Blown fuse and smokin’ resistor on Technics SA5170

[Anonky]

2SB616 is the compliment, whatever transistor number that happens to be. It'll be the other one mounted to the heat sink.

Apologies for being slow here. There are four transistors mounted to the chassis. I just need to know specifically which one 2sb616 is the replacement for. Also are you suggesting I replace both of the output transistors on each channel (the transistor that sd586 replaces)? I just need very explicit direction. Thanks

 

[cademan]

Don't need to touch the good working channel but if only one out of the two were shorted on the bad channel, its good practice to replace both even if one might be good.

If only one out of the two was shorted and the other one was good, what usually happens is the shorted one stresses the other good one and is now no longer trusted or reliable if you replace just the one that was shorted.

When talking about complimentary pairs of transistors where two transistors have different numbers, be sure that the replacements are in the right positions as the old one's. If you swap them around or are reversed, it will not work and the new transistors could be damaged.

 

[gadget73]

Apologies for being slow here. There are four transistors mounted to the chassis. I just need to know specifically which one 2sb616 is the replacement for. Also are you suggesting I replace both of the output transistors on each channel (the transistor that sd586 replaces)? I just need very explicit direction. Thanks

ok so I looked at the schematic and realized this isn't an NPN/PNP pair like I expected. Assumptions and all that. Apologies for the confusion.

Its all NPN transistors so both in the channel will be a 2SD586. Either way, I'd still replace both of them on the dead channel.

also the 2SD586 is one of those annoying obsolete package parts. I don't think anything currently made is a perfect drop-in for it. Depending on exactly how they mounted it will determine what makes sense to replace it with, and you may be stuck drilling and tapping a hole for a new part. 2SC5200 would be my best guess for a sub, but someone may have a better suggestion.

 

[Anonky]

ok so I looked at the schematic and realized this isn't an NPN/PNP pair like I expected. Assumptions and all that. Apologies for the confusion.

Its all NPN transistors so both in the channel will be a 2SD586. Either way, I'd still replace both of them on the dead channel.

also the 2SD586 is one of those annoying obsolete package parts. I don't think anything currently made is a perfect drop-in for it. Depending on exactly how they mounted it will determine what makes sense to replace it with, and you may be stuck drilling and tapping a hole for a new part. 2SC5200 would be my best guess for a sub, but someone may have a better suggestion.

Previously you mentioned using MJE15003/MJE15004. Would you recommend 2sc5200 instead?

 

[gadget73]

bad assumption on my part, I hadn't looked up the datasheet for the 2sd586 to see what the actual package is. I thought it was a TO-3, the "flying saucer" type, but its an XM20 "batwing".

TO3

XM20 (some are all plastic without the metal tabs)

The 2SC5200 is a TO-3P or a TO-264 depending who makes it. I have no idea what the difference is since they appear identical but they list both ways. This is what that looks like

the mounting differences are pretty obvious, probably gonna have to drill and tap a hole to make that work. Maybe there is a better solution, honestly I don't know. I've never had to replace a "batwing" transistor so its not something I've personally dealt with. I suspect it would be less trouble to mount this vs a flying saucer though.

 

[petehall347]

bad assumption on my part, I hadn't looked up the datasheet for the 2sd586 to see what the actual package is. I thought it was a TO-3, the "flying saucer" type, but its an XM20 "batwing".

TO3

XM20 (some are all plastic without the metal tabs)

The 2SC5200 is a TO-3P or a TO-264 depending who makes it. I have no idea what the difference is since they appear identical but they list both ways. This is what that looks like

the mounting differences are pretty obvious, probably gonna have to drill and tap a hole to make that work. Maybe there is a better solution, honestly I don't know. I've never had to replace a "batwing" transistor so its not something I've personally dealt with. I suspect it would be less trouble to mount this vs a flying saucer though.

i have in the past used the old transistors to clamp the subs down using longer screws . that way at a glance it appears the old ones haven't been touched

 

[gadget73]

hm, thats a thought, provided the screws clear the body of the replacement part.

 

[Anonky]

i have in the past used the old transistors to clamp the subs down using longer screws . that way at a glance it appears the old ones haven't been touched

So here is the verdict:

2sc828–>KSC1845

2sc2003–>KSC2316

2SA954–>KSA1220

2SD586–> 2sc5200 (x2)

seem like a reasonable order for my problem transistors?

 

[Anonky]

So here is the verdict:

2sc828–>KSC1845

2sc2003–>KSC2316

2SA954–>KSA1220

2SD586–> 2sc5200 (x2)

seem like a reasonable order for my problem transistors?

Ok so other than the ksc1845 these are all obsolete unfortunately. Ugh

edit: nevermind I at least found the 2sc5200 on mouser. There are a few options including 2sc5200N made by toshiba (which is a couple bucks cheaper). Would that work? I understand I may have to tap another hole

 

[gadget73]

should be fine, slightly different letters after the part number depending on if OnSemi or Toshiba makes it but its the same part where it matters.

grab the insulator micas for these too, otherwise it will short to the heat sink.

in place of the KSA1220 you should be able to use a TTA004b

 

[Anonky]

should be fine, slightly different letters after the part number depending on if OnSemi or Toshiba makes it but its the same part where it matters.

grab the insulator micas for these too, otherwise it will short to the heat sink.

in place of the KSA1220 you should be able to use a TTA004b

Insulator micas? Any specific product you’d recommend? New to replacing parts and not familiar with this product. You you just stick that on the side so it doesn’t make physical contact with the heat sink?

 

[gadget73]

big transistors usually have the collector tied to the metal part of the back. If you don't put the mica under there it shorts to the heat sink and then you've got a whole whack of other problems.

this is standard mica, it will need a very thin application of thermal compound on either side. This sits between the output transistor and the heat sink

https://www.mouser.com/ProductDetail/Aavid/56-02-101G?qs=na02DS%2BV37QRPZKlvao/bg==

this is a more modern thermal pad. No thermal poop needed, sits in the same location

https://www.mouser.com/ProductDetail/Wakefield-Vette/CD-02-05-264?qs=5aG0NVq1C4zA2V6Q7RBAXg==

 

[Anonky]

Wanted to finally provide an update on this. The room with my workbench had a leak so I’ve been out of commission for the longest time. But since the leak is repaired I’ve finally sat back down at the bench today and finished up this project.

Wanted to say thanks to everyone helping me through it. @gadget73 @cademan this was only my second component level repair I’ve had to troubleshoot so I appreciate yall’s help immensely. The receiver is singing once again and might I add, it looks immaculate.

 

[Anonky]

Ok guys @gadget73 @cademan i spoke too soon. After playing on DBT for a while I moved to an outlet without the DBT. After 10-15 mins roughly Power fuse blew again and I had some smoke. Not sure where the smoke came from but there are no obvious resistors burned up. There is a zener diode with a black line down it which looks possibly burned? On diode test I’m reading 1050 ohms in one direction and 800 in the other. Could this be my problem?

 

[Anonky]

I also think I may have erroneously been using the wrong wattage bulb initially. I was using 100W bulb on the DBT when testing it while it was “working properly.” After I had problems just now i was investigating what I may have done wrong and realized I was using a 100W bulb. I changed to a 60W bulb and something is definitely shorted as I’m getting very bright bulb when I power on for first couple seconds (doesn’t dim) so I shut it back off… ugh

UPDATE: I put the 100w bulb back in the DBT and powered the unit on again and it’s glowing brightly so I suppose I didn’t make a mistake with the bulb. There is now a new short in the receiver that occurred since my repairs. Any tips appreciated to help locate the short.

 

[gadget73]

what did you end up doing with it to make it go initially ?

honestly its probably a circle back to the initial troubleshooting steps. Sounds like a shorted transistor somewhere.

 

[Anonky]

what did you end up doing with it to make it go initially ?

honestly its probably a circle back to the initial troubleshooting steps. Sounds like a shorted transistor somewhere.

I replaced all of the faulty transistors and resistors that we had previously discussed. Made sure the pin outs were correct. Only thing I can see that would’ve caused smoke is that this zener diode looks to be black? Based on measurements taken, it seems that it is possibly shorted so I’m ordering replacements.

Again I played it on the dim bulb for like an hour without issue. Moved to an unprotected outlet as it seemed to do fine. Then after 15ish mins power fuse blew. Now when I plug it in DBT is indicating a short (after replacing fuse). I tested all of the resistors I replaced and they’re still in spec so the resistors are good. Might replace the zener diode and give it another go? If still indicating a short at that point I’ll pull the transistors.

edit: that cap in the pic kinda looks bulging but I don’t think it is. The wrap is just too low maybe? Or maybe it is bulging? Idk. Might just replace the cap anyways.

 

[gadget73]

zeners do tend to fail shorted, but that begs the question of why? They normally only fail like that when subjected to excessive current flow. I can't pull up the whole schematic right now so no idea what thats connected to.

 

[Anonky]

zeners do tend to fail shorted, but that begs the question of why? They normally only fail like that when subjected to excessive current flow. I can't pull up the whole schematic right now so no idea what thats connected to.

Here is a photo

 

[gadget73]

looks like its being powered from a 27.5v line through a 1k resistor, so the max that could flow is 27.5 milliamps, or about 3/4 of a watt. Might be worth confirming that 27.5v line is in fact 27.5 volts. I also dont see that being enough current to make the DBT run at full brightness.