subwoofer amp repair

[electrolust]

Had a power outage recently. When the power came back on, I lost the dishwasher and my subwoofer power amp. I am not sure that the power amp was working before the outage. It was in storage for awhile and I had a little downtime so was working on setting it up. It definitely worked when I put it away a few years ago. It is about 15 years old and I used it for about the first 8 years then stored it. It was plugged into a surge suppressor / PDU type device and everything else survived, although everything else are super modern devices, all new (and newly designed) within a year ... so very modern products, maybe all with excellent TVS protection. It's the age of the amp mostly that leaves me unsure whether it was the outage recovery that killed it. Everything is plugged into the PDU "unswitched" or bypassed from PDU control, which maybe also bypasses the surge protection.

The amp is an artison RCC-600-SA. It complements my RCC-600-SM subwoofer pair so I'd like to keep this amp. It drives a pair of 8ohm amps in parallel. It is spec'd for 600ohm RMS into 4ohm and designed to bridge the pair of matching subs. It has some EQ built in which matches the subs, which is why I would like to keep it. You couldn't normally buy the sub and amp separately back when this was on the market.

I do see one obviously damaged component, looks like a ceramic cap. (This also suggests to me it was the power recovery.) I guess I should test the nearby caps and assume they are identical? While I'm there should I replace the electrolytics? There are a number of large ones (10000uF) near the damaged ceramic, then another set of medium looking ones nearby, and lastly maybe a dozen pretty small ones on the input board which also contains phase and low pass filters. None of the electrolytics look bad. I could test them but really at that point why not just replace.

Anything else I should proactively replace, considering the likely cause of failure? What I guess is the mains regulation and a large toroidal transformer both look perfect visually.

[electrolust]

I guess if there is a TVS device somewhere I should replace that?

[dietert1]

The blown part may be a small ceramic capacitor. Sometimes those go bad. If you take them away and replace the mains fuse in case it is blown, does the amplifier turn on? What is the symptom? Without schematic, the repair can be a bit of try and error.
The big power supply capacitors are less prone to getting dry than the small ones. I would do the caps after the repair, in order to avoid a difficult to solve double error situation.

Regards, Dieter

[TheMG]

That ceramic capacitor looks like it burnt up a long time ago, I see some hair and dust in the burn hole, which most certainly has ended up there after the event in which it burnt up.

Anything else I should proactively replace, considering the likely cause of failure? What I guess is the mains regulation and a large toroidal transformer both look perfect visually.

Don't "proactively" replace anything just yet. Troubleshoot!

Get your DMM and start measuring. First, is that ceramic capacitor shorted or has it blown open-circuit? If it's open, it's likely not causing the problem (based on its location, its purpose is likely for a bit of extra high frequency bypassing in the power supply, not a critical function). It should eventually be replaced as it is obviously faulty, but may not necessarily be the cause of the amp not turning on.

Fuses - have you looked around for fuses and tested if they are any good? If fuses are good, start checking voltages. Where do you have power and where do you not. Analyze the basics of how the circuit is probably supposed to work. For instance, is there a standby supply and relay or is the main toroidal transformer supposed to always be energized?

[electrolust]

Thanks, very helpful. Yeah I now realize it's likely old damage. The fuse should have popped if it was a line surge.

OK will start with some basic troubleshooting.

I maybe should have noted first, the amp does power on, including in response to 12v trigger if I put it in that mode. It just doesn't produce output. There's also a signal trigger which I didn't think about trying so I'll make sure that part works (or not) also, which might help narrow it down.

[electrolust]

So, I have a "spare" 400w x 7ch amp. It's also of 20 year vintage, but it was in service earlier this year, doing just fine. I figure I'll put 2 of the channels to use with the subs, then I have all the time in the world to do the repair on the first amp.

Now, I have 2 amps to repair ...

With the power toggled off, I start plugging it in and hear a relay click. Ok that's odd. I pulled teh plug before fully inserting it. Double check (it's a 3 way toggle), yep it's off. Plug it in fully, relay click. 2s later it lets out the magic smoke. ARGH.

119.4 VAC at the plug with no load. So probably not some crazy accidental 240v or anything weird from the PDU. I'll try a load of some sort later before I try an amp again.

[electrolust]

The power supply board appears to be bad. (This is the board with the blown cap.) It does give a low voltage out to the other boards, and an internal fan, but there are +/- 45V rails that are dead. Very hard to extract it for further testing. A bunch of wires, including the split 45V wires, are not connectorized. They are through hole soldered with glue or epoxy to fix the wire. I can get the board out far enough to twist it 90* and have access to both sides for the component replacement.

[electrolust]

I found that the power supply board has this large bridge rectifier. It looks to take the AC post-transformer and split it into + and -45V rails.

I expected to find some AC voltage on the two ~ terminals of the diode bridge but did not. So I'm thinking that the front panel is either not correctly signalling the power supply board, or the relay/signal circuit on the power supply board blew up.

There is this 5 pin terminal on the front panel that has power. It is directly connected to the power supply board. It's the only direct connection. If I turn the amp on from the front panel, the internal fan comes on. The fan is connected to a 2 pin terminal right next to the 5 pin terminal on the power supply board.

The front panel has connections to both the input board and the power supply board. So indirectly through the input board and through a dsp board, it has these additional connections to the power supply board, but only a single 5 pin cable that is directly going to the power supply board.

On the 5 pin terminal I found 12.0, 5.0, GND, 0.7, 10.3 volts. When I turn the front panel on, the 0.7 drops to 0.1 and the 10.3 drops to 9.3. The other 2 voltages stay the same. A 5v fan turns on as well. That led me to find the other 2 small diode bridges that filter 5v and 12v power, they seem to work with the power switch.

So I was thinking there are 3 transistors that turn on the 3 diode bridges? And one of them is bad? I'm thinking the 10.3 power rail should be dropping to 0? If I'm headed in the right direction, should I be able to visibly see a bad transistor? They all look fine.

But then I identified the two 3-pin terminals to and from the main xformer. There are only those 2 terminals (makes sense). The "input" terminal always has 120V, no matter whether the power is on or off. The output terminal always has 0V. If there is input voltage into this large toroidal transformer, shouldn't there always be output voltage? Now I'm wondering if I should try applying 90VAC to the output terminals, bypassing the xformer, thus identifying a bad xformer. I'm hard pressed to understand how it could have gone bad with no visible damage. It's wrapped in paper and saran wrap and looks perfect.

Another thing that would help is being able to release these very low profile connectors (last photo, J4). There are a number of them. It prevents me from removing the power supply board entirely. Is there a special tool? I think I'm lifting all the fingers but the connector is still not coming apart, so either I'm not getting them all or my makeshift way of doing it doesn't release them properly.

[electrolust]

forum isn't allowing me to post all the pics in a single reply

[electrolust]

transformer

[electrolust]

transformer showing label. interesting that the legend for the rails are +/-45V but the transformer says +/-35V

[SpecialK]

I don't think you can release J4.  It probably isn't a detachable connector.  I see these from time to time.  Generally the other end of the cable will detach. 

[SpecialK]

On the 5 pin terminal I found 12.0, 5.0, GND, 0.7, 10.3 volts. When I turn the front panel on, the 0.7 drops to 0.1 and the 10.3 drops to 9.3. The other 2 voltages stay the same. A 5v fan turns on as well. That led me to find the other 2 small diode bridges that filter 5v and 12v power, they seem to work with the power switch.

Can you measure +15V on the 7815 and -15V on the 7915?  Can you test the small bridge rectifiers and the large one two?

I would guess the relay isn't switching to allow the big toroidal tansformer to get AC.

[CaptDon]

It is kind of unlikely a transformer of that capacity has gone bad. If you can verify that it does in fact have 120vac on its primary then look at the output. It is surely center tapped  35-0-35. If you have no output voltage by measuring each side 35v to the center or 70 side to side then you have something wrong with the transformer. Perhaps measure the resistance (with power off of course) I would suspect the primary to be under 10 ohms and the secondaries under 2 ohms. If the resistance is much higher or even open you found your problem. I suspect there may be an open in the center tap of the secondary which could cause no D.C. voltages. Look for bad solder or burned circuit board at the transformer connections. Do you know why you get 45vdc from 35vac? How about the 1.414 peak factor of 35vrms? Answer 49 volts no load!!!!

[electrolust]

On the 5 pin terminal I found 12.0, 5.0, GND, 0.7, 10.3 volts. When I turn the front panel on, the 0.7 drops to 0.1 and the 10.3 drops to 9.3. The other 2 voltages stay the same. A 5v fan turns on as well. That led me to find the other 2 small diode bridges that filter 5v and 12v power, they seem to work with the power switch.

Can you measure +15V on the 7815 and -15V on the 7915?  Can you test the small bridge rectifiers and the large one two?

I would guess the relay isn't switching to allow the big toroidal tansformer to get AC.

I'll test the VRs next. I tested the 3 rectifiers. The two small ones work but the large one doesn't get AC power. AFAICT the transformer ALWAYS has power as long as the mains is plugged in. even when the hard power switch in the back is off. Is that a proper configuration? Maybe it keeps the DSP board ready to go at all times, to avoid long startup time.

I'll also check the transformer. If it is open or otherwise bad, is that a part I can readily substitute? It has a part number 13020133 and a name 'sub1' so it seems like an internal part number probably custom?

[electrolust]

oh one more question. this was a family of amplifiers of different wattages. Any thoughts on if it is likely that the transformer from a lesser wattage version would be a drop-in replacement? I mean if you were buying off the shelf I'm sure they would be different. But if you're making them in house maybe you'd just make them all of the larger variety for inventory and production reasons? The name 'sub1' on the sticker seems pretty generic ...

[SpecialK]

My guess for theory of operations is thus:

Small transformer has many taps.  This is always on and creates a few low voltage supplies from these taps using the small round rectifiers. +15V, -15V chiefly, then other regulators use these supplies to create 5V and such.  All chips would use these supply voltages.

When switching from standby, the contacts allow AC to the large toroidal transformer.  This would create the +/- 45VDC for the power amp. Otherwise the unit sleeps to conserve electricity and be efficient.

The 10.3V seems like it might be a 12V or 15V being loaded down a bit too heavily.

[electrolust]

It is kind of unlikely a transformer of that capacity has gone bad. If you can verify that it does in fact have 120vac on its primary then look at the output. It is surely center tapped  35-0-35. If you have no output voltage by measuring each side 35v to the center or 70 side to side then you have something wrong with the transformer. Perhaps measure the resistance (with power off of course) I would suspect the primary to be under 10 ohms and the secondaries under 2 ohms. If the resistance is much higher or even open you found your problem. I suspect there may be an open in the center tap of the secondary which could cause no D.C. voltages. Look for bad solder or burned circuit board at the transformer connections. Do you know why you get 45vdc from 35vac? How about the 1.414 peak factor of 35vrms? Answer 49 volts no load!!!!

main transformer resistance is primary 2.85 / 1.4, secondary 0.4 / 0.25. Those values are side-to-side / side-to-center. So that part seems good.

[strawberry]

by construction one of those cheap amps. so don't expect any kind TVS or other protection
does it have microcontroller that is powered from standby transformer and all voltages are good/shorted then go to garbage bin
if not then is there voltage at relay supply. and what is doing its control transistor

reduce picture resolution to 1080x720pix(~200kB/0.2MB), no one needs 12Mpix images

[electrolust]

But then I identified the two 3-pin terminals to and from the main xformer. There are only those 2 terminals (makes sense). The "input" terminal always has 120V, no matter whether the power is on or off.

Above, I was referring to the input to the main xformer primary. I don't know why I thought that was always hot. It isn't.

Can you measure +15V on the 7815 and -15V on the 7915?  Can you test the small bridge rectifiers and the large one two?

I would guess the relay isn't switching to allow the big toroidal tansformer to get AC.

I do measure +15v and -15v. Those feed the input board. The small bridge rectifiers work and the large one does not (no AC input).

I did some visual tracing and a little more probing of the power supply board and yeah that relay must be the culprit. Either the front panel is somehow bad or that relay is bad. That let me to finding a lot of diags of Alesis RA series amps that use this same relay and it goes bad. Perhaps the speaker out got shorted and the protection is that the relay blows. I suppose it makes sense that I also blew a filter cap along the way.

Testing the relay will be annoying because I can't easily extract the PS board from the chassis. I can flip it through 180* to operate on it but I can't power it at the same time. I'll have to solder on some wires for probing. I guess not that bad. I went ahead and ordered an equivalent replacement part in advance as I now have reasonable confidence.

Will I be able to desolder this with just an iron? It's an 8 lead part, 6 leads together 2x3 grid 5mm spacing, and 2 more physically way "over there". I've never tried to remove such a large part. Or will I really need to get hot air on it. I don't currently have hot air.

Original: Dong Woo DW321-D12S
Replacement: NTE R25-11D10-12

Thanks for sticking with me through this long troubleshooting. I'm in a busy time and only get short breaks to get back to it.

[abdulbadii]

seems forget the indispensable one, a quality ESR meter?

[electrolust]

So I was finally able to get back to this. Wow, over 120 days ago per the necrothread warning lol.

I successfully tested the relay in situ and it was bad. I removed it and retested and yes it is bad. Yay. Confirmed testing by comparing to a new part.

I then easily removed the blown filter cap. The adjacent one (and all 4 others) are marked 104m 100v but I wanted to test the one next to it and it wouldn't register on the LCR meter (portable agilent cheapo). hmm, so I remove the good adjacent cap but lifted the pad in the process.  (see photo)

I then tested the cap separately and it read spot on 100n. That's when I realized the reason i got no reading in circuit is because all the filter caps are in parallel, plus whatever else might be in circuit. The cap body was slightly damaged anyway and I don't know how much was due to my removing it, vs perhaps it picking up collateral damage from its neighbor.

So to repair the lifted pad I figure I'll attach a wire to the next cap over. They are in parallel anyway. If I probe with the DMM pointy tip I do get connectivity, so I guess if I get solder on the lead it would hold it well enough, but I guess better safe than sorry. I wish there were a way to test the others in circuit, so I could have more assurance that they are still good. With my level of skill, it's too risky to remove them.

I have 50V caps in my inventory but sadly not 100V so I'll have to order. Hate to make such small orders.

[electrolust]

The cap markings are

TM
104M
100V

Is TM a manufacturer abbreviation? The cap is the matte brown circular pancake body type. Digikey doesn't have a cap of this spec with that physical appearance. I suppose I can use anything that physically fits.

[EHT]

104 would be 100n, 10 * 10^4 pF.
I haven't really been able to follow the rest though, maybe you could draw the schematic out for the AC part of the PSU. Why have you drilled new holes in the PCB?

[electrolust]

I didn't drill new holes. After removing the parts solder reflowed into the holes. I drilled them through.