Motor transistors keep dying on me

[EmilFrid]

Hi!

I have an issue with my Teac 90-16 that is driving me insane. The transistors driving the take up and supply motor keep dying. I today managed to rewind (without load) for a minute or less before yet another transistor died. Take up is even worse. Few seconds. Starting to feel a bit hopeless as I don't really find anything out of the ordinary. Collector measures 150 VDC and the control voltages feeding the bases of the transistors read 1-4 volts depending on mode. No shorts (except for the one emanating from a broken transistor), I just tried measuring the voltage feeding the bases to see if there was any spikes but couldn't find anything.

Anyone know how to fix this, or at least present a different perspective as I'm starting to feel blind?

Thanks

 

[radardoug]

There is a suppression network R and C across those transistors. Have you checked that? On old Ampex MM1200s, the motor transistors died regularly because they were not rated high enough. Swapping to a video deflection transistor stopped the failures. So I would be looking for a more rugged transistor.

 

[EmilFrid]

Thanks @radardoug , I actually changed to a transistor that could handle a lot more, but it didn't help. I also changed the caps in the suppression network. But here is an odd thing. With the Collector disconnected I measure 300 volts on the Collector cable and as soon as I start measuring it slowly drops (maybe for a minute or so) to 150 volts. I found that strange. I've checked connections to the suppression network and everything is where it's supposed to be.


If it is of any help, the failing junction here is usually the base-emitter

 

[radardoug]

You should only have 100 volts ac supply, so you should never get 300 volts.

 

[EmilFrid]

@radardoug yes, I found it strange. It should be 100VAC rectified, after it stops discharging I'm left with 150VDC which is more in the ballpark than the initial 300VDC. I did use the BUX48a which should be able to handle that though

 

[JohnRoberts]

Semiconductors generally fail from overheating or over-voltage. 5W emitter resistors suggest some power in the area. If not overheated (literally melting the silicon) voltage spikes could punch shorts between collector emitter. Can I ASSume the bad transistors measure shorted collector-emitter? Less common failure modes are open bases (from excessive base current). Back in the 60s as a junior technician I would grind the tops off metal can transistors to look for voltage spike punch throughs (visible as a dot where the collector-emitter is shorted, with some local melting around the dot from the heat caused by that short. )

I see a RC spike snubber, and a CR snubber (?), across the transistors. Dried out caps in the snubber circuit could diminish effectiveness. I am not even sure what a CR component is maybe some kind of capacitor?

Good luck

JR

 

[ruffrecords]

What was the originally specified transistor? Did it include a catch diode?

Cheers

Ian

 

[EmilFrid]

@JohnRoberts
Actually it measures short (or rather close to short) base-emitter. I haven't seen that before. And yes, the snubber is a small value capacitor and I tried replacing them but without any success. I also replaced the caps in the RC network.

@ruffrecords
The original transistor is 2SC1114, and it does not include a catch diode.

 

[JohnRoberts]

@JohnRoberts
Actually it measures short (or rather close to short) base-emitter. I haven't seen that before. And yes, the snubber is a small value capacitor and I tried replacing them but without any success. I also replaced the caps in the RC network.

@ruffrecords
The original transistor is 2SC1114, and it does not include a catch diode.

I am not very familiar with a short between base-emitter fault. Does this mean that the collector is open circuit?

Does anything look overheated in the area?

JR

 

[EmilFrid]

@JohnRoberts hmm I can't remember right now, but i don't think the collector was open. I will have a look tomorrow. Nothing looks overheated and the fault is consistent with all 8 (!) transistors I've replaced. I even tried to connect a variable resistor in series with the base to see if the problem was located there, but as soon as I dialed it in at roughly the correct torque the transistor failed again.

 

[JohnRoberts]

If the fault was overheating it could exhibit a short base-collector-emitter.

JR

 

[EmilFrid]

What I remember from the measurements right now is that the collector didn't short to anything but the emitter and base were shorted, or rather showing around 200 ohms of resistance in both directions. I will write down some proper measurements tomorrow and present it here. This issue confuses the hell out of me

 

[Evertide]

All 8 diodes ( D20 - D27 ) are OK ?

 

[EmilFrid]

@Evertide i will check that tomorrow and get back. I have to get some sleep now. What I find peculiar is that both transistors fail in the same way.

 

[CJ]

What part number and where did you get them? (working the counterfeit angle)

 

[Cqwet Dbdfte]

The design was likely proven long time ago, so I'd look for parts failures.
Inductive loads generate interesting back EMF, and a rectified 100V AC is an easy 141V, plus back EMF.
Motor insulation, shorts to case, shorts inside windings, shorted or dried out caps, open resistors etc. Test motor separately, check voltages with a scope.
The power transistor insulation is an issue. Driver transistor may also get wiped out in the process.
The split-phase caps on the motor need to be checked, along with every single part in the circuit.
Using beefier BJTs should be considered. These are acting as variable resistors absorbing all the junk those motors produce.
Look for higher voltage parts, 600V and up. If motors are shorted, they may damage any transistor.

 

[EmilFrid]

@CJ here's the part:

@Cqwet Dbdfte

Thanks for the reply. Oy vey, that's a lot of stuff to check. I already know the motors are fine and there are no shorts to case anywhere. That was my first thought. Maybe motor caps are the problem? And why does my meter read 300VDC that gradually climbs down to 150VDC on the collector cable when it isn't connected to the transistor? Confuses me a lot.

And @JohnRoberts I just measured the transistor and it shows 100 ohm in both directions base to emitter and a normal continuity between base and collector.

@Evertide the diodes measures ok.

I'm thinking out loud. Would adding a catch diode solve the problem? What I find strange is that the short is between base and emitter. That must have to do with it getting reverse biased, because the maximum voltage that the base reaches is 3.8VDC and the transistor should be able to handle this

 

[EmilFrid]

...and here's the other transistor. Most of the BUX48's that I've installed are of this kind

 

[sahib]

I think we are shooting in the dark here. Can you post the schematic in page 9-7 for the motor connection? I find that schematic alone confusing . Is that circle supposed to be the motor? Whatever it is the two tabs are shown shorted which makes it and the capacitor above useless.

I am not even sure what a CR component is maybe some kind of capacitor?

Good luck

JR

That is a capacitor with integral shunt resistor in the same package.

 

[EmilFrid]

Yes @sahib the circles are the motors. Here's the diagram you asked for: