checked some components only to find that on the right channel, I have a partial short as the 1K resistors I installed measure .5K instead of 1K. I have to trace it down now.

Don't spend too much time analyzing the left channel (Q510/Q509) as I have to get rid of this short.

Will be back with new measurements once I find it. Sorry for the inconvenience.

[rddube]

Ok, so I think I finally found the problem in that the .5K measured on the 1K resistors seems to be ok - explanation: right channel was giving .5K on those resistors while the left channel was giving me 1K...searched everywhere only to find out that the .22K double resistor on the left channel R527 was open! Fortunately I had a spare one and replaced it and now both the right and left channel read .5K on the 1K resistors when installed.

Everything seems to balance well, but now R520 and R544 measure 196 ohms, while R573 (the center and good channel) measures 217 ohms, not too sure why but is this normal with 1K resistors instead of finals?

[eccerr0r]

Whenever you measure a resistor, it needs to be measured out of circuit. Too many side paths will affect your measurement. I kind of expected that emitter resistor to affect your reading and there was never a "short" on the right channel.

So it looks like the fried transistors fried the resistor too.

and yes, the skewed in-circuit measurement of the resistors with that hack added would affect reading, my calculation resulted in around 198Ω which matches up quite well within about 1% of 196Ω.

[petehall347]

just so you know those double resistors also known as emitter resistors in this case are 0.22 ohms . the k is tolerance .

[rddube]

Originally Posted by petehall347

just so you know those double resistors also known as emitter resistors in this case are 0.22 ohms . the k is tolerance .

Oh sorry, my bad, you are right they are 0.22 ohms, and that is what I installed!

[rddube]

Originally Posted by eccerr0r

Whenever you measure a resistor, it needs to be measured out of circuit. Too many side paths will affect your measurement. I kind of expected that emitter resistor to affect your reading and there was never a "short" on the right channel.

So it looks like the fried transistors fried the resistor too.

and yes, the skewed in-circuit measurement of the resistors with that hack added would affect reading, my calculation resulted in around 198Ω which matches up quite well within about 1% of 196Ω.

Ok, thank you for the explanation, I'm learning a lot while doing this thanks to you! So I think we are in business, will take measures again and report back - just got notice that the finals have arrived in Canada, so probably next week will be here. Thanks again!

[rddube]

So here are my latest measurements with the .22 ohm resistor changed and after having removed and checked Q506, Q507, Q508, Q562 and Q563 to be good:

Q506 C to gnd +0.465V
E to gnd +0.389V

Q518 C to gnd +1.7V
E to gnd -1.57V

Q530 C to gnd +1.785V
E to gnd -1.63V

Looking at board such as the photo I uploaded:
R5143 from left to gnd +0.365V
from right to gnd +0.386V

R5146 from left to gnd -1.04V
from right to gnd -1.05V

R5149 from left to gnd -1.052V
from right to gnd -1.053V

Q504 B to gnd -62.7V

[rddube]

Eccer0r and Petehall347, do these readings look ok to you? The only one that is close to 600mv is the left channel (Q506), in fact it is below, but it is positive on C and on E, and also on each side of the resistor, whereas the other two channels have positive and negative on each side but are more in the 1.7V area. Not sure how to interpret this, is there something wrong with the negative side of the left channel? Many thanks for your input!

[eccerr0r]

As in previous messages still would like to know the voltage at Q504-E (or voltage at R513) and what happens to the voltage across Q506 as you change R522. Yes the spread is wrong compared to the other channels, but I'm at a loss as why all the transistors are good except that it burns the finals. Again the closer the voltages are together on Q506, the safer for the downstream transistors, but it produces odd-order harmonics in your music when playing which may be unacceptable to your ears.

[rddube]

Originally Posted by eccerr0r

As in previous messages still would like to know the voltage at Q504-E (or voltage at R513) and what happens to the voltage across Q506 as you change R522. Yes the spread is wrong compared to the other channels, but I'm at a loss as why all the transistors are good except that it burns the finals. Again the closer the voltages are together on Q506, the safer for the downstream transistors, but it produces odd-order harmonics in your music when playing which may be unacceptable to your ears.

Sorry Eccerr0r, had missed your question about this. So the voltage at R513 is -61.25 and at Q504-E it is -60.86V.

Will do measurements across Q506 as I change R522 and report back in a few minutes

[rddube]

How weird can you get, as soon as I attach my probes across Q506 and try and start it, it gives me no voltage reading and goes immediately into protection mode (red light). If I detach my probes, it starts up with green light ??? I tried to take a reading across Q518 and got -3.35V and the amp started normally (green light)...what's going on?

[eccerr0r]

A good test meter should not cause going into protection mode. Maybe you have a flaky Q506 or shorted your probes to something again?

Perhaps a broken R518 would also cause the device to go into protection.

BTW that hole near C572/R528 looks very suspicious, if there is a screw there, is it shorting C572 with that jumper J48?

[rddube]

My test meter is a Fluke 87V, so I think it is ok. Tried it with another one I have a Uni-T and it does the same thing - as soon as I connect both probes from the meter, it goes into protection mode. R518 measures 3.275K so I think it is ok.

The hole near C572 is the hole to insert the screw that holds the transistors to the heat sink, it is a pass throught hole.

So the only thing left, is a flaky Q506 because I didn't short my probes again. When I do the same type of reading on Q518 it doesn't do this - it gives a reading and stays on green light.

So I guess I'll replace Q506 and try it again. Will report back later.

[rddube]

While I remove the board, I'll remove R518 and check it out of circuit just to make sure.

[eccerr0r]

mainly inspect for broken traces, as long as there is some conductance that works acceptably from C-B of Q506, it should be acceptable and not cause finals to fry.