Problem with Rotel RA-1062

[soren5]

I have a Rotel RA-1062 that started shutting down after playing for a few hours. If I turned it off completely and let it cool down it would work again for another few hours. However, eventually it got too hot, the fuses blew, and it was completely dead.

I'm no expert in amps, but I know how to solder, so I downloaded the service manual and looked at some other threads with people having similar issues with this amp. I tested and realised some transistors were bad, so I ended up changing a bunch of them:

(Q627-630) 2SD1047P-DE -> NJW0281
(Q631-634) 2SB817P-DE -> NJW0302

(Q623,624) 2SC3902-RS -> KSA1220
(Q625,626) 2SA1507-RS -> KSC2690

(Q901,903) 2SD600K-EF -> KSC2690
(Q902) 2SB631K-EF -> KSA1220

Then I soldered everything back, replaced the F603-604 fuses, but they blew instantly when I turned it back on.

Anyone who can help me out?

Thanks.

 

[Lojzek]

Check 2sc3902 against ksa1220. The first one is an NPN, the second one a PNP transistor.

 

[JonSnell Electronic]

Which components were the faulty ones?
Fault find, replace faulty components and then test with a load limiter.

 

[soren5]

Check 2sc3902 against ksa1220. The first one is an NPN, the second one a PNP transistor.

You're right I was looking at this post stating that 2SC3902 could be replaced by 2SB631K, and then this post saying that 2SB631 could be replaced by KSA1220. But I guess that wasn't the case. Do you know the correct substitute for 2SC3902?

 

[soren5]

Which components were the faulty ones?
Fault find, replace faulty components and then test with a load limiter.

Half of the transistors on the heatsink were faulty, but the original 2SD1047P and 2SB817P had been replaced by 2SC5242 and 2SA1941, so I decided to change all of the output transistors, including the predrivers. From reading other threads I decided to test Q901-903, which were also faulty so I replaced them.

What's a load limiter, and how do I test with it? Sorry but I'm a bit of a noob here – guess I'm out of my depth

 

[Lojzek]

Hi soren5,

I meant to say an NPN should be substitued with another NPN.

(Q623,624) 2SC3902-RS -> KSC2690
(Q625,626) 2SA1507-RS -> KSA1220

 

[soren5]

Hi soren5,

I meant to say an NPN should be substitued with another NPN.

(Q623,624) 2SC3902-RS -> KSC2690
(Q625,626) 2SA1507-RS -> KSA1220

Is that it? Simply switch the KSC2690 and KSA1220 in Q623,624 and Q625,626?

 

[Lojzek]

You have probably made a mistake and installed wrongly a pnp transistor in place of where should have been an npn.

 

[soren5]

I think you're right. I will check all of them and see if I have made other mistakes. When I'm done and have corrected the mistakes, is there anything I can do to test before powering up and risking burning another pair of fuses?

 

[mbz]

I think you're right. I will check all of them and see if I have made other mistakes. When I'm done and have corrected the mistakes, is there anything I can do to test before powering up and risking burning another pair of fuses?

As previously suggested use a current limiter aka Dim Bulb Tester, google for plans. Just about mandatory in this situation.

 

[soren5]

So I googled Dim Bulb Tester and made one with a 100W dimmable halogen bulb (incandescent wasn’t available). I then changed the NPN/PNP transistors that had been put in reverse.

Now when I power it up the bulb lights up and stays on and the amp seems to stay in soft start and never gets to the point where input selector works.

What to do?

 

[Mark Tillotson]

Halogen bulbs _are_ incandescent!
You may have to defeat the soft-start circuit with a DBT.

 

[soren5]

Defeat like bypass? Any advice on how to do that? I'm at a bit of a loss here :/

 

[soren5]

So what I called soft start is probably just the speaker protection relays that click when switching on the amp. Should I try bypassing those?

Maybe I could try connecting a test speaker to the negative terminal and the heatsink side of the fuse holder as suggested in this post?

Any advice would be greatly appreciated 🙏🏼

 

[mikeAtx]

Bulb should dim to almost off if there is no signal. You could hook a inexpensive speaker up to the fuse directly. Before you do, you should probably test the voltage at the fuse with a DMM. It should be near zero volts DC and AC. If the bulb stays bright, then you have a bad fault in the amp, which the bulb is protecting against things burning up.

 

[soren5]

Bulb stays bright so I expect there's still a fault in the amp. Unless the startup protection relays are causing the bulb to stay on?

How should I go about measuring the voltage at the fuse? There are four fuses, F601-604, where F603-604 blow when turning on the amp. Should I remove those two and measure the voltage between the two poles on each fuse holder?

 

[Mark Tillotson]

So what I called soft start is probably just the speaker protection relays that click when switching on the amp. Should I try bypassing those?

No, those are protection relays. They save your speaker from DC offset and thumps at turn-on.



A soft-start circuit limits the current to the power supply at switch on to reduce the inrush current from the mains.[/QUOTE]

 

[audiopro]

Do not try to defeat anything. The reason the bulb is bright is you are drawing huge current. The reason you are drawing huge current is you blew something, or installed something incorrectly.
Don't just blindly change transistors. Test them, look up proper equivalents, and then replace what's needed.
You have now turned an inexpensive job into a disaster. I hope you are able to sort it out.
Start at the outputs and test your way to the front. Check and double check.
FYI - your initial problem was caused by bias drifting, which can be caused by a number of things, including bad caps, or it was set too high to begin with.

 

[mikeAtx]

I took a quick look at the schematic and it looks like it is pretty easy to pull the "amp" boards of the system. I'd pull all that and see if the preamp side is working ok and not faulting. So if the preamp is alive (voltages look ok at B+/B- and supply voltages for preamp are good, I'd put in the amp boards one channel at a time to isolate. All the while keeping the bulb protection in. The bulb should go very dim if all is good. Initially bright as the supply caps come up, but then so dim you may not see the filament lit.

 

[soren5]

Do not try to defeat anything. The reason the bulb is bright is you are drawing huge current. The reason you are drawing huge current is you blew something, or installed something incorrectly.
Don't just blindly change transistors. Test them, look up proper equivalents, and then replace what's needed.
You have now turned an inexpensive job into a disaster. I hope you are able to sort it out.
Start at the outputs and test your way to the front. Check and double check.
FYI - your initial problem was caused by bias drifting, which can be caused by a number of things, including bad caps, or it was set too high to begin with.

That’s a lesson learned then! I didn’t realise that was the way to go about it, though it makes a lot of sense now As I said I’m not an expert in amps / electronics, but I still hope that I’ll be able to fix it. I’ll start at the outputs and test my way to the front, and by this I assume you mean test for faulty components?