uh, and i ******* up again, caps and rectifiers got ugly warm - dead?

[demogorgon]

Yes, i suprise myselfe sometimes with how much it's possible to mess up such a simple litle thing as a gc, but i managed to do it again.

what did i do this time?

i grounded the "+" on the rectifier and tried to use the "-" as a source for negative voltage.

what happend?
the rectifier and the main resiervoare cap got dirty hot, and I pulled the plug on the amp.


question:
is the cap and rectifier dead? would it be wise to trie to use them again?
have i killed the lm3886 to at the same time? it's well protected isn't it?
and the opa627's too..

Regards
Marius

 

[carlosfm]

You must correct the error and try again.
Measure everything, including DC-offset, before connecting any speakers.
Oh, you better wear sunglasses before switching on.

 

[demogorgon]

Carlos, are you serious?

correcting the error? trying again?

well, the problem is, i dont know what i did wrong, and i dont see or measure any wires where their not suposed to be..

I was a litle tired when i wrote the previous post, so to clairify:

can i connect the "+" out of one of the rectifiers (hereby refered to as nr1) to the others(nr2) "-" and take negative voltage off the "-" of number 1?

edit: ok, just to impress you all with my morvelouse paint skills..

 

[carlosfm]

edit: ok, just to impress you all with my morvelouse paint skills..

That is correct.
Check it out if you have made it as the sketch.

 

[demogorgon]

now, i checked, and checked again, but this thing is still acting so wierd..

could it be that the cap is shot, and that there is contact to some extent inside it, why in the hell would it get hot otherwhise?

this thing is bugging me out..
the lm ic's seem fine though, getting warm, but nothing more than a lm3886 turning on for the first time is supposed to get..

BAH!!

thanks
marius

 

[demogorgon]

heh, the **** cap was shot as i hoped. now for resoldering it all again

this buffer better be woth it carlos..

 

[demogorgon]

not the cap after all!

i replaced it with a new one, but nothing: still gong warm.
what is cousing this? everything is in order!
every solderjoint to itself, every wire where it's supposed to be..

I have to constantly battle my self and my own crazed desire not to pick it up and smash it to the wall.

can anyone me some pointers as to what to look for?

regards
marius

 

[wintermute]

Sorry to ask this question, but have you double checked the polarity of your caps? (the simple errors are often the hardest to see)

The one on the +ve rail should have the -ve on the zero volts line, and the one on the -ve rail should have the +ve on the zero volts line.

Tony.

 

[Morse]

Hi Demogorgon;

Hmmm, just some ideas, based on my (possibly faulty) understanding of your situation -

1. Are you sure that the parallel primaries of your PS trafo are wired in phase with each other?
2. Did you wire the 2 caps so that the "lower" one has it's "+" attached to the "upper" one's "-" at the system ground? That's the way it should be.
3. V- isn't attached to system ground, is it?

On a circuit like this, having a pair of cheap DMM's with alligator clip leads is nice, since you can measure both secondaries at once and you don't have to worry about reversing phase so much. I picked up a $20 Chinese cheapie DMM to go with my (somewhat better) Chinese cheapie DMM for just this sort of thing....

One last thought is to wire and debug the PS circuit before the amp - that way you won't risk pricey chips un-necessarily.

Good luck!
All the best,
Morse

 

[wintermute]

Just one comment on morse's last suggestion. Make sure that you have a safe way of discharging the PS caps if you don't have a load connected! I normally use a standard 60W house light bulb, with some wires soldered to it.

Tony.

 

[carlosfm]

Sorry to ask this question, but have you double checked the polarity of your caps?

It's very likely that this is the problem.

 

[demogorgon]

--and once again i win the mongo-prize!

you were right guys, i put one of the caps the wrong way. another of those thing you do wrong and just dont think of.

will this kill teh cap?
it seems to have buildt up some internal preasure from its ill treatment..
and now it's quite clear why the rectifier and tranni got warm too.

geez, i suprise myself, i should know what way to put the caps by now..

well, thanks for pointing the way.
lets see if it works this time around.

-marius

 

[wintermute]

I'd say throw it and get another one. You're probably lucky it didn't explode

Otherwise you might think it's ok, and have noise problems etc, also I'd say even if it was working it's life span would be pretty drastically reduced.

Since I would assume you are probably using 35V electro's it shouldn't be too expensive to replace

Tony.

 

[Morse]

Hi Demogorgon;

>>>...will this kill the cap?...<<<

Yep. The oxide layer was damaged by the reverse current it's received (which is why the "pressure built up inside" - that's from a current arc that boiled part of the electrolyte!) Even if it were properly re-formed under controlled current conditions, the spots where it's arced between the plates will always be prone to failure and that cap should NOT be trusted.

Hi Wintermute;

>>>...Just one comment on morse's last suggestion. Make sure that you have a safe way of discharging the PS caps if you don't have a load connected! I normally use a standard 60W house light bulb, with some wires soldered to it...<<<

Excellent point - I always wire bleed resistors in parallel to my PS caps (generally DIRECTLY to the caps...) so it's not an issue here - but it is with anyone who uses the amp circuit to bleed out his caps on shutdown. Definitely bleed off the charge on any PS caps before working on it!

All the best,
Morse

 

[demogorgon]

Since I would assume you are probably using 35V electro's it shouldn't be too expensive to replace

Tony.

nahh, they were 80V 10.000uf caps, and quite expensive ones to boot, so i would have liked to use it again, but now i see why i shouldn't, so i wont.

Hi Demogorgon;

>>>...will this kill the cap?...<<<

Yep. The oxide layer was damaged by the reverse current it's received (which is why the "pressure built up inside" - that's from a current arc that boiled part of the electrolyte!) Even if it were properly re-formed under controlled current conditions, the spots where it's arced between the plates will always be prone to failure and that cap should NOT be trusted.

Hi Wintermute;

>>>...Just one comment on morse's last suggestion. Make sure that you have a safe way of discharging the PS caps if you don't have a load connected! I normally use a standard 60W house light bulb, with some wires soldered to it...<<<

Excellent point - I always wire bleed resistors in parallel to my PS caps (generally DIRECTLY to the caps...) so it's not an issue here - but it is with anyone who uses the amp circuit to bleed out his caps on shutdown. Definitely bleed off the charge on any PS caps before working on it!

All the best,
Morse

thanks for the info, learned something new today as well then!
i have two 4.7k reseistors in paralell on the rails to gnd and i figure carlos put them there for exactly that purpoce.

well, just to uppdate ont eh amp though, a new problem surfaced when i fixed the other one, i have a dc offset of 17.1 and 15.1 on each channel, and 14.5 of the buffers.

guess i should have gone with a kit, not that i ever give up, i dont know my limitations THAT good

 

[carlosfm]

well, just to uppdate ont eh amp though, a new problem surfaced when i fixed the other one, i have a dc offset of 17.1 and 15.1 on each channel, and 14.5 of the buffers.

Do you gave an input resistor to ground, on the buffer's signal input?
Just wondering...

Otherwise, it can be another reversed cap.

 

[wintermute]

Hi Wintermute;

>>>...Just one comment on morse's last suggestion. Make sure that you have a safe way of discharging the PS caps if you don't have a load connected! I normally use a standard 60W house light bulb, with some wires soldered to it...<<<

Excellent point - I always wire bleed resistors in parallel to my PS caps (generally DIRECTLY to the caps...) so it's not an issue here - but it is with anyone who uses the amp circuit to bleed out his caps on shutdown. Definitely bleed off the charge on any PS caps before working on it!

All the best,
Morse

Hi Morse,

having found out about this the hard way myself (was really obvious once I thought about it) thought I'd just mention it to potentially save demogorgon from a rather large fright

Mine was on a +-63V with 18,000 uF/rail PS with no bleeders, temporarily disconnected from the amp to troubleshoot I can't remember if it took a chunk out of my soldering iron, but the blat noise and bright flash was enough to scare the pants of me!!!

demogorgon,

I'll leave the dc offset problem to those more knowlegeable than me

Tony.

 

[Franz G]

it seems to have buildt up some internal preasure from its ill treatment..

Thats exactly, why CarlosFM was not joking: wear sunglasses the first time you try your amp!

Just in case of an exploding cap or chip, to protect your eyes!

Franz

 

[ROVSING]

You could try to unsolder the caps and "load" them with a 9v battery, put on your multimeter and see how fast it "unloads"if the current is falling slowly they will be ok,i actually made the same mistake but none of the capacitors were damaged (12 pieces) they work fine in my clone and again : the sound is amazing ! It probably depends of how long you had dc current on them.

 

[steenoe]

i actually made the same mistake

Oh no Rovsing, not again
Are you ever gonna learn that


Steen.