Weird protection relay behavior: Realistic STA-2000

[kendo581]

Hi AKers,
I've owned a STA-2000 for a couple years now, and it has performed flawlessly until just recently. I've always had it hooked up to a surge protector, but I decided (probably stupidly) to remove the surge protector and had it plugged straight into the wall outlet.

Not surprisingly a few days later, when listening to a record, I heard a pop in the speakers and the protection relay clicked open. About 3 seconds later, it clicked closed again, and everything seemed fine. A few days after that, the same pop occurred but this time the relay never closed again. No more sound, but all lights, tuner, etc. power on. I figured it was something with the protection relay protecting the speakers. After I found the service manual on this site (thanks 71Challenger):
http://www.audiokarma.org/forums/showthread.php?t=291764

I began checking all the diodes and transistors on the protection board as was recommended in the troublshooting section (#36 in the manual). I'm not a expert, but I found something weird. When I check the continuity of transistor TR601 by touching one lead (black or red) to the B leg and the other lead to the C leg, the relay closes! It doesn't matter if the multimeter is set for impedence or DCV. Is this normal? The leads definitely aren't touching, so I sure I'm not shorting something. None of the other transistors do this.

I was hoping that this transistor was the problem, so I hooked a test speaker up to the STA-2000, and checked the R and L channels separately for sound when I closed the relay "manually". To my delight, the R channel worked fine when the relay closed. But, the L channel only emits a LOUD high pitched noise when the relay is closed! It doesn't seem to matter if the L channel is hooked up to the A or B speakers.

So now I think it is something wrong on the L channel main amp board. If there was something wrong on L amp board, would this prevent the relay from closing? Do these symptom sound like a short somewhere? Overload to the amp? I've posted some pictures of the protection board (with TR601 highlighted), and the curcuit diagram from the manual. Any suggestions? Thanks in advance!

http://s1265.photobucket.com/albums/jj506/kendo581/Realistic%20STA-2000/

 

[kcollins4]

So you're checking the transistors while they're still in the circuit?

 

[kendo581]

Yeah, I've been checking them while in the circuit, both with the power on and off. I'm definitely no electronics expert, so if this is a big no-no, let me know!

Based on my internet research, the transistors act like diodes (most on the protector board at NPN), so I have been trying to checking voltage drop across each of the legs. The way the board is install in the STA-2000, I can't easily access the underside of the board to check the soldering connection or desolder the transistor. I've been trying to avoid removing any boards up to this point.

 

[jaybirdss396]

The protection circuit most likely is doing its job because there is DC on the output of the Left channel. Find where the output of the left channel goes into the relay and check for DC. Most likely something is fried on the Left channel amp board.

 

[kendo581]

Thanks for the tip jaybirdss396,
I'm thinking that something is fried on the L amp curcuit as well. One question, since the relay is open (unless I "manually" close it using TR601), is there no DC going to the L amp board? Or does power go to the amps first, then to the protection curcuit? Sorry for the beginner questions!

 

[kendo581]

I think I've answered my own question. Yes, there is power going into the L amp curcuit. I measured 40V DC coming out of the L amp into the protection curcuit. The R meaures 0V DC. Since the R channel works properly, I assume that I shouldn't have 40V DC coming in from the L channel? Does this L channel voltage indicate a short somewhere on the L amp curcuit?

 

[Hyperion]

I think I've answered my own question. Yes, there is power going into the L amp curcuit. I measured 40V DC coming out of the L amp into the protection curcuit. The R meaures 0V DC. Since the R channel works properly, I assume that I shouldn't have 40V DC coming in from the L channel? Does this L channel voltage indicate a short somewhere on the L amp curcuit?

In your first post you mention a LOUD high pitched noise from the L channel with the relay closed. This is the problem you need to look at as the amplifier seems like it is 'taking off' in other words self-oscillating. If this is what is happening it could very well place a large DC voltage on the output of the L amplifier, thus triggering the protection circuit.

When you are probing the protection circuit with your meter, you are probably loading it just enough to energise the relay. Your meter will appear to be an additional resistor added to the circuit as far as the protection circuit is concerned.

 

[dshoaf]

Yes, if you have pretty much the same voltage as is found on the input to the power amp from the power supply (usually about the same as the voltage on the big filter caps - +-49v according to the schematic), then you've likely got output transistor(s) shorted. This means that the drivers are also likely candidates for replacement. Don't forget to check the emitter resistors for the output devices as well.

As for the intermittent popping noises you heard that kicked out the relay, those still need to be taken seriously even after you have diagnosed what devices are, indeed, bad at this point. In other words, diagnose and remedy what you can detect as being bad but beware that there may be other failures lurking until you can confirm the amp remains stable after getting the immediate failures out of the way. Let's just hope that the intermittent failures heard first were related to the failed devices.

This also means that the protection relay circuit was doing its job - yea!

Cheers,

David

 

[kendo581]

Thanks for the responses Hyperion and dshoaf,

Just to clarify since I'm still new to this:

The output transistors are on the amp curcuit located where the terminals are (TR509, 519 and 521 on the L amp or TR510, 519 and 520 on R amp)?
What are you referring to when you say drivers?
And are the "emittor resistors for the output device" the E leg on the transistors above (ie. TR509, 519 and 521)? Or something else?

Apologies if I'm a little slow at this. Thanks again!

 

[kendo581]

So a quick update...

I tested all the output transistors on the L amp (driver) board and compared them to the output transistors on the R amp board (which is working fine). It appears that a T03-type transistor is faulty (C to B connection allowing current flow in both directions...)

This faulty L amp transistor (NEC 2SB539) is located on the L amp heatsink along with another T03-type transistor (NEC 2SD287); both of these are listed in the service manual. Interestingly, the R amp heatsink transistors are a different model (SYL ECG285 and SYL ECG284, respectively), which I think is a Sylvania/Phillips ECG transistor. The SYL transistors have higher rating for max. voltage and power dissipation:
http://alltransistors.com/transistor.php?transistor=33553
http://alltransistors.com/transistor.php?transistor=10494

Could this be why the NEC broke and SYL didn't? Is this common to have different parts in one amp v. the other? Also, some of the caps and transistors on the board appear to different as well (SLY vs. NEC again).

My next step is to decide whether to try and replace these transistors, which would require quite a bit of desoldering/disassembly. Any thoughts on this? Should I replace with the same parts that are on the R amp? I've include some pics of the transistors/amp boards highlighting some of different components.
http://s1265.photobucket.com/albums/jj506/kendo581/Realistic STA-2000/

 

[dshoaf]

Excellent for posing pictures - they help the most.

The EGC parts are replacements indicating that a repair was done earlier on that channel. If it is still working, no need to mess with it. I'd not recommend you start swapping parts between the channels. This is a good way to make sure the Right channel will fail along with the Left.

On the Left channel, then, you've found only 1 bad transistor? Look at the schematic for the 2 output transistors and see that there are 2 driver transistors connected to them. These are highly suspect as well since they are direct coupled to the output devices. With the big TO-3 transistors removed, check the drivers for shorts. If you do not catch all the bad transistors, then just replacing the output devices will result in immediate failure of them again. As for the caps and other devices you see that are different, don't worry about them until the Left channel is up and running again. I see that the Right channel has had some of them replaced already.

I'm sure some of the other guys who work on Realistics and other models will recognize the 2SD539 device and can recommend replacements. I'm not up on all that these days. My experience goes back to the 70s when you could still get the originals.

BTW, on the power amp boards, those big white ceramic blocks are the emitter resistors I mentioned. Make sure they're still in spec.

Cheers,

David