Phil:
Check how the voltage regulation takes place... it could be either a tl431
and an opto or it could be from the aux winding that feeds the chip... if
the diode on the aux winding is poorly (open) this may cause the op voltage
to exceed the PIV of the diode... and also check the snubber RC across the
flyback fet... open circuit R is also a possibility.

Steve

-----Original Message-----
From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of
Philip Pemberton
Sent: 11 June 2010 21:35
To: Microcontroller discussion list - Public.
Subject: [EE]: SMPSU repair -- secondary schottky shorts and similar stuff

Hi guys,

(Pardon the alliteration in the subject, I think I'm suffering from 
excessive solder smoke exposure)

Here's a "tale from the trenches" for you...

I have -- on my bench -- a "Nichimen" branded DVR. It belongs to my 
grandfather, who is quite attached to it (and the stuff he's recorded on 
it). They don't sell 'em any more (gee, I wonder why) and he doesn't 
want to buy a new one and have to learn how to use it all over again.

So I offered to fix it. Looks like a fairly simple PSU problem -- 
multimeter says 0V on all rails, even though the B+ is up (+330V 
off-line, which is about right for the peak voltage of a 240V AC line). 
The power supply is a cheap-junk little Chinese switcher based off a 
Philips TEA1522 chip.

A bit of debugging revealed a ton of junky "Capxon" and "JPCon" 
capacitors, which were all duly replaced with Matsushita-Panasonic FC 
and FM-series low-ESR electrolytics of the same or greater voltage 
rating. I also found a dead Schottky in the 12V rail regulator -- this 
was replaced with a higher-rated substitute.

I'm now in the situation where when I power the thing up, it kicks in 
for about a second, then the output Schottky blows dead-short, and the 
Philips chip goes BANG. Literally -- as in, sparks flying, and its 
output power-FET blows open. There's SUPPOSED to be an over-current 
limiter for the FET built into the chip -- but this never works. I've 
lost all three of my spare chips already. So I guess it's fair to say 
this chip is about as good as everything else I've seen Philips/NXP 
produce: that is to say, it's a pile of steaming crap.

Now onto the real question: what could cause the output Schottky to blow 
shorted in this manner? All that's on the output is a Pi filter:

           SB650         =====
  XFMR >----|>|-----*----^^^^^----*------> Vout
                    |             |
                   ---           ---
                   ###           ###
                    |             |
                   ===           ===

The two capacitors are 1000uH 25V low-ESR electrolytics, the inductor is 
a 10uH "wire round a cylindrical former" thing.

So far the best theory I've come up with is that the design was marginal 
to begin with, and expects that the capacitors have a higher ESR than 
those I've got installed. That seems silly though -- surely you'd want 
the ESR to be as low as possible to keep the ripple down?

Any comments (or enlightenment) would be most appreciated... I've spent 
two days on this thing and it's still being a complete ass. If it wasn't 
for the HV rail it uses to drive the VFD, I'd have probably replaced it 
with one of the 5V/12V mixed-mode PSUs from my junk box...

Thanks,
-- 
Phil.
piclist@philpem.me.uk
http://www.philpem.me.uk/
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist