J
John Crighton
Guest
On Wed, 16 Jul 2003 08:44:35 +1000, Mike Harding
<mike_harding1@hotmail.com> wrote:
you solved it.
Years ago I worked at a place called Maitec
that made UPSs. They used strips of that sil-pad
material stuck onto the heatsink just like a stick on
label. Peel and press on. The strips they used were
just over an one inch wide by several inches long.
The FETs protruded out from the edge of the board
and were clamped down by springy steel clamps.
One clamp for two or three power FETs. One springy
finger per transistor.
The advantage of the clamp is that positioning the
device when soldered to the board just has to be neat
and not accurate.
No plastic insulating washer involved that can soften up.
I only mention this as a second best approach, in case
you have to do it again for a less strenuous application
and happen to come across some lengths of that sil-pad
stick-on strip at a good price.
I enjoyed reading your report. Thanks for that.
Regards,
John Crighton
Hornsby
<mike_harding1@hotmail.com> wrote:
thanks for sharing that heat sinking problem and howAnd finally:
Mounted the 3 MOSFETs with the heaviest load on their own
0.37 deg C/W without any insulating pads and with two local
3" fans - ran them for 20 minutes+ with each mosfet passing
6A25 and dissipating 58W plus 3 power resistors (on the same
heatsink) each dissipating 18W (grand total of ~220W). The
MOSFET case temperatures went to around 110 deg C
which means the junction temps are within spec. for those
devices. Problem solved
Moral: Don't use insulating pads on heatsinks if you can
possibly avoid it.
Mike Harding
Hello Mike,
you solved it.
Years ago I worked at a place called Maitec
that made UPSs. They used strips of that sil-pad
material stuck onto the heatsink just like a stick on
label. Peel and press on. The strips they used were
just over an one inch wide by several inches long.
The FETs protruded out from the edge of the board
and were clamped down by springy steel clamps.
One clamp for two or three power FETs. One springy
finger per transistor.
The advantage of the clamp is that positioning the
device when soldered to the board just has to be neat
and not accurate.
No plastic insulating washer involved that can soften up.
I only mention this as a second best approach, in case
you have to do it again for a less strenuous application
and happen to come across some lengths of that sil-pad
stick-on strip at a good price.
I enjoyed reading your report. Thanks for that.
Regards,
John Crighton
Hornsby