Texas star dx-350 hdv repair (toasted)

[Eldorado828]

Eldorado828,

I feel your frustration!

Yes, you may have a shorted component that goes from the B+ line to ground, but because this is a Texas Star amp there is another very common possibility.

you may have a component lead on the underside of the board that is too long, or one of those washers that go underneath the board may be shorting the positive trace to ground.

since you know that its before the power switch, that narrows the areas you need to look at.

you're going to have to take the whole amp apart again and remove the PC board and look everything over.

get a small file or other similar implement and file down any component leads on the bottom of the board that look suspect (too long).

also, either glue or solder those washers right to the bottom of the PC board so they won't move around on you when putting the board back in.

its a pain in the butt but welcome to working on Texas Star amps!

hope this helps.
LC

That's good stuff and I appreciate it.
I haven't been back to it yet but hopefully by the weekend.
I will say, when I had it apart the first time I actually did glue the washers and trimmed anything that was sticking below the board for that very reason because there was a lot of reflowing involved. The board had a very minor warp to it so I even added some thin foam spacers in a couple of areas where it kind of just floated or didn't have anything for spacers.

I was pretty meticulous when I put it together but it has crossed my mind that something might have gotten underneath the board.

Keep in mind, the amp was working well and had been under use for 3 hours or so on and off. That's the reason I stray away from the thought of a short underneath but it isn't impossible.

Once I'm back on it I will report back.... but complete disassembly isn't out of the question.

 

[nomadradio]

The hard part is to clip all the protruding lead from the solder side of the board without scalping the foil trace itself, and lifting it from the laminate. Oddly enough this is a bigger hazard with a fresh, sharp diagonal cutter than it is with a used one that's not so sharp anymore. The sharp blade wants to dig into the solder and pull up the foil.

73

 

[Eldorado828]

Update:
I finally had a little time to dig into the amp and I think i may have found the short. @loosecannon and @nomadradio , I think you fellas nailed it but I can't absolutely verify until I get back to it tomorrow.

I'll add a little back story that I think factors into this equation. So when I put the amp to use the day it began having issues, I had installed a fan kit like this one

Now the Texas star is a 3 part aluminum case or housing. The faceplate and chassis bolt to the heatsink via the 4 screws and washers along with the pills and then the top cover screws to it.
These fan kits bolt right to the existing top cover holes and the amp sort of hangs there by the screws and it puts a tension on the housing causing it pull upwards or towards the pcb board.

So in this situation, when I assembled it before I soldered the meter bulb after installing the meter to avoid accidentally touching the bulb housing. Well guess what, there was a tiny solder sag on the bottom side of the board. Add that with the fan kit putting tension on it and I believe after it all warmed up it made contact unexpectedly.

How I've come to this conclusion? A very minor discoloration to the aluminum cabinet right underneath where this solder blob is.
Here's what I'm talking about on the underside of the meter bulb.

Hopefully nothing else took a hit during this issue.

More to come

 

[Eldorado828]

Gentlemen, i have an update and it's what I hoped that it wouldn't be. One of the pills is shorted.

So I proceeded to investigate the last theory on board shorting to ground and it didn't pan out. Next I went on to start pulling parts to test and narrowed it down to the transistors. Pulled the pills and one of them was shorted, go figure..... bias resistors were still good so that's what had me hanging on to hope that it wasn't a pill.

WTH. All voltages were good, bias was fine, pass through and input swr was good... Did I get a weak or bad pill? Any suggestions?

 

[Ranch55]

It can and does happen.....
When installing new power transistors, it is recommended to also install new bias resistors......

 

[Eldorado828]

It can and does happen.....
When installing new power transistors, it is recommended to also install new bias resistors......

I appreciate it, I did in fact replace the bias resistors because they were toasted to begin with. Maybe it was just a bad pill to begin with. I'll try again and see what happens.

 

[nomadradio]

Um it's usually the internal failure of the transistor that roasts the bias resistor. The excess current that damages the resistor is coming OUT of the transistor's base lead and into the resistor. Supposed to be the other way around, pushing current into the transistor's base, not flowing out of it because of an internal short from collector to base.

73

 

[Eldorado828]

Update, I finally had some time to get back to this project and should finish up this week.

So after testing dang near every component on the board I only came up with a bad transistor. Go figure... So I'm replacing the bias components and feedback circuit so I can give it another go. I'll update once I get power to it.

More to come

 

[nomadradio]

I recommend powering it up the first time from a small current-limited power supply, set for a max of around 1 Amp. Keying the relay with a gator lead will reveal the transistors' combined idle current with no drive. And if there is any sort of fault, the power supply will shut down, rather than cause damage. We call it smokeless smoke testing. Adopted this policy years ago after popping a brand-new pair of MRF421 because of a solder bridge.

73

 

[Eldorado828]

Well fellas I got everything put together. Powered up, bias impedance and voltages this time were right where they are supposed to be. I was going to call it good for the night but couldn't help myself so I put rf into it to verify that it is in fact working like it should. I'll do some power checks this evening and push it a little bit to make sure it is going to hold up.

When I installed the previous pills, I had bias and voltage readings that were just a little bit high so now I'm wondering if I missed something.
Could it have been that a little higher bias voltage contributed to the bad pill?

Anyway, I really appreciate everyone who helped me out on this one. I can't thank you enough.

I'll report my final checks soon.

 

[Eldorado828]

Gentlemen, I know this thread has been long and drawn out so this should put it to a close. I finished up the amplifier this evening and put it through its paces so I can say it's definitely doing the job once again. New feedback circuit, added C11,C13 new bias circuit and pills. Pass through and input tunes are great and power is where it should be with hg pills.

While I was in there I also did the power wire upgrade and replaced the so239's to make it "mo betta".

Did I install a faulty pill the first time? Who knows.... but it's working really well with the help of you fellow wwdx forum members.
Thank you @loosecannon
@nomadradio
@Ranch55

Maybe this thread can help someone in the future.

 

[SuperLid]

Gentlemen, I know this thread has been long and drawn out so this should put it to a close. I finished up the amplifier this evening and put it through its paces so I can say it's definitely doing the job once again. New feedback circuit, added C11,C13 new bias circuit and pills. Pass through and input tunes are great and power is where it should be with hg pills.

While I was in there I also did the power wire upgrade and replaced the so239's to make it "mo betta".

Did I install a faulty pill the first time? Who knows.... but it's working really well with the help of you fellow wwdx forum members.
Thank you @loosecannon
@nomadradio
@Ranch55

Maybe this thread can help someone in the future.
View attachment 73849View attachment 73852View attachment 73851

When you say "new" feedback and bias circuit, do you mean new design? Or, replaced? It looks like the factory TS design to me from the photos.

 

[Eldorado828]

When you say "new" feedback and bias circuit, do you mean new design? Or, replaced? It looks like the factory TS design to me from the photos.

I apologize if the wording isn't to your liking, I replaced the parts with new parts.

 

[SuperLid]

I apologize if the wording isn't to your liking, I replaced the parts with new parts.

It's not a matter of semantics. I wanted to know if you changed the circuit.

 

[Ranch55]

Good job.....