Help with blowing transistor NAD T752 receiver

[rddube]

Originally Posted by R_J

J504B connects to V+ and V- of J781B, via JMP58~61, it will have the full +_ supplies on it. (see page 17) The jack could be either just a test point or it is for a different model.

Wow you have an hawk's eye (or lots of experience ) to have seen that. So is 63V + and - pretty much ok?

I've been testing what Petehall mentionned with my osciloscope a 10khz sine wave along the path of the signal - center and right channels are perfect from the transistor to the ouput coil, but left channel the sine wave gets fairly muddy...but I don't have the transistors in there as they were short and am waiting for the new ones.

[rddube]

Originally Posted by petehall347

ok when transistors short it can take other things with it . check everything .
also if replacing complementary transistors replace as pairs . like drivers and output pairs .
aso be sure check emitter resistors and current limiter resistors .
you could have a bad bias pot or bad solder connection in the biasing circuit .
also there may be high frequency oscillation happening that can only be seen on your scope if you have one .

Could you point me in the right direction by telling me an example of a emitter resistor and a current limiter resistor (part of schematic attached). Also, how do I check if there is high frequency oscillation (I do have a basic scope). Many thanks!

[Per Hansson]

Originally Posted by lotas

Maybe the transistors are not original (remarked) and they will not match their parameters, the Chinese love to do this

Originally Posted by rddube

Any suggestions on where I should look, what component should I test? Please, anything?

I would take your current destroyed parts and open them up:
You have one original from the unit and the one you bought.
Open them both and see if the die is the same physical size.
If it is not you have your probable cause why it failed.
(Lots of counterfeit and relabeled chips from China!)
Two examples below, a big old hammer or a vice might work.
Nothing to loose since they are already dead!

https://twitter.com/johndmcmaster/st...10403701624833

https://www.youtube.com/watch?v=W_hIVtcx6CA

[rddube]

Originally Posted by Per Hansson

I would take your current destroyed parts and open them up:
You have one original from the unit and the one you bought.
Open them both and see if the die is the same physical size.
If it is not you have your probable cause why it failed.
(Lots of counterfeit and relabeled chips from China!)
Two examples below, a big old hammer or a vice might work.
Nothing to loose since they are already dead!

https://twitter.com/johndmcmaster/st...10403701624833

https://www.youtube.com/watch?v=W_hIVtcx6CA

Ok, so I opened an original (which is the one on the right and was much more difficult to open as it was quite more solid) and one I got from China (the one on the left with it's housing beside it). Do you see any difference in the die?

[rddube]

Help anyone?

[stj]

they look the same to me.

[Per Hansson]

Yup that looks to be the same size to me also, another very important factor is how well they are attached to the heatsink, some cheap crap is made very poorly but this looks soldered so I'd say they are ok from this look.

[rddube]

Ok, but if we get back to my original problem i.e. Q510 (2SC5669) popping while taking with it Q509 (2SA2031) - I've checked every component on the left channel and compared it to the right channel and they are very close.

Petehall453 suggested to check emitter resistors and current limiter resistors but I am unsure which are which. Can someone identify one of each on the attached schematic? Or what else should I be checking? There are 2 100uf 100V capacitors per channel, could they be the culprits? I took them out and checked them on my tester and they come up ok?

Many thanks!

[petehall347]

concentrate here including the bias trimmer r522 .

[rddube]

Originally Posted by petehall347

concentrate here including the bias trimmer r522 .

Thank you Petehall347! Will measure and report back including the bias trimmer. Tks again!

[rddube]

Originally Posted by petehall347

concentrate here including the bias trimmer r522 .

OK so here are the results of every resistor - the only one not corresponding to the schematics is R514, but see my note.

As for r522, when turned fully clockwise I get 1.2K ohms, and fully counterclockwise I get 2.6ohms.

Many thanks for your input!

[rddube]

I tested r514 out of circuit to make sure and and it tests 10.086K ..as for my note in the schematic, I made a typo as r549 reads 2.986K ohms and not 1.986K ohms. Correction attached.

Also, I tested D503 out of circuit and it tests good.

[eccerr0r]

Ensure Q562/Q507 and Q563/Q508 aren't shorted and Q506 not open...
Curious are you measuring the voltage across R523 when it's operating? Is this what you're getting the 1mV you were measuring earlier?

BTW R514 is fine the way it is, we know it's working because your amp went into protection. Maybe too late to protect the devices...

You might need to do some power-on voltage tests with Q510 and Q509 removed and so you don't fry anymore, though R523 voltage tests will require these two to be installed...

[rddube]

Originally Posted by eccerr0r

Ensure Q562/Q507 and Q563/Q508 aren't shorted and Q506 not open...
Curious are you measuring the voltage across R523 when it's operating? Is this what you're getting the 1mV you were measuring earlier?

BTW R514 is fine the way it is, we know it's working because your amp went into protection. Maybe too late to protect the devices...

You might need to do some power-on voltage tests with Q510 and Q509 removed and so you don't fry anymore, though R523 voltage tests will require these two to be installed...

Hello Eccerr0r! And thank you for stepping in!! Ok, I will remove Q562/Q507 and Q563/Q508 and Q506 and report back. As for your question, I am measuring R523 at the tespoint P501 for the 1mv that is stated in the manual. I have to admit however that contrary to the right channel that when turned completely clockwise gave me 0mv until I turned it counterclockwise about half way to get to 1mv, the left channel (P501) turned completely clockwise started at 1mv and when I turned it to about halfway counterclockwise is when it went from 1mv to 2mv, I therefore stepped back to 1mv and left it there. While I was turning counterclockwise, it would go from 1mv to 0mv sporadically and it is only when it hit 2mv did it become stable and going back to 1mv it stayed there for at least 5 minutes (which is what is recommended in the manual).

I will report back and I can do any measurement you will ask me to do..again thank you for your reply..be back in a few....

[rddube]

Ok, so removed Q562/Q507, Q563/Q508 and Q506 and all are good, they show NPN and PNP (accordingly) on my transistor tester. So it comes down to only Q509 (2SA2031) and Q510 (2SC5669) shorted before going into protection mode.

[eccerr0r]

What speakers are you using, are you simply expecting too much from the output?
Do the transistors get hot when idling?

If they are getting hot while idling, and R520 getting hot implies that something upstream is pulling the drivers apart... but this contradicts the 1mV seen at the test point. So I'll have to ignore testpoint data assuming perhaps you have a speaker attached or something. What is the voltage between the collector and emitter of Q506 when running (okay if Q509 and Q510 are out of circuit)? Also what's the voltage across R5143?

[rddube]

I'm using 4 ohm speakers that sound very good, they are Mission speakers. I don't know if the transistors are getting hot when idling, I didn't measure that. You have to see that the output transistors are folded in the back of the board screwed into the heatsink, so i'ts a bit difficult to read temperatures and voltages down there.

As for the voltage measurements, you want me to put it back together and maybe solder myself some wires to Q506 collector and emitter to be able to measure voltage and same for R5143, as all these components are down at the bottom and I think I have to reinstall all the boards in front of the Front/Center amplifier board. I could always try it without all the other boards, just the main board and see if it trips the protection. What do you suggest, and again thanks!

[eccerr0r]

If they're getting hot to fry, you should feel the heat on the heatsink. As a class B it should not be getting (that) hot when there's no audio signal minus the (rail voltage*2)*(1mV/0.22Ω) loss per channel. IMHO this seems too low, but giving benefit of the doubt, it should not even get warm unless you have like 10 channels or so, plus whatever the other circuits dissipate.

Does that 220Ω resistor burn up when idling too, or does it burn when you're really running it? That's another clue...

[rddube]

Well there's only 3 channels on that board and I don't think they get hot, but I could be mistaken. I do have a thermal camera and a laser temperature probe so I guess I could check their temperature but right now Q509 and Q510 are shipping (from China, the last source I got the others from).

To be perfectly honest, the last time the 220ohms resistor burned, was just after one of the output transistors had popped, it went into protection mode and I stupidly turned it back on to see. That is when I saw a glow and by the time I had turned it back off that 220ohm resistor had not completely fried, but let's say cooked.

I will be getting the replacement IC's within a few weeks, and so last time I had put it back together everything tight, it was fine on idle and low volume - so I could definitely take measurements then. It's when I cranked it up and after listening to 1 song about 1/4 of the way that that transistor popped and went into protection mode. I did not turn it back on, rather I dissassembled it and found Q510 and Q509 shorted.

So do you think I should wait to have the transistors back in (I got 4 pairs just in case) and do the tests then, or would it be better to do as you had sais in your previous post, test it without the Q509 and Q510?

[rddube]

I also have to admit that Q509, the hole in the heatsink is worn so the screw holding it is not as tight as the others. Now that I have the right insulator pads with non conductive thermal grease, do you have any tricks on how I can get that screw to hold tighter? I've been using a wooden toothpick and not tightening too tight so as not to lose the grip. Maybe it's Q509 that trips Q510 and not the other way around?