• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Sony STR-2800 Two Blown Fuses

A

AFishyFish

New Member
Hi everyone.
My Sony STR-2800 has blown two 4AT fuses (F803 & 804). I wasn't there to witness it, so it's hard to say how exactly it happened and whether someone maxed out the volume knob or something.

It powers on, but there's no signal coming through the speakers and headphones outputs.
A few months ago I sent it to be repaired when the right channel stopped working. The tech did not specify exactly what he'd done, but I believe that he replaced the two output transistors of the right channel. The original transistors are 2SC1061 (Q707 & 757) and 2SA671 (Q708 & 758). They were replaced with D613 and A769 respectively. It looks as though no new thermal paste was reapplied.


From what I could gather, AT is a slow-blow fuse. Is that correct?
Are the new transistors a suitable replacement to the originals?

I was planning on replacing the fuses, and if they blow again, start checking each of the output transistors. What would you recommend doing in this case?

Thank you for taking the time to read. Any suggestion or guidance is much appreciated.

Sony STR-2800 Repair Manual
Photo of new output transistors
Photo of fuses
Photo of the receiver's insides
 
Register to hide this ad
I would find the short before putting more fuses in. Looking at the schematic, you must have an E model with adjustable input voltage. Is it set for your mains voltage?
If you have a multimeter, check the bridge rectifier diodes for shorts (with the unit unplugged).
 
Yes, it's the European model, set to receive local mains voltage.
I'll remove and check diodes D801-804 as soon as I can. Thanks for the help.
 
Too late... I disconnected one lead of each diode. They're all fine and give out a reading of around 0.5 to 0.6 Volts. What do you think I should check next?
 
Go ahead and check the output transistors next. If they are ok, put the fuses in and power it up.
 
I flipped the receiver to reach the board and a dead bug dropped from somewhere inside. How likely is it that it caused a short?
Anyway, transistors Q757 and Q758 are completely shorted. The two other (original) output transistors are fine. Could it be that the replacement transistors weren't up for the job from the get go or could it be something else that caused them both to fail? Are they the cause of the problem or just a consequence like the fuses? Should I replace the transistors and fuses and attempt to turn the receiver on or should I check something further along the flow chart?
 
The dead bug would not have caused a problem. The transistors used would hold up just fine but the tech did not use a proper complementary pair. The 2SD613 should be paired with a 2SB633. He just used whatever he had laying around. I would not go back there. It is possible that there is something else that is wrong that caused the original failure and since it was not addressed the amp blew again. It is also possible the replacement transistors were counterfeit and blew when the receiver was played at high volume. It's possible you have a frozen woofer on that channel. Check you speakers carefully. Gently push the woofer evenly in and check for rubbing sounds or binding. It should move freely. Check your speaker wires at both ends to make sure there are no stray wires that can short together. The wires should be stripped so that only 1/2" of bare wire is visible and the strands neatly twisted together. Any splices should be insulated with electrical tape.
Now to the receiver. Replace the 2SD with a MJE15030G and the 2SA with a MJE15031G. These are much better transistors than the originals. In fact you should replace them on the other channel as well.
Check RT751. If it were to open up or get dirty, the outputs would receive full current and blow up. Check Q753 - Q756. Check D751. If all are ok then you can replace the outputs and power it up.
 
Thank you for taking the time to write this extremely helpful information. I'll report back with more information after checking everything you've suggested.
 
RT751 and Q753-756 are fine, although none of those transistors are original - Q753 (2SA896) is A899B, Q754 (2SC1811) is C1904B, Q755 (2SC1475) is C1509) and Q756 (2SA706) is now BC640.

I'm not sure about D751 (SV04F). I'm getting an OL reading both ways, same with D701. It's not a simple diode so I'm probably approaching it wrong. Here's some information about it - http://electronics.stackexchange.com/questions/243654/what-can-i-use-to-replace-an-sv-04f-bias-diode
How can I reliably test them to be certain?

My local shop has MJE15032 and MJE15033 in stock. Only difference from the MJE15030 and MJE15031 seems to be the voltage rating, unless I'm missing something. Is it okay if I use 250V instead of 150V?
 
To test the diode you can make a circuit with a 1K resistor in series with the diode and either a bench power supply or a 9V battery. Connect the + end of the battery or supply to the anode of the diode and connect the cathode of the diode to the 1K resistor. Connect the other end of the resistor to the - terminal of the battery or supply. If it's a supply, set it anywhere from 9V - 12V. Connect your DMM, set to volts, across the diode. I would expect a good diode to measure 2.4V - 2.8V across it.
The 15032 and 15033 are fine to use. For now you can leave the other transistors in place but I would eventually replace them with proper complementary pairs. I will have to look them up and see what is currently available.
 
Diodes measured 2.6-2.7V, so I installed the new transistors in both channels and new fuses and... R764 went up in smoke. Q757 and Q758 probably fried again, but I haven't tested them individually yet. Q756 is probably affected as well. I installed the transistors with a thin layer of thermal paste on both sides of the insulator and checked that there's no continuity between the collectors and the heat sink. I hope I didn't miss anything when testing the smaller transistors or ruining them when soldering them back. I was pretty careful with that. Any thoughts as to what caused this?

Update: I connected D751 and D701 backwards when soldering them back. It was the black anode cable which threw me off. What a dumb mistake. Where should I check for more possible damage because of this? If the diodes are to blame, shouldn't there be similar damage on the other channel?

Update 2: Q758 and Q757 confirmed not working. Q756, D751, D701, RT751 are fine, so is R764, which is charred but still spot on 470 ohms. I'll test the rest of the transistors and update as soon as I can.

Update 3: I'm now getting continuity between Q708/701's collectors and the heatsink. No continuity between the heatsink and the other channel's fried output transistors' collectors. I couldn't count on the thermal paste being non-conductive so I really tried to make sure that the transistors are insulated, but it wasn't enough.
 
Last edited:
There should be mica insulator sheets between the metal tabs on the transistors and the heatsink, and insulating washers on the screws for the transistors.
 
Yeah all of the above including maybe you got rat who somebody cranked the bias pots the wrong way and blew the drivers too. CCW = postive CW is off and usually cheap plastic. I'd rebuild the p.s. in parts and look and any scr's for age or test said chips, too. usually hung in space and should have an aluminum heat sink.. many models seem to skip the heat sink so the regulator goes south. Well of late tests on str's da333-555 es era.. if the scr is south you won't get above idle... i.e. 63vdc rails only mearure 33-38vdc roughly half or less? so hasn't kicked in the drivers.

Sony layouts are sort of predictable aside from fancy schitt.
 
dr*audio - I used the insulator sheets after cleaning them and spreading new thermal paste on both sides and also used the washers. I checked for continuity after installing them and for some reason there was none. Now there is. It could be that the paste is highly conductive and got on the mounting screws, or that there is something that I'm completely missing. I'll need to test some more. Could it be that the two misoriented diodes caused the properly insulated transistors (Q757,Q758) to fail, yet didn't harm the other pair (Q708, Q707) because they shunted the current to ground through the heatsink? I'm just throwing a very uneducated guess. Sorry if it sounds preposterous.

Binkman - thanks for joining in. I'm afraid that most of what you wrote passed over my head. My knowledge of electronics is very limited and I'm trying to use the little I know to understand the suggestions made by the good people of audiokarma. I know that it can be frustrating to help someone who seems to know next to nothing about the subject at hand. When I tested RT751 (bias pot), I made sure to dial it back to how I found it - 13.66kΩ. I should check the other channel's bias pot for comparison.
 
You didn't use the heatsink paste that is meant for computers, did you? The arctic silver? That stuff is conductive. A common newbie error is to over - tighten the screws and crack the washers, or mis - install the washers so the shaft on them that is supposed to go through the hole in the transistor tab, doesn't go through and gets crushed on the surface of the tab. Sometimes the original transistors have a larger hole than the modern replacements and the old washer won't fit through the hole. sometimes the insulator gets torn.
 
It's a compound that's meant for this task. This is the stuff I used - http://www.farnell.com/datasheets/1792677.pdf?_ga=1.237356038.768867356.1481373872
The screws are certainly the prime suspects. I'll check all the parts involved.

What I plan to do is install the old functional output transistors in place of Q757 and Q758 and triple check continuity. Now that the diodes are oriented correctly, do you think it's time for another power-up or should I check something else?
 
If you have not adjusted the bias I would not attempt it as they are really cheap plastic. if using orig parts it will probably be okay. But you should become familiar of how to measure safely without shorting out the test points if equipped on the pcb. Others above have the lead so wait on their replies.

You might invest in building a dim bulb tester if you plan to repair more stuff.
 
Test the transistors on both channels in case the other channel got blown as well.
Set the bias pot to minimum resistance before powering it up.
 
If adjusting bias to 0 remember it's Clock wise. CCW = +mv I would pull both pots to be sure you have range before test run.. and defective pot and you can fry stuff.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom