With the center tap connected and the 38v/52v pair connected, I still get squarish waves AC output on the transformer (much lower Vpp) and still have good signals at the MOSFET gates. But with the center tap and the 50v/63v pair connected, the signals on the MOSFET gates are back to junk and there's no more AC on the transformer. So it must be something down stream on the 50/63V side of things. I was not able to disconnect the circuits past the diodes yet, I'll have to take the PCB out when I've got a bigger block of time so I can do that.

Aside from the power inputs, there's not much DC. Some spots have 5V relative the chassis, but not relative the power supply negative (battery -).

I really wish I could find a schematic. I'll ping Phoenix Gold to see if they'll share it. I'm not expecting them to, but there's no harm in asking. In any event, we've really narrowed it down. Next I'll take the PCB out so I can see the traces and have access to start removing things.

Check all your output power transistors. I have had a couple where these went bad. If so, replace them all. Common on this unit.

I wonder if this is going to turn out to be the case. What's the best way to test them?

 

The amp voltages will have no continuity with the supply voltage,

Best way to test them is to unsolder the collector and emmiters, and check for open circuit between them, if its the high voltage thats down, sounds like its the driver stage,first pair of output transistors, which are bolted on the other heatsink. You posted a diagram of the amp in post #1.

 

OK this sounds like the next logical step, I'll let you know what I find!

I had the thought of just ordering a bunch of parts and having them ready, but these old parts are hard to find! And the few I found are expensive, so I guess I'll test them first and then order lol... I guess these old parts just aren't used much these days.

Does anyone have experience with the cross-reference guide on this web site: http://www.weisd.com/ I'm wondering if it's good enough to trust blindly, or do I need to dig out the datasheets and compare manually.

Just for fun, here are the transistors I found on the output side:

B1163
D1718
A1306
C1567-Q
C3299A

 

Dont go ordering anything till you have proved its faulty, the transistors are marked different on the circuit you posted, maybe a different chassis, but the principal is the same, two npns two pnps and a npn C1567 current control.

The B1163 will be npn. The D1718 is pnp. The C1567 /C3299 will be npn, and the A1306 prob pnp.

 

Man, it takes a lot of heat to get these things out! There must be a ton of copper in the PCB. I ended up having to use the hot air and got a couple of the big guys out, without doing too much damage to neighboring parts. I'm open to any tips for extraction when there's a ton of copper sucking your heat away.

Are the big ones the driver stage first pair of transistors that you mentioned (B1163 and D1718)?

Data sheets are hard to find, but it looks like the D1718 is NPN, and the B1163 is PNP. Here is a data sheet for a supposed equivalent.

So far I've extracted 1 each of the B1163 and D1718. On both it is open circuit between the collector and emitter, no matter which way I put my probes. But I do get about 330k and 348k ohm between the base and collector, and 345k and 355k between the base and emitter. Is the open circuit between collector and emitter good or bad?

 

They sound ok to me, you can measure them in circuit between collector and emmiter, or just lift out one leg the middle is usually the collector.

The big ones are the final output stage, the smaller two will be the driver,the centre is the base current controler.

You need the diagram of the amp really.

 

I haven't given up on this! And I think I'm getting somewhere. Do diodes fail often?

Long story short, after removing the PCB and a pile of components and them all testing OK, I started poking around the board. I removed some fuses and still had the problem, so I assumed (no schematic) by the physical location of the fuses that I disconnected the output side, meaning the problem must be in the power supply side itself. So I started poking around and found a diode that measures a perfect short, .2-ohm in both directions. Looks like part of a rectifier. The diodes around it all had just over .48v forward voltage, and open in reverse so they appeared good, but this one is a dead short. It's D7 in the pictures below, I highlighted it. I'll replace that diode and see what happens.

Side note: When viewing the image, you can click the little arrow in the top-right of the screen to get the full resolution picture.

 

Yes they can blow easily, replace it, looks like it is one of a bridge rectifier set up, possibly the negative side,

Keep going with your tenacity!

 

I don't give up easy.

How closely do I need to match the diodes? If I've got one that's similar but not identical is that OK? Or should I replace them all if I can't find the same part number?

The one I removed says MUR 120 030. I'm on the hunt, but like other parts in this old relic, the original doesn't seem to be around anymore.

 

Try mur160, ebay have them, i would replace all of them on that output.

 

Also get yourself a better iron, hakko are very good and cheap too.



http://m.ebay.com/itm/Hakko-FX888D-...06-Heavy-Duty-Cutter-/380911584152?nav=SEARCH

 

I was able to find some mur120 on mouser, but the lowest forward voltage I could find was .875v, and the ones that are in there all measured around .48v. Would that be a significant difference? The mur160 all had much better reverse voltage limits, but an even higher forward voltage at 1.25v or more. Then again, the spec sheets indicate forward voltage @1A, and I'm sure my meter is giving far less than that so maybe my comparison isn't valid? A quick search didn't turn up the spec sheet for the originals, but maybe it's not worth the effort if these specs aren't that significant for this case.

Thanks for the iron suggestion. I've been pretty happy with my iron, I'm using this one, and it will melt the solder everywhere if I turn it up hot enough. It's my electric solder sucker that is failing me in the high copper areas, it just can't overcome the heat loss to the copper. Maybe I need to try that spring operated sucker in combination with my iron, instead of trying to use a electric solder sucker. The electric sucker is really nice where it does work.

 

the spec sheets indicate forward voltage @1A

That is the important point. Diodes have more voltage drop with higher currents. Comparing 1 amp to what a meter measures with is a very long distance.

 

Try byv28_200 ultra fast diodes from Vishay

 

Arg, there must be something else bad... I had a diode here, part PTC205, supposedly equivalent to this one, so I dropped it in just to see if I got any output from the amp. The wave forms on the power supply MOSFET gates now look good which is an improvement, but still no audio output and pretty low voltages on the high side of the transformer. I knew it wasn't a perfect match to the other diodes, but I was expecting something on the outputs.

What strikes me as really odd is the input side of the transformer has what seems like a very high + DC bias, and pretty small Vpp. The screen shot below is both inputs to the transformer (low side), they are out of phase which I get, but notice the Vbase is 11.5v. So the center of the waveform is offset about +13v from ground (same as the amplifier input voltage). Should this really have this much +DC bias?

 

Try byv28_200 ultra fast diodes from Vishay

I will try this if I end up ordering them, thanks!

 

New data point. The MOSFETS that feed the low voltage side of the transformer are part number IRFZ44, here's the data sheet. The impedance between the left two pins (gate and drain) is 2.7-ohm, and the diode test shows current will flow in both directions between gate and drain. I'm thinking this is bad, am I correct?

 

Is this test out of circuit, if yes then they are duff.

Otherwise i think they have a gate,drain feedback coil from the pulse transformer.

 

It was out of circuit. I was getting similar readings in-circuit so I decided to take one out to double check, and sure enough bad readings. I wonder if that diode went bad, causing more stress on these MOSFETs causing them to fail. I'm thinking this might explain the hot resistors and the big DC bias on the transformer. I'll order some new transistors and those diodes you recommended. If only mouser had same-day delivery. Stay tuned!

 

I removed all 4 and they all flow current both directions between gate and drain, and they all have low resistance between gate and drain, from 2.7 ohm to about 19 ohm. So it looks like the whole lot is dead. A new set are on the way and they might just arrive this weekend, cross your fingers and toes!