Burned amp

Hello guys!

I worked on some improvements and I applied some change which caused some short circuit (honestly - I messed up).
I reverted my changes back to original, but the two 100ohm resistors R105 R107 ohm will get very hot and burn again.
Any idea what I should replace? That two transistors near resistors?

Thank you for help!!!

See the scheme please:
burned.jpg
 
Last edited:
First of all, make sure that there is no excessive loading left on +/-15 V - a quick resistance check might already give some clues.

What exactly went wrong in the first place? Maybe draw the changes you were performing. This may give some clues as to where the problem is now. That said, if those resistors overload that badly, the regulator transistors may in fact be toast. If you had previously managed to create a short across +/-15 V, it wouldn't take much to fry them, they are just little TO-92s with little over half a watt of rated dissipation.

If you don't have any 2SC1815/2SA1015 on hand, I'd say BC639/640 would make a good replacement. BC337/327 would require dancing the ol' leg twist to match ECB pinout.
 
@KarlvdBerg I did new PCB for CD input module, because I wanted 70um and wider copper paths, also silver plated. Yes, I try to do maximum to my Yamaha AX-10. And I guess I messed up some DALE resistors. I tried to be detailed and precision, but it turned into failure... (wasted 40 USD + time). The little board is taking 15V from the mainboard.

@nigelwright7557 I've checked the zener diodes. Yes, they are dead. Now I soldered-in good ones.

@sgrossklass and @nigelwright7557 Do you think I should replace also Q105 and Q107 just to be sure? I have only one try left, if I will burn it again, I will run out of these old original components.
 
Before I changed Q105 & Q107 I'd replace C101,102, 106, 107 on something this old. Leaks in any could burn this circuit.
Lift base of Q105 & Q107 to test with diode test of DVM. Forwards from base twice to C & E, 450-700 mv, backwards 9999 or ----.
These aren't special transistors, just match pinout. EBC plastic TO92 use MPSA06/56 or BC546/556 or 10W metal can farnell has 2n5320 2N5321 npn and pnp 2n5323 I saw yesterday. BCE TO92 use BC639/640, metal can TO39 use BD139/140 backwards with bit of aluminum as heat sink. If TO5 use the TO39 parts & add heat sink. Designers tend to leave out heat sinks to save cost, and cut the life of their product. Add them back in and forget replacing this part again. A bit of sawed up window frame with a hole in works for a small heat sink, it doesn't have to be factory made and it doesn't have to be black. Heat compound can cover up surface mismatch at low wattage.
 
Last edited:

Karl vd Berg

Member
2012-07-16 12:49 am
I'd replace also C105 and C115 (100uF/6.3V), because of little Vdc margin. Something like 100uF/25V or 35V would be safer...

Need to know the amp model and if it's still working under 220V, so 16.3Vdc after rectifiers would be bit higher under 230V in CZ...
 
Last edited:
I'd replace also C105 and C115 (100uF/6.3V), because of little Vdc margin. Something like 100uF/25V or 35V would be safer...

Need to know the amp model and if it's still working under 220V, so 16.3Vdc after rectifiers would be bit higher under 230V in CZ...

It is Yamaha AX-10 for 230V, I am running it on 240V. As I mentioned, I caused the issue with my modification.

I hope I will have a chance to solder working parts this evening, so I will let you know what is the actual situation.

I am glad I have almost all parts, sadly just one pair of transistors near resistors and one pair of zener diodes.
 
@nigelwright7557 I've checked the zener diodes. Yes, they are dead. Now I soldered-in good ones.
I do have to wonder how the hell you managed to fry those.

Well, if the transistors were a dead short across B-C it could definitely happen... Would explain why your 100Rs were smoking, too.
@sgrossklass and @nigelwright7557 Do you think I should replace also Q105 and Q107 just to be sure?
Ask the diode test first, which I presume will indicate some rather shorted pn junctions. It is very likely that they are as dead as any dodo.
 
Hello to everybody, I wish you all the best in 2020!

I apologise I stayed silent. Life sometime will make us busy in different way as we would like.

Back to amp. It is burned, because ... oh, what a shame.... I connected power transformer incorrectly. Instead of 72-0-72 I connected 72-72-0. That was really silly. :headbash:

So, I have second Yamaha AX-10 as backup, so I am not in silence, but I have to change lots of components in the burned one, I want it back alive ...
 
Happy new year.
I replaced 126 parts in a many time shorted PV-1.3k. Most were free, salvage from old PCAT power supplies, like diodes & such. Learned a lot about what different parts of the circuit did.
As you probably have concluded, measure power supplies before use, especially ones you build. Load resistors are handy on switcher supplies, which don't like zero load. Also mating connectors to whatever connector the PS has. I buy old wall transformers for $1 to $3 from a charity resale shop, definitely check those with DVM before use. Some hum, some have 25% greater voltage without load.
These leaded component projects are a pleasure to work on. I Had a problem with turn on switch in a 2016 samsung TV since July, couldn't trace a mult-level board to see where the bad via (through hole plated) was. Would work when I warped board. So luckily this month I found a matching board on ebay for $30. Didn't learn much from that experience. Fixed a PV8 mixer with a bad via making left outs not work in October, learned a lot from that. Surface mount, but 2 sided board, traceable.
When working through a system subject to overvoltage, a leakage current tester at say, 12 v, is more use than a capacitor or transistor gain tester. 12 v supply 47 k resistor, dvm current scale in series, see how bad they leak. Lots of zotted 50 v ceramic caps in the PV-1.3k. Some transistors passed double diode test on meter but failed leakage check.
 
Last edited:
I do have to wonder how the hell you managed to fry those.

Well, if the transistors were a dead short across B-C it could definitely happen... Would explain why your 100Rs were smoking, too.

Ask the diode test first, which I presume will indicate some rather shorted pn junctions. It is very likely that they are as dead as any dodo.
Yep. Shorted (or open) Zeners would not cause those resistors to overheat. Almost has to be direct C-B shorts on the two transistors.
 
@indianajo, Yes, it is the way how to learn something more. I have lots of components, so I should be able to manage it. The question is mainly about time. But I will panage it, board by board :)

@Ylli I fixed the mainboard + I changed Zeners. Now I have sound, a very weird sound. No smoke. I think I will start with checking voltages on board. I am glad to have few extra old transistors for testing purposes.