Screwing PCB to heatsink may cause flexion and damage to ceramic capacitors?

[ocset]

Hello
We have a   SMD_only SMPS   PCB   of 10W which we screw to a flat heatsink  surface. There is a rubbery thermal pad between PCB and heatsink.
The PCB is 6cm x 15cm x 1.6mm thick.  There are five screws which screw the PCB to the heatsink….one in each corner, and one in the middle.
There are multiple  SMD ceramic  capacitors on the PCB. Do you think that as  the screws are tightened up one by one,  that there is a risk of PCB flexion causing some of the ceramic capacitors to break.?

[TJ232]

about 48%

[ocset]

Thanks, thats what  i was worried about, so i wonder how people get round this?
I mean, as you know, its very handy to have a power PCB sat on, and  afixed  to,  a heatsink.

[TJ232]

Thanks, thats what  i was worried about, so i wonder how people get round this?
I mean, as you know, its very handy to have a power PCB sat on, and  afixed  to,  a heatsink.

Proper design & stuff like that?

I would suggest you to look at Dave DPS5020 Power supply module repair video, he was quite badly hit by a SMD cap crack:

[ocset]

Thanks, good video...we're not quite as bad as that.....our screws arent right over the caps....but board flexion could result with us......i suppose our rubbery thermal pad kind of lessens the flexion of the PCB...but still, it would be nice if there was a cheap way to get round this and avoid any board flexion.

Doing the screws up alternately and bit by bit solves it, but you cant expect production staff to be so careful and painstaking like that

[ogden]

Doing the screws up alternately and bit by bit solves it, but you cant expect production staff to be so careful and painstaking like that

Torque limiting screwdrivers/wrenches is a must in factory where overtightening by staff can cause fault. This is in addition to properly designed PCB layout that does not put mechanical stress on sensitive SMD components.