Nakamichi PA7II Blowing output transistors.

[gto127]

Hello everyone. I have a Nak Pa7II which is immediately blowing Q 116 & Q117 in the output stage. I just replace all 18 transistors along with any 1 ohm resistors that were out of tolerance. I was thinking it may have aproblem in the bias but not sure where to start. I have a schematic but the bias current is not listed. I have the pot at center position which is the same as the left side of amp. Also I ordered the transistors from Mat electronics & almost every one of them had different lot numbers. How important is matching transistors in this amp?

[Nelson Pass]

You have to fire it up slowly to find the fault before the
output stage blows. Could be the drivers or the bias
circuit.

[gto127]

Thanks for reply Mr Pass. Would you happen to know any probable high failure parts I could check before firing it up? Also would you please explain the difference between this Stasis design & a regular AB amp. I have the brochure for this amp but don't totally understand the concept.

[anatech]

Hi gto127,
I did warranty service on those amplifiers in Canada. Not one died under warranty and the only PA-7 that died had been shorted by one of those large copper box staples. The staple did not survive the ordeal.

As Nelson has said, you must power these amplifiers up slowly to avoid causing extra damage. There is an inrush current limiting resistor you can blow if you try hard enough. You would normally short that with a clip lead to prevent damage when using a variac.

Getting down to this repair, there are no parts that are more apt to fail than any other. So the first question must be, why did it fail to begin with? What happened? What I can tell you is that blindly replacing transistors will not result in a successful repair. There may very well be other failed components, although this design tends to limit damage.

What is Stasis? As far as you are concerned, the output stage is not included in the feedback loop. It must run at a higher bias current to be more linear. The feedback comes from a point just back of the drivers (from memory). This information doesn't really help you with your problem though. You still approach this repair as you would any other amplifier.

This is a really nice amplifier, I like it anyway. Treat it well and do careful work. One last comment. When replacing transistors, you should completely clean the insulators and all mounting surfaces (even the new transistors). Use fresh thermal compound and preferably, new insulators. Make sure you keep the area super clean. Transistors should never be "cranked down". There is a specific torque that should be used, but the parts will be snug down and not be able to rotate on the mounting screw. Tighten just enough to achieve this and no more.

-Chris

[Shaun]

Originally Posted by gto127

Would you happen to know any probable high failure parts I could check before firing it up?

First thing is to know what caused the amp to go faulty in the first place, of course. The only time I have experienced new transistors blowing up at startup was when they were counterfeit devices. It might help to ask the retailer why their transistors are from so many different batches; it's not a good indication.

Sound advice given so far...

[gto127]

The symptoms before amp failure according to the original owner was that the right side had been running much hotter that the left for about a year then finally one day just went out. I sure wasn't crazy about the transistors being so mismatched. Using a diode check they measured slightly different from each other more than I think should be normal. example one measured .456 another .6. These were out of circut measurement before replacing them. I wish some of the better retailers such as digikey & mouser still carried these items. I'm going to have to borry a friends higher amp variac before going any further with the repair. Mine only goes to 5 amps. Thanks for all advice so far.

[Shaun]

Perhaps you could look at the bias circuit. "Running hot" suggests that it could have been faulty or incorrectly set before it failed.

[anatech]

Hi gto127,
The diode test tells you exactly zero - sorry. At worst, it misleads you. There are ways to test transistors that have been discussed many times before here. You need a transistor tester that is accurate and reliable, I would recommend something like a Heathkit IT-18. It's my best friend and I have 5 others. The IT-18 lives on my bench and I have a spare I bought last year I think. The one I'm using I bought new - way back when ...

Failing that, you can easily built a jig that will allow you to make the measurements you require. It's easy enough to do, but a bit less convenient than a dedicated meter. The beta test on many digital meters is not something I trust, I almost never use it when I'm using a meter that has that function.

You need to know beta for matching transistors, you can also measure across each emitter resistor when it's all installed to see how close they match. These are about 1 ohm in a Stasis design. No problem.

There are questions I have as to why you are trying to repair a high powered amplifier by yourself when you don't understand how to even match outputs, or know what the proper substitution for them would be. It's too nice an amp to learn on, and I'm not convinced you can make a proper repair, I might be wrong here and I hope I am.

What are you using for a meter? 'Scope?

BTW, your 5 ampere variac is enough to run this amp up for testing and bias set.

-Chris

[gto127]

Originally Posted by anatech

Hi gto127,

There are questions I have as to why you are trying to repair a high powered amplifier by yourself when you don't understand how to even match outputs, or know what the proper substitution for them would be. It's too nice an amp to learn on, and I'm not convinced you can make a proper repair, I might be wrong here and I hope I am.

What are you using for a meter? 'Scope?

BTW, your 5 ampere variac is enough to run this amp up for testing and bias set.

-Chris

Well for one there is no one left in the area to do this type of repair.I had a VCR repair shop for years but did very little with amps or receivers. I spent most of my time with DVD player CD players & VCRS. I had someone that repaired the AMPs when they came in. I only did simple repairs with them so I need to build my skills in this area. Second I love Nak equipment & this is my dream amp.I have a OMS7AII,CA5 & a ST7 tuner to go with it. I'm using a Fluke 87 meter & a Hitachi 100mhz scope. I have a sencore TF26 transistor tester but It's shot. That's good news that my variac would work. I was concearned with the inrush current when first turning amp on might peak too much since the amp has a obvious problem.

[anatech]

Hi gto127,
Your test equipment is good, so that's a plus. The Sencore is broken, another plus. Now you can get a proper transistor tester. Yes, I know the Sencore (Cricket?) was expensive, but that company marketed expensive stuff to TV shops.

A curve tracer is overkill and is not easy to interpret until you are very use to them. The Heathkit IT-18 is really good on the bench. If you want more functions, then go for an IT-121 or similar. You must be able to measure rough gain and both leakage modes.

I only did simple repairs with them so I need to build my skills in this area.

Wrong amp for learning on. I have to say that your second reason is even less defensible. Did you at least get the service manuals for these? Keep in mind that service manuals do not go over any information that they deem is common for amplifier service people. Also, it's been my experience that TV repair techs think differently than audio technicians do. Not better or worse - just different. For instance, you would be far better at servicing TVs and VCRs than I would. Knowing my limitations, I do not make any attempt to service these things.

I didn't see you list a distortion analyzer either. Another thing you must have in order to confirm the amplifier is operating properly. Some amplifiers require you to advance the bias until the crossover notch disappears. You can not see this unless you have a THD meter, with your oscilloscope connected to the output of the THD meter. You need to look at the residuals.

I was concearned with the inrush current when first turning amp on might peak too much since the amp has a obvious problem.

I have to ask. Why do you think we are using a variac?

-Chris