Onkyo TX-DS 575 / Integra DTR 5 in circuit protection mode

The most likely scenario is that it is detecting a DC offset on one of the channels, but if you are new to faultfinding, you will probably struggle to get to the bottom of the problem with any certainty that you can confidently switch back on, without finishing up with another pile of dead components. The reason that you read infinity on the speaker lives, is because these are connected to the output stages via the output relay(s), which will be remaining open, if the unit is in protect.

To check for DC offset, you need to measure at the output transistors' midpoint, which will be at the junction of the two low value emitter resistors (assuming that the amp uses conventional output devices, and not an STK hybrid). At idle, you should have zero volts, or very very close to zero, on this point. If you find any offset on any channel, then it's likely that one or other or both of the output transistors has failed, but that is unlikely to be the end of the story. With a DC coupled amp, the problem can be coming from virtually anywhere right back to the front end, and other transistors besides the outputs, will likely be faulty. If it does use an STK hybrid - and I seem to recall that Onkyos don't - then any DC offset will normally be cured, without any other issues, by simply replacing it. If you don't detect any offset, then the problem is with the protect circuit itself (not unusual), but sometimes, these monitor many parameters, and can be an absolute bitch to troubleshoot.

If you have a full-rail offset on all channels - say +45v - then look for a missing negative rail or vice versa.

I don't mean to put you off, but honestly, if you are new to faultfinding, this kind of project is not something I would recommend to a beginner. Most of us who are involved in repairing this kind of gear daily, bypass the power switching, and use a variac to bring up the supply slowly, so that we can measure what's going on (wrong !!) before all those shiny new transistors fail again ... Good luck with it.

Arfa

Thanks for the concern. It's already a dead pile of parts, so I might as well learn something new. :) Saw the relays. That explains that... The receiver has separate front channel and center/surround channel amp transistor boards. I think it's been narrowed down to the surround board - when I unplug the surround board's rail power, the circuit protection does not kick in. I tried to catch the offset voltages on the surround board before the protection kicks in, but the DMM doesn't register fast enough.

Thanks - I see that (.22 ohm middle tapped resistors after the amp stages, but before the coils). I've got to get external voltage to the surround board rails somehow. The rails are 45V, but would driving them at 12V (two 12V batteries, one for each rail, to give 24V delta) provide enough voltage for a diagnosis?

This receiver has the C5198 and A1941 transistors attached to the heat sinks (no STK's that I can see). It's a 5 channel + sub pre-out, along with separate two channel B speakers (quasi-multi room). The surround output circuit is fed by a LC4966 chip, which is fed by an M62447SP chip (that feeds the front and center amp circuits directly as well). Data sheets show the LC4966 as a quad bilateral switch and the M62447SP as a 6 channele electric volume chip. Hopefully it's something on the transistor board. I can almost follow analog circuits, and get a concept of whats going on... I'd love to power each individual channel, but there is one feed to the entire surround board, and the rest are PCB traces and hard jumpers soldered in place (which I suppose I could unsolder, if thats what I'm supposed to do).

Thanks - will check.

I appreciate the advice, and love a challenge. I wish I had a Variac (I see many flavors on Ebay, for $10=$100+) or knew of someone locally (Farmington, CT) that had one, but I'll have to figure out another way to bring up the supply voltages gradually. It's been a good learning experience so far, and if this works, I've got a bookshelf system the lost its right channel after it was dropped that would be nice to use in the bedroom, and another one for the basement exercise room. I think every point you mentioned generated about 2 hours of googling, schematics, and testing, but I'm more comfortable already, and feel like I'm starting to get the concepts. Thanks again!

Andris

The very first thing I would now do, before going any further, would be simple ohms checks on the output transistors, and those 0.22 ohm emitter reistors that you found. You may well find shorted devices quite quickly. The chances are that only one channel out of the 5 main ones, has failed. This means that you probably have an identical channel that is perfectly OK, and will serve as a reference channel for comparative ohms checks. If you find shorted outputs, initially, just remove them, and re-power the unit. You will probably find that it then comes up OK on the other channels. You can then go ahead and replace the faulty transistors, but do not then repower with full input voltage, as there will almost certainly be other issues that will reblow the outputs.

In order to carry out further checks safely, you will need to get the rails on the output stages down at a very low level. This is where I would be using my variac, but in order to do this, it is first necessary to bypass the main power relay. To do this, simply solder temporary wire links across the relay switch contact pins. If you do not have a variac, a crude but effective way of reducing the input voltage to the unit, is to string 40 watt light bulbs in series - say 3 initially -and in series with the power input. If you feel that you will be going down this route, you might want to give this a try before fitting new output transistors, just to see what sort of levels you get, as this will depend a lot on the idle power consumption of the unit.

Once you have established low rails - say +/- 10v - you can start making voltage checks to ground between similar points on a working channel, and the (possibly) repaired one. If there are any significant differences, chase them down, and replace the faulty components. You may well be looking at driver transistors, coupling resistors to the bases of the output transistors, output transistor emitter resistors ( the 0.22's ) and possibly pre-driver transistors. If all voltages look similar between the reference and repair channels, and the mid-point volts are zero or very close, you can reduce the number of light bulbs by one, to bring the voltage up a bit higher. Always re-check all the voltages each time you increase the input volts, and let it run for at least 5 minutes, checking the output transistor temperatures, all the time.

Trust me, it probably won't be an easy ride, and it's not a very good project to pick as a starter, but you seem to have the enthusiasm and at least a basic understanding of what you're doing, and as you say, you haven't got a lot to lose by having a go. Again, good luck ...

Arfa

The most likely scenario is one or more bad muting transistors. It's a fairly vommon problem on these - usually caused by a mild lightning strike or perhaps a static shock somewhere along the line.

I'll research the parts - there's about 10 transistors involved, and troubleshooting down to the individual bad one(s) is pretty difficult - and repost.

The good news is the parts are cheap.

Mark Z.

Here's your part numbers: The parts are on the U1 board, a vertical pc board - of which there are 2 if memory serves - it's the larger board.

Q3383, Q3684, Q3183 - 221631 Q3483, Q3683, Q3583, Q3283 - 2213632 Q3186 - 2215960 Q3187 - 2215780 Q3188 - 221282

Certainly check for any emitter - collector shorts first. Also, There are substitute part numbers for some of these - Onkyo can fill in the subs when you call them. Their US number is 800-229-1687 - option 5 for parts department.

Mark Z.

Thanks again for the pointers. I checked the .2ohm resistors, and although I wasn't able to check all of them, one may have been shorted. It's in a tough spot to get at with the DMM probe (which may have been grounding out somewhere else), so I'll remove everything from the chassis. That looks to be a pain, but I think if I'm going to end up replacing components, I'll have to do that anyways.

The U1 board is the pre-amplifier board (vertical, on the front channel side of the unit, parallel to the heaat sinks). Your memory is spot on! I found the offending transistors on the schematic, and will add that to the list of things to check. It's odd that the circuit protection would kick in only when the surround board is plugged in. Q3684 was easy to get to, while still int he chassis, and seems "normal" with the DMM. B-E and B-C was .2Mohm's (and .6-.8V in diode mode), and C-E

65KOhms and dropping (I see a 56K and a 1K resistor in series across the C-E circuit). If the muting circuit goes, do all the diodes usually go? I had everything behind a surge protector, but maybe it was junk. Anyways, you guys gave me another evening of things to do! :) Thanks again.

Andris

Just a point. The 0.22's *should* read short circuit or thereabouts - 0.22 ohms is almost a piece of wire ... What you are looking for with those is them having gone open circuit, which they often will if the transistors attached to them, are short circuit ( path = "+" rail, Tr C-E s/c, 0.22R,

0.22R, Tr C-E s/c, "-" rail ) = lots of current between rails, = blown emitter resistors.

As far as your meter probes go, I keep a piece of silicone rubber sleeving over my probes, so that only the last 1mm is exposed. I also keep them filed up to a needle point, which helps to stop them slipping.

Arfa