transistor testing.

[catstrat]

I know you can test transistors with a meter. But what if you have a leaky transistor? Will the meter test show that to?

[The Dude]

short answer: most of the time, yes.
long answer: sometimes not. Sometimes a transistor junction will not show its "true colors" until it gets warm or has a higher voltage applied to it. You can also often find a leaky transistor by checking them on a high resistance scale as opposed to using the diode check function.

[catstrat]

transistor testing

I have a QSC RMX 1450 in here now that blows a fuse on the right side. I'm trying to avoid a fiasco like i had on the Gallion Krueger. The amp has a hard short. When i test the o/p transistors in circuit they all test as shorted across all pins, on both voltage rails. I checked the flyback diodes, that's what i call them, and they test good out of circuit. If i remember right, one shorted O/P trans can make all of them test as shorted. The bulb limiter is a life saver. I guess the next step is remove the o/p trans and test out of circuit. All the filter caps test good in circuit. The owner said he thinks the building,a church, was struck by lightning. I didn't see any damage to the board. Back to my first qustion, leakage occurs after the transistor heats up. If i set my meter to the highest resistance range i can, which on my meter is 20 meg, what should i look for in the case of leakage?

[The Dude]

Yes, normally outputs are paralleled, so you need to pull them and check them out of circuit. The diode check function will find the shorted ones. To look for leakage, on a high resistance scale, test each junction both ways. One direction it will give you some sort of reading. The other direction (probes reversed), it should read at least 10 times the previous reading. If a junction is leaky, it will normally show the same resistance both directions and this type of leakage is not always recognized by the diode check function, depending on the amount of resistance or leakage. Saying that, transistor leakage is not a "common" problem. They normally just short.

Since you have at least some shorted outputs, be sure to check your emitter resistors. They often open when a transistor shorts. Also, check associated driver circuitry. Be thorough. If you miss even a single defective part, you can end up starting over, although the light bulb limiter should help to prevent that.

[catstrat]

Yep . I don't want to go through again. lol. This is valueble info for me. As it stands right now the bet is a shorted o/p transistor. We'll check all those things tomorrow. Thanks.

[catstrat]

Wizard333 just suggested a possible bad filter cap. I'll pull those out tomorrow and test them out of circuit to. Thanks.

[loudthud]

An old Phase Linear schematic has the following tip for finding the shorted transistor. Measure the Collector to Emitter resistance of all the output transistors. The shorted one will measure the lowest and the others will measure two times the emitter resistor value higher. Remove the shorted transistor and check again until the short goes away.

Another way to find the shorted transistor may be easier in some circumstances. Just unsolder one end of all the emitter resistors, then check E to C with the ohm meter.

Always check any of those low value cemet type resistors for opens. They can fail without any outward signs of distress. Some amps these days have Base stopper resistors, a 1/2W or 1/4W resistor in series with each Base. Check those.

[52 Bill]

Wizard333 just suggested a possible bad filter cap. I'll pull those out tomorrow and test them out of circuit to. Thanks.

Are there separate power supplies for each channel? What type of transistor is used in the output bipolar or mosfet?

I ask because each type of transistor tends to have specific failure modes that more common, like E to C shorts in bipolars. Also check your schematic as I know older QSC amps had grounded collectors that can confuse you if you, if you approach it like a more traditional design.

When dealing with parallel outputs, I unsolder one collector and emitter at a time until I find all of the shorted ones. Once the shorted one(s) is removed the amp can be tested to see if there are any other problems before replacing sets of outputs.

I remember that Enzo has a trick for this as well, I'll try searching for it, or maybe he'll chime in here.

[Enzo]

The Enzo trick is what Loudthud already described. Isn;t "my" trick, just one I like to use. When a row of transistors seem shorted, measure E-C, and if they are all measuring about 1 ohm, the one that is half an ohm is probably the shorted one. Pull that one first.

A meter can tell you that two points in a circuit are shorted together. It cannot tell you which part on a parallel is the bad one But there is nothing that will make a shorted part appear not to be. So we measure resistance in circuit, if it appears shorted, then we use further means, but if it does not measure shorted, then the part is not likely shorted. yes, of course there are situations where something only acts shorted at higher voltage. Every tool has its limits. A tech has GOT to know his limitations.


So in all this troubleshootin' and testing, I forgot to count how many transistors we tested. So you have to ask yourself one question: do you feel lucky? Well, ...


Oh... never mind...

[Enzo]

For those following along

QSC RMX 850 RMX 1450 RMX 2450 Service Manual free download, schematics, eeprom, repair info for electronics

Wait for the wording under the sample image to change from processing to Get Manual, then click the words. That is the download button. The manual includes some troubleshooting tips.

The caps can short, but note the caps are not grounded, their cold ends are at the P/S center tap common. So if your emeitters are sshorted to ground, it is far more likely a transistor.

[Jazz P Bass]

Service Manual.

[catstrat]

testing transistors

Good morning everyone. So far ,so good. I pulled each o/p trans out of circuit and 5 out of the 8 test shorted. Question; i read somewhere i should short all the pins with each other before testing. Is ths true? The O/P transistors are 2SA 1943 and 2SC 5200. I believe those are just BJTs. The drivers are MJE15032 and MJE 15033. Both sides have a dual voltage power supply. Someone else on here said some amps have a base stopper resistor, i assume they do the same job as grid stoppers in a tube amp. If they're open then the transistor would go into thermal runaway. Is that right? Of course i need to check the emitter resistors to. At this point i think it's best to replace all 8 o/p tranistors plus the drivers and any bad resistors. Does that sound like a viable plan of attack? I have some transistors in stock already but i think the best thing for me is to order parts as needed for each job. As many types of transistors as they're out there, i just wouldn't have the room to keep all of them. I hope i was able to answer everyone's questions. Thanks so much everyone.

[catstrat]

transistor testing

The other thing, when i pulled the first one,a 2SC 5200, it tested shorted. I tested the other output trans. in circuit and the short was gone. Is that what Enzo and 52 Bill is talking about ? I sided with caution and pulled all of them for test. The question i have, if they don't show a short in circuit, but show a short outside of the circuit, am i missing something here?

[catstrat]

Oh, thanks for the service manual Jazz P.

[Jazz P Bass]

The transistors are in parallel.

If one is shorted they all will appear to be shorted until the bad one(s) is (are) removed.