diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Power Supplies (http://www.diyaudio.com/forums/power-supplies/)
-   -   Strange Behavior w/ Diodes (Heater Supply) (http://www.diyaudio.com/forums/power-supplies/225458-strange-behavior-w-diodes-heater-supply.html)

steve516 11th December 2012 03:33 PM

Strange Behavior w/ Diodes (Heater Supply)
 
So I've finished my 6SN7/6SL7 phono preamp and I'm experiencing a few issues with the PSU.

Namely, the filament supply has been acting funky. I'm using the 6.3V taps on my power transformer into a bridge rectifier using 20V/40A Schottky's with a 10,000uF cap. At first, I was getting 9V unloaded and only 4.85V loaded (definitely too low). I got the transformer out of an old amp from a friend and I've never been sure of the current ratings for the filament windings, so, since it had an additional 6.3V tap, I hooked that up as well (so windings in parallel) and I had an exact 6.3V loaded. Perfect, except that when I turned it on this morning, one of the schottky's literally exploded with a loud pop.

I can only fathom it has to do with inrush current and the parallel windings, but it's not blowing fuses so I'm really at a loss.

Possibly just a bad diode? I will try and replace tonight, I have 10 or so laying around...

Thoughts?
Steve

KatieandDad 11th December 2012 03:39 PM

Try each transformer winding with a purely resistive load to see what you get.

DF96 11th December 2012 04:07 PM

Your heater windings need an AC current rating of about twice the DC current you want to draw. You may then also need a soft start circuit to avoid blowing the diodes on switch on, because the big cap will be almost a short circuit at first and the cold heaters will quickly discharge it.

AC heaters can be a lot less trouble!

steve516 11th December 2012 04:56 PM

Thanks for the tips. I'll try just running on AC, and see how good my twisted pair wiring really is! :P I already am getting a little bit of hum that I haven't been able to nail down, so I'm cautious but it's worth a shot.

SK

Elvee 11th December 2012 05:17 PM

Quote:

Originally Posted by steve516 (Post 3280463)
Perfect, except that when I turned it on this morning, one of the schottky's literally exploded with a loud pop.

What type number of diode is it?
Also, there is a high risk the other diode of the same rail is damaged: semiconductors generally fail short before they open, and this could mean the other diode also had to sustain a short-circuit current.

Quote:

I can only fathom it has to do with inrush current and the parallel windings, but it's not blowing fuses so I'm really at a loss.
It looks like a possible cause, but 10,000uF at such a low voltage is normally not that much for 40A diodes

Quote:

Possibly just a bad diode? I will try and replace tonight, I have 10 or so laying around...
Possibly, but take care to replace both diodes at the same time, otherwise your stock could vanish very quickly...

steve516 11th December 2012 07:36 PM

Quote:

Originally Posted by Elvee (Post 3280626)
What type number of diode is it?
Also, there is a high risk the other diode of the same rail is damaged: semiconductors generally fail short before they open, and this could mean the other diode also had to sustain a short-circuit current.


It looks like a possible cause, but 10,000uF at such a low voltage is normally not that much for 40A diodes


Possibly, but take care to replace both diodes at the same time, otherwise your stock could vanish very quickly...

OK, so I was mistaken about my diodes. They are type 1N5822. So average forward current of 3A should be fine for my tubes, which at 4x 600mA gives 2.4A. 40V, and max surge current of 80A. So it would seem that I had a bad diode? Will replace all four just to be safe and report back on results.

SK

Elvee 11th December 2012 08:34 PM

Quote:

Originally Posted by steve516 (Post 3280813)
OK, so I was mistaken about my diodes. They are type 1N5822.

The picture is quite different: schottky's are not very resilient compared to ordinary Si diodes of the same caliber, and the surge event now looks much more likely.
Quote:

So average forward current of 3A should be fine for my tubes, which at 4x 600mA gives 2.4A. 40V, and max surge current of 80A.
That would mean a ~2 x 3A rating for a bridge, but the power factor of a normal rectifier+cap is also in the region of ~0.5, this means that with 2 x 2.4A you are just within the range, but not much more, and with a low impedance winding, you could easily overstress your diodes, both at the start-up and during steady state.
Ordinary diodes like 1N5402 would have no problem coping with such surges, but your output voltage will be insufficient.
Quote:

So it would seem that I had a bad diode? Will replace all four just to be safe and report back on results.
Now, I am not sure at all. It could be a fluke, but it could as well be recurrent.
If you have enough of them, you can try with 4 fresh ones, but if it happens again within 6 months, switch for larger ones: 10A in schottky seems more than reasonable for this application.

steve516 12th December 2012 03:24 AM

OK, so replaced all four and actually just completely rewired/ soldered the whole filament supply circuit. All is working well in 6.3V land!

Still have some hum in the preamp, not sure where from yet but I'll squash it accordingly...

Thanks for all the help!

Steve


All times are GMT. The time now is 11:36 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2