Late 90's Peavey Transtube Supreme repair (power transistor questions)...

[RJBAudio]

First of all, I'll add that I'm new to this forum but have been involved with music and electronics for about 20 years now and stumbled upon this site while trying to help a friend out with a broken amp.

First inspection of the amp revealed a blown 4 amp line fuse and I determined that one of the DC voltage rails was just about shorted to ground which led me to a faulty power transistor (shorted between the emitter and collector). This amp only uses two power transistors in a push-pull configuration in the output stage and the evidence of solder flux tells me that both have been replaced in the past. The current "bad" transistor has the part number "70473100" (PNP) and it's compliment which measures to be good is a "70483100" (NPN).

I noticed that some online parts sources dedicated to guitar amp parts carry the original or equivalent replacement parts but I came across a thread on this forum where someone planned to use some MJ15015/MJ15016 power transistors as a substitute in a different Peavey head. It seems that these alternative transistors might be a bit more robust and make it less likely that my friend will be bringing the amp back to me any time soon ("10" isn't loud enough for him most of the time) but since I can't find much info on the original parts (only the substitutes) I'm not sure how well the substitution will work or if it has any advantages.

On a side note, (although I don't think it is) is it likely that something else might have caused this transistor to blow and replacing it with a new one might cause the new one to fail as well?

So I guess my main question is... should I just buy the original replacement for the 70473100 PNP transistor or should I go with a new set of the MJ15015/16 substitutes?

[oc disorder]

Welcome to the forum RJBAudio yeah this is a great resource for repairing amps etc !
You may find this handy
Peavey factory semiconductor cross reference
I'm the newbie! And I got questions, many questions!
(the above is the URL !)
posted by Enzo.

I think that you will find the 70483100/70473100 are in fact MJ15015/MJ15016.
Its highly likely the driver transistors have been stressed and need checking too.
Also check the nearby diodes.
The MJ15024/MJ15025 would be an upgrade in "ruggedness" but shouldn't
be necessary.
I gather Peavey U.S. are very helpful and should be able to supply you with schematics and spares.

[Enzo]

PV wil in fact send you the schematics, and oc is right, the MJ15015 and the 70483100 are the same part.

You can buy parts right from Peavey or buy the generics. The MJ15015 won;t be any more robust, it will be the same part. The part didn;t fail because it wasn;t robust enough, it just failed. That amp's brother may work 40 years and never need a transistor. Chances are the outputs are all that is bad, but oc is also correct that you should check the drivers and any resistors associated with the output stages.

With that single pair of output transistors, I recommend replacing the bad one and the "good" one as well.

ANd yes, if you wanted to go heavier duty, the MJ15024/15025 pair would be it.

here's the guide, I'll save you the trouble of finding it in the other thread.

[RJBAudio]

Once again, thank you both for all of the information.

Since there was evidence that both transistors (70473100/70483100 and there are only two in the output stage) had been replaced before and the PNP one was bad now I decided to go with the more "robust" option of the MJ-15024/25 set. I drove over to a local parts distributor (where I usually go for pinball machine electrical parts) and checked to see what they had and picked up a set of equivalent replacements to the MJ-15024/25 set (models NTE388/NTE68) which have much higher voltage/current ratings than the original set. I figured that since this part of the amp seemed to be the weak spot of the design and that my friend really pushes it pretty hard that I would spend a bit extra to make it a bit less prone to failing in that stage.

I checked as much as I could in the power amp section to make sure everything else was good. I got everything together and the amp works great now. I told my friend, who owns the amp, that it was fixed and he was very excited (I only charged him for the price of the parts and he responded with a very nice "You're the man!"). I gave him a rough analogy to help him understand the "upgrade" so to speak by saying that it is like having a car with a specific engine that has a transmission that has failed twice and now the car still has the same engine (power) but a new transmission that is less prone to failure under the same high-stress conditions. Afterwards, I kinda felt bad saying this because I had forgotten that his car isn't on the road right now due to a bad transmission.

On a side note I just picked up a 2,700 watt Carver Sunfire MkIV plate amplifier for my basement home theater subwoofers (15" TC Sounds woofer in sealed enclosure x2 in parallel) from a friend on another forum for $75 with one minor issue (the auto-on circuit requires a stronger signal than usual to turn the amp on, so I'll have to look into it but it is a pretty complex amplifier). The amp sounds great and there was no way that I could have found a better deal on an amp for this role (my brother took the old Ashley PA amp that I was using before).

So all in all it was a pretty good day for me and I really appreciate the help. This seems like a great forum and you'll probably be hearing from me again in the near future (I have an Acoustic 370 bass head that works but has some turn-on static noise that lingers at a lower level during operation... I've looked into it a bit a while back but didn't get very far).

Thanks again!

[Enzo]

RJB - please oh please don't get addicted to the NTE stuff. They are not the real parts. They are not made on the same lines nor by the same processes. In the case of one opposing pair of outputs, OK, but NEVER mix them with other real parts - for example in a powr amp with three outputs per side instead of just one. They won't share current properly nor will they track thermally.

But one very practical consideration is cost. Allied sells MJ15024 and MJ15025 for $3.86 each. They want $12.38 for the NTE388. I shudder to think what they sell them for over the counter.

I recently needed a power triac for a repair, and we wanted it as fast as possible. My local parts store said they had them, so I ran down ther. Only to find they didn't have them they had the NTE cross instead. The $6 from Allied or Mouser was going to cost me about $24 in NTE. I politely declined. WHat on earth could I charge my customer for a part I had to pay $24 for, and still keep a straight face?

DOn;t get caught up in specsmanship either. The stock parts are quite reliable and sturdy. I have replaced many many MJ15024 type parts in amplifiers too. There is no part made that can't fail or be made to fail.

In an amp like this, the power rails are what, about 50v? 55 maybe? That means the parts never see more than about 100v across them, so the difference between 200v and 250v part is meaningless. Likewise, the 100 watts into 4 ohms means they conduct about 4-5 amps at full out.

If those transistors were marginal, we'd see a steady stream of Bandits in the shop with blown outputs. (Bandit and SUpreme share the same board.)

There is also something in transistors called "safe operating area" which refers to the graphic representation of their voltage and current ratings. That 16 amp 250v part cannot do both at once. If it conducts 16 amps, it cannot go near 250v, and likewise if it is to handle 250v, it does so at substantially reduced current. SO either/or, not both.

I suspect what happened to this amp is that one output failed, and someone replaced it. They may not have considered the stress placed on the survivor, and it failed at some later point as a result. SO both were replaced. Quite possibly the second time through they again only replaced the bad one, meaning the replacement from the first repair was now the stressed survivor. That is why I suggested replacing BOTH.

Like all PV solid state amps, the power ratings are realistic, and the amps are designed to run all day at their rated power.

But I am glad you made it work and your friend is happy.

[RJBAudio]

I guess I should have been a bit more patient and ordered the parts from the net. I'm more of a jack of all trades, master of none (except maybe DIY "passive" speakers) type of guy and went by the recommendation of the guy that works the counter at the local parts distributor who considered the NTE parts equivalent to the part numbers that I supplied to him. I paid about a dollar more than what you quoted for that NTE part and I guess that all that I saved is shipping costs. I guess all is well that ends well but hopefully I'll know better the next time. I often order parts from Mouser and have browsed the Newark and Digikey sites but wasn't aware of Allied.

[Enzo]

Allied is a huge parts house, and I buy almost all of my Motorola transistors from them. Motorola is now On Semiconductor. Same parts.

Allied Electronics - Electronic Components Distribution

Your parts guy carries the NTE line of general replacement parts. he looks up what you want and finds the NTE equivalent. The stuff may have ratings that are close, so they are "equivalent." What is he going to say to you anyway, "These won;t work?" Hardly.

Look, it isn't like I think they will blow up. I am on a personal crusade to wean people from NTE. Other techs may not be so strident about it as I am. I just want techs to understand that NTE crosses your part number and then offers the part in their line that is the closest match to the real part. That is very different from being identical to it.

What is real dangerous is when folks find a part, look it up in the NTE guide and assume the specs for the NTE part are the specs for the real part. Your real part may have a rating of - making up some numbers - say 40v and 300ma. NTE crosses it to their part with an 80v rating and 500ma. Their part may work just fine, but you cannot assume that because their part is 80v, 500ma that your original part was too. SO it is always shaky to spec your real parts using the NTE book. yet I know plenty of folks who when faced with an unknown transistor, will look it up in NTE rather than finding a data sheet for the original part online somewhere.

I find that most any small signal transistor I want to cross comes back down the NTE123. Odd that a thousand different transistors all are the same as their one part, maybe? I find that any time I need a sub, I can find one just as well as NTE can from my stock of general transistor types.

In signal transistors it doesn;t matter a lot, but output transistors in high power amplifiers, like a Peavey CS800 or something, it can mean the difference between working or burning up.

ANd I do realize the circumstances for a hobbyist shop are different from a commercial repair shop like mine. I keep drawers full of all those parts across the room from me, the hobbyist often goes out and buys what he needs on a case by case basis.

[tube power]

Hi, I'm new here. I found this thread through a google search of this exact subject.

I am about to replace the 70483100 w 9804
and the 70473100 w 9803.

I am using mj15024
and mj15025.

Would the lower number, 24 go where the 03 went
and the higher, 25 go where the 04 went?

Does it make a difference?

Thank you.

[Steve Conner]

The 9803 and 9804 are the manufacturing date codes. 3rd and 4th week of 1998.

Now in the first post:
"70473100" (PNP) and... "70483100" (NPN). I checked that in the PV cross ref.

The MJ15025 is PNP and the MJ15024 is NPN.

So the MJ15025 replaces the 70473100, the MJ15024 replaces the 70483100.

Yes it makes a difference, if you get it wrong the amp will go up in smoke.

[tube power]

Thanks Steve.

Obviously I have no business even doing this, but I want to salvage the amp if I can,
and no one around here will even look at it for less than $100.
I've done a little soldering and know enough to keep one hand in my pocket.

[tube power]

I have one other question.
I've built a few simple circuits and modded some pedals.
Transistors always had 3 legs, these only have two.
What is the situation in reguard to polarity with these?

[Enzo]

Look at them closely, the two legs are off center. The transistor will only go in the holes one way and still have the screw holes line up. And that third lead is there, it just isn;t a piece of wire. The metal case of the transsistor is the third lead. The two screws complete the circuit to the rest of the board. Also that is why such transistors usually have an insulator between the transistor and the metal heat sink it is mounted on.

[tube power]

Waiting for the new ones to arrive from Hong Kong.
Had to desolder the old ones to see what you meant.
Now it is obvious.
Thanks.

[J M Fahey]

Hong Kong???

[Gtr_tech]

Oof....watch it there. There are tons of fake (counterfiet) power transistors out there right now. I wouldn't order from "hong kong phooey" or any other unknown dealer. Get the parts from a reputable parts dealer (Digi-Key, Mouser, Allied, Newark, etc) or you will find out the hard way about whats going on.

Read this:
Counterfeit Transistors

[tube power]

2x MJ15024 + 2x MJ15025 Audio Power Transistor TO-3 - eBay (item 260631231083 end time Nov-03-10 03:27:40 PDT)

Yup, I took a chance on an otherwise for sure paper-weight.

I'm just tinkering and have more time than money.
Maybe I can salvage it, but as you can see, I'm not qualified, so hey...

[tube power]

Oof....watch it there. There are tons of fake (counterfiet) power transistors out there right now. I wouldn't order from "hong kong phooey" or any other unknown dealer. Get the parts from a reputable parts dealer (Digi-Key, Mouser, Allied, Newark, etc) or you will find out the hard way about whats going on.

Read this:
Counterfeit Transistors

The ones I ordered seem to have some of the signs of fakes, some signs of genuine.
Fortunately they come in sets of two, so I can dissect one or more and if they check
out I can use the others.

I looked around and found that I can get them from authorized dealers for about $10 more. So these may not turn out to be much of a bargain.

Note to self, ask questions from experts before you order...

I will say this, the sellers are excellent Ebayer's so far. Good communication, nice emails
will shipped alert etc.

http://cgi.ebay.com/ws/eBayISAPI.dll...m=260631231083

[Steve Conner]

Yeah, I'm fairly sure those are fakes. The print looks weird, and weren't Motorola already renamed to ON Semi in 1999? The seller seems like a nice guy, but maybe he doesn't know they're fakes.

The MJ15024/5 are overkill for the job, so maybe whatever is in those cans will be OK anyway. But it would be really interesting if you could spare one to cut open.

The real devices should cost about $5-7 each from a reputable US dealer.

[Enzo]

Mouser has both MJ15024/25 in stock at $4.56 each. Discounts start at 25 pieces.

[tube power]

Yeah, I'm fairly sure those are fakes. The print looks weird, and weren't Motorola already renamed to ON Semi in 1999? The seller seems like a nice guy, but maybe he doesn't know they're fakes.

The MJ15024/5 are overkill for the job, so maybe whatever is in those cans will be OK anyway. But it would be really interesting if you could spare one to cut open.

The real devices should cost about $5-7 each from a reputable US dealer.

Find out how nice he is if they are counterfeit and I demand a refund...
Anyway, Ebay might be interested in that if it happens, or not.
I'm definately going to wreck at least one out of curiosity since I doubt I'll use any of these.

Mouser it is, lesson learned, the hard way, again.

[Steve Conner]

I'm curious too. He ships worldwide for free, so I'm going to order a set and test them in our lab.

[J M Fahey]

the sellers are excellent Ebayer's so far. Good communication, nice emails
will shipped alert etc.

Ha ha, sure they are.
Besides, they inserted so many caveats and disclaimers in the ad, that very politely they can show anybody *you* are wrong.
Just a free sample:

working PERFECTLY for ANY RADIO SATELLITE PROJECT

Ahem !! cough cough !! You *aren't* repairing a Radio Satellite, are you?
Nobody challenges your technical level.
On the contrary, we want to help, but using bad parts definitely won't help you.
.
Hi steve:

The MJ15024/5 are overkill for the job, so maybe whatever is in those cans will be OK anyway.

*Maybe*, but I have already repaired amplifiers with fake MJ15003, and been offered fake 2N3773, with the following peculiarity:
The base material is 1/8" thick, against the usual 1mm or less.
They *look* incredibly rugged, that fact was pointed to me by the salesman.
I actually bough one, to be immediately subjected to brain surgery (literally, I opened its "skull" )
Fact is, the (iron) base is so thick, because they do not use the "heat spreader" copper "coin" that's soldered inside.
The chip itself is soldered directly to the iron base, which has 1/5 the heat transmission of copper.
The thick iron base does *not* compensate for the lower conductivity.
I have it somewhere, I'll post a picture.

[tube power]

Can I use this to check these "bad" ones even though I took them out?
It says "In Circuit Tester"
(Got it from grandfather)

I see pnp / npn setting, I get that.
Low/High?
Short/Leakage/Gain?
I get the idea it has to be in the unit to use this?

[tube power]

Ha ha, sure they are.
Besides, they inserted so many caveats and disclaimers in the ad, that very politely they can show anybody *you* are wrong.
Just a free sample:

Ahem !! cough cough !! You *aren't* repairing a Radio Satellite, are you?
Nobody challenges your technical level.
On the contrary, we want to help, but using bad parts definitely won't help you.
.
Hi steve:

*Maybe*, but I have already repaired amplifiers with fake MJ15003, and been offered fake 2N3773, with the following peculiarity:
The base material is 1/8" thick, against the usual 1mm or less.
They *look* incredibly rugged, that fact was pointed to me by the salesman.
I actually bough one, to be immediately subjected to brain surgery (literally, I opened its "skull" )
Fact is, the (iron) base is so thick, because they do not use the "heat spreader" copper "coin" that's soldered inside.
The chip itself is soldered directly to the iron base, which has 1/5 the heat transmission of copper.
The thick iron base does *not* compensate for the lower conductivity.
I have it somewhere, I'll post a picture.

Thanks, I'm with you, how do you argue counterfeit via email with seller and ebay, right?

I'm going to cut the oem open and see what's in there right now...

[Steve Conner]

I've ordered mine, we'll see.

If it were me, I'd not contact Ebay or the seller. I would go straight to my local On Semi sales office and send them pictures, test data etc. If I can convince them, their word would carry more weight with Ebay than mine.

If they don't care any more because the transistors have the old Motorola brand, then it would be time to take it up with Ebay. But I like to think they would care.

I wouldn't contact the seller, don't want to give him time to think up an escape plan. (Except if he's reading this and sends me real MJ15024s printed to look like his alleged fakes. )

I'd also send the pictures and test data to Rod Elliott for his counterfeit transistor site.

I've never seen one of those "in-circuit testers" before, not sure how it works or if it's suitable. I doubt it would do any harm to the transistors, and the front panel suggests that the "short", "leakage" and "gain" modes are for use with the transistor out of circuit.

[tube power]

I'm guessing I would clip black to the case and try the yellow and red on the legs?
I think I read in a description that these are bipolar, so either way?

[tube power]

Hmm...

This is the reading I got on both transistors pretty much exactly the same for each.
Reading stays around there for "short" "leakage" and "gain"

[52 Bill]

You have the leads hooked up wrong. The case is the Collector (red wire), the emitter and collector leads are reversed.

[tube power]

You have the leads hooked up wrong. The case is the Collector (red wire), the emitter and collector leads are reversed.

Thanks.
Do the black and yellow go to either leg?

Anyway it reads a little lower with red on case, 15 on that top scale.
If I switch legs with black and yellow it drops down to 5 just on Leakage and Gain.

[J M Fahey]

Do the black and yellow go to either leg?

Nope, they are clearly marked on the measuring instrument:
E(mitter) Black
B(ase) Yellow
C(ollector) Red.
To each pin/leg you attach the corresponding alligator.

[tube power]

Tough first day of kindergarten.
All readings are around this range.
Readings look ok I guess, anyone think my transistors are not the problem, or am I reading this wrong?

Thanks...

[J M Fahey]

Fine. So far so good.
You are testing DC Gain/Beta, it shows around 20 which is normal for a power transistor.
Please set it to "short" and see what happens.
Or you can recheck with a modern multimeter in the diode check position.
You test continuity between 2 legs at a time:

You should have:

Red to Collector, Black to Emitter: open/over 1999 (no conduction)
Inverting :still no conduction.

Red to Base, Black to Collector *or* to Emitter: in both cases, around 600.
Inverting, in both cases open/no conduction.

Sorry but I'm not familiar with that transistor tester, find the Digital Multimeter more reliable.
Good luck.

[Enzo]

The tester checks for basic transistor function - does it "transist" or not. But the tester does not put real world conditions on it. Will that transistor still work when we try to drag 10 amps through it? Will it still be OK when it has 100v knocking on the door? Note that the tester has no adjustments for voltage or current. It tests that high powr transistor the same way it tests a little 2N3906.

Like most component testers, if it tells you the part is bad, it probably is. But if it tells you a part is good, it only is telling part of the story.

When the tester says "in circuit" that generally just means it uses enough testing voltage to turn on a semiconductor junction.

[tube power]

If this is correct for both pnp and npn...


Then here's what I get...
PNP

NPN

[Steve Conner]

OK, those transistors are f**!?d for sure. Even if you got the pinout wrong, a good transistor shouldn't give a zero reading for any combination of legs.