On 22/07/2016 19:19, Gareth Magennis wrote:


"N_Cook" wrote in message ...

On 22/07/2016 10:03, wrote:
On Friday, July 22, 2016 at 9:02:43 AM UTC+1, N_Cook wrote:
Shinco DVD-2610 on Elektrotanya
assuming the SJK is an HV diode and 2x 4148 in the pic, then probably
much the same circuit


Right, now I'm back in the lab, turns out this really is just a bypass
cap, it's connected directly accross the main HT cap. I had at first
assumed it was connected to the current sense circuit or thereabouts.

No bypass caps on any of the other circuits using this controller
though, I guess a high voltage 0.1uF will do. (not SMD this time)


Thanks for the schematic links,


Gareth.


I wonder if it was a multi-layer cap in there with metal migration
growth leading to short.
Any other similar caps on there?







Not my photo, but this is a new PCB:
http://i852.photobucket.com/albums/a...1/IMG_8676.jpg

My blown cap is the one in the middle of the bottom of the PCB, below
the 3 parallel sense resistors and between FET TR200 and Tx T200.



Gareth.



I suppose a 400V MLCC can suffer from metal migration as the numerous
50V notorious MLCC failures.
I've not looked for any manufacturers of 400V rated devils, I assume
they exist, should anyone be foolhardy to use them