PL20 REV C board capacitors C3L & C3E went POOF

[62vauxhall]

I thought I'd be buttoning up the amp tonight.

The unused board was mounted, connected and worked in this 400 amplifier about a month ago. The face plate has been off and meters disconnected ever since awaiting the install of a new light board which I did today. Ironically, I just bought a variac and this was the first time I used it. Part of the delay was waiting for the arrival of some white LED's to replace the blue ones which I received yesterday.

All went well, the variac brought the amp up to power and the LED's illuminated. Before putting on the face plate but after attaching the meters, I thought to re-check the bias once again. The amp was powered up for 30 or so minutes then turned off as I connected the DMM for left channel bias check which was a bit low so I adjusted to 350mv. Turned the amp off again, attached DMM for right channel bias check, turned up the variac and as I waited for the reading to stabilize, heard a noise and witnessed a fair bit of vapor being released. The tops of C3L & C3E had opened and vented.

I assume that replacing those two capacitors is not the answer unto itself. Anyone know what may have gone wrong?

 

[Gepetto]

I thought I'd be buttoning up the amp tonight.

The unused board was mounted, connected and worked in this 400 amplifier about a month ago. The face plate has been off and meters disconnected ever since awaiting the install of a new light board which I did today. Ironically, I just bought a variac and this was the first time I used it. Part of the delay was waiting for the arrival of some white LED's to replace the blue ones which I received yesterday.

All went well, the variac brought the amp up to power and the LED's illuminated. Before putting on the face plate but after attaching the meters, I thought to re-check the bias once again. The amp was powered up for 30 or so minutes then turned off as I connected the DMM for left channel bias check which was a bit low so I adjusted to 350mv. Turned the amp off again, attached DMM for right channel bias check, turned up the variac and as I waited for the reading to stabilize, heard a noise and witnessed a fair bit of vapor being released. The tops of C3L & C3E had opened and vented.

I assume that replacing those two capacitors is not the answer unto itself. Anyone know what may have gone wrong?

Yes that means your +20V regulator is not working.

 

[jbeckva]

Yes that means your +20V regulator is not working.

Think it went from 20 to 85 volts?? Yikes that's not good...

 

[62vauxhall]

I'm using my phone trying to locate that +20 volt regulator and have not found anything to identify it. Any chance of getting a part and location number?

 

[laatsch55]

They are the TO-220's in the middle of the board, mirroring each other..along with the TL-431 --TO-92's....also in the middle of the board....

 

[Gepetto]

I'm using my phone trying to locate that +20 volt regulator and have not found anything to identify it. Any chance of getting a part and location number?

Emailed you docs Gary...

 

[WOPL Sniffer]

I am NOT saying that this is what happened, just saying I have seen this before (Premature Failure of Components), or, it could have been a shitty part. I try to make sure it's not ME, when building boards (I have screwed up though). As I had a discussion before, try to use some form of heat sinking boys and girls. You can't do squat about crappy parts, but you can try to minimize the effects of the soldering iron. HEAT KILLS

 

[62vauxhall]

Thanks G, thanks L. When I thought to check voltage to the new LED board, a (-) meter terminal got away from me. In addition to hitting and "interrupting" a trace on the LED board, now fixed, I suspected it might have whacked somewhere on the main board as I thought I heard something at the time. I guess that is what happened. Serves me right for not taping off those terminals and I had no real reason to check that voltage in the first place, just got curious. Wish I'd had that variac a day or two sooner.

 

[WOPL Sniffer]

ouch

 

[laatsch55]

Thanks G, thanks L. When I thought to check voltage to the new LED board, a (-) meter terminal got away from me. In addition to hitting and "interrupting" a trace on the LED board, now fixed, I suspected it might have whacked somewhere on the main board as I thought I heard something at the time. I guess that is what happened. Serves me right for not taping off those terminals and I had no real reason to check that voltage in the first place, just got curious. Wish I'd had that variac a day or two sooner.

Always learnin...

 

[62vauxhall]

Seems I smoked a couple of resistors too, just for good measure.

Thanks G for the parts list. Mouser has on hand all but one which is Q2 - MJE15033G. They show the ship date for new stock as 2017-09-04. That could mean September 4th or April 9th and they're late.

But Digikey have them and I believe I saw Q1 - MJE15032G as being on hand as well. Would obviously prefer to order everything from the same supplier but it is too late at night right now for me to start checking Digikey for all parts. I have 4 hours to sleep and I want all of it.

If Q1 & Q2 are to matched to less than or equal to 500uV Vbe, will a supplier do that? I don't relish the thought of buying a bunch of each and hope that two are close enough. Plus, I have no device that makes that measurement.

 

[laatsch55]

You do not need to match Q1,Q2. They drop the voltage from 80 to 20, then the TL-431, TO-92'S regulate the voltage.

 

[laatsch55]

Check the diodes in that section too.

 

[62vauxhall]

The Digikey site is all but impossible to access via my phone. Too big, won't completely load then I get punted off.

I since learned on the Mouser site (which does load) that their part #512-TL431CLP is obsolete so it's in my interest to get all from Digikey.

They don't seem to carry Xicon resistors but I assume they'll have 10 ohm & 22 ohm 1% metal films by someone else.

Parts list did not say (or I did not see) - 1/2 watt....1watt?

Lee, regarding the diodes. I was reading through Rocuss's 400 thread. Seems to have been something similar going on. So it's the Zener's @ D80 & D81 I'm to inspect? There are a multitude of IN4148 / 49 in the neighbourhood but I have a quantity on hand.

I believe once before you advised diode checking in circuit is sufficient to determine dead or alive? To recap, is that done in DMM diode check mode with the power caps discharged or does one leg need to be pulled?

Would not look forward to multiple diode testing in that manner with this type of board. Had that bad experience before changing relays on a speaker protect board with those through hole sleeves.

 

[laatsch55]

D80, D81 are Zeners, a 1N4148 IS NOT, do not sub that. In circuit is OK for the dead or alive test. Meter in diode check mode, caps discharged and amp powered down...

 

[62vauxhall]

Yes, understood. Only mentioned the non-Zener standard diodes because some are in/near the affected area.

EDIT: Was planning to include some of those Zener diodes with what I order from Digikey. But IF I needed IN4148/IN4149 for direct replacement, I have those. Would not attempt to substitute non-Zener for Zener.

 

[laatsch55]

Understand Gary....

 

[Gepetto]

The Digikey site is all but impossible to access via my phone. Too big, won't completely load then I get punted off.

I since learned on the Mouser site (which does load) that their part #512-TL431CLP is obsolete so it's in my interest to get all from Digikey.

They don't seem to carry Xicon resistors but I assume they'll have 10 ohm & 22 ohm 1% metal films by someone else.

Parts list did not say (or I did not see) - 1/2 watt....1watt?

Lee, regarding the diodes. I was reading through Rocuss's 400 thread. Seems to have been something similar going on. So it's the Zener's @ D80 & D81 I'm to inspect? There are a multitude of IN4148 / 49 in the neighbourhood but I have a quantity on hand.

I believe once before you advised diode checking in circuit is sufficient to determine dead or alive? To recap, is that done in DMM diode check mode with the power caps discharged or does one leg need to be pulled?

Would not look forward to multiple diode testing in that manner with this type of board. Had that bad experience before changing relays on a speaker protect board with those through hole sleeves.

These are the TL431 part numbers:
[TABLE="width: 172"]
<colgroup><col></colgroup><tbody>[TR]
[TD]595-TL431AILP
512-TL431CLP
863-TL431ACLPRAG[/TD]
[/TR]
</tbody>[/TABLE]

These are the Zener part numbers:
[TABLE="width: 172"]
<colgroup><col></colgroup><tbody>[TR]
[TD]78-BZX55B30-TAP[/TD]
[/TR]
</tbody>[/TABLE]

 

[62vauxhall]

Once again, thanks y'all.

Finally spotted where those resistor's wattage is identified. Must split the order between Mouser & Digikey as Mouser does not have one of those MJE devices so getting both from Digikey.

 

[62vauxhall]

These are the TL431 part numbers:
[TABLE="width: 172"]
<tbody>[TR]
[TD]595-TL431AILP
512-TL431CLP
863-TL431ACLPRAG[/TD]
[/TR]
</tbody>[/TABLE]

These are the Zener part numbers:
[TABLE="width: 172"]
<tbody>[TR]
[TD]78-BZX55B30-TAP[/TD]
[/TR]
</tbody>[/TABLE]

Parts are coming. Ordered most of what I will or may need from Mouser except the MJE's which Digikey still have.

Those different TL431 suffixes....they refer to packaging? I wound up requesting the one ending PRAG.

BTW Joe, those bright white LED's look lovely behind the 400's tinted lenses. Still seems odd PL used meter lenses that dark with the amount of illumination the original incandescent bulbs produced. It was hard to tell that the amp was even on in a lit room without looking inside directly at the bulbs. With room light off, a pale glow and not enough to read the scale from more than a foot or two away.