torque on silca pads

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
MaccAu

MaccAu

Member
Joined 2009
What is the best torque for applying to the silpads? for the following. I dont have a torque set but maybe i can get away without or get one if needed

Re: torque on SPK4-0.006-00-58 - for Thyristor to heatsink?
Tensile Strength 5000 psi

Re torque on SP900S-0.009-00-105 - for Sanken Output transistors on an amplifer - shape of the sanken output transistors 200w MT200
Tensile Strength:1300 psi

Sam

Mouser Parts links (digikey has actual photo of item) for Sanken
SP900S-0.009-00-105 Bergquist | Fans, Thermal Management | DigiKey
 
Last edited:
Elvee

Elvee

Member
Joined 2006
Since this one is fiber-glass reinforced, it has no practical limit by itself, but the object you want to cool certainly has, especially if it is partly or totally made of plastic (remember though that even full-metal parts also have a limit)
 
multi-volti

multi-volti

Member
Joined 2005
Old Motorola (obviously old) semiconductor databooks used to have app notes discussing things like heatsink mounting surface flatness (25-50 micro-inch iirc) and hardware torque for different packages. 6-32 screws on a TO-220 package probably had only a single-digit-to-low-double-digit torque (#-inch) need (wild guess 6-12 oz.-inches?). The flatness and grease or silpad assure no air gap so there is good heat transfer/low thermal resistance. Tensile strength probably relates to limits on the screws...shouldn’t be approaching any failure limits...finding semiconductor mounting requirements would probably be more meaningful than limits associated with failure. The heatsink flatness probably assured the bulk of the work for good heat transfer. There was probably no pressure spec for thermal compound used before pads because the flatness made it happen.

I also remember hole diameter recommendations for mounting screws to avoid deforming the semiconductor surface into the hole, causing potential die damage inside the device package.
 
Kay Pirinha

Kay Pirinha

Member
Joined 2011
llwhtt said:
I didn't notice your torque specs, I was just adding the document # that shows all of that data. AN1040 shows in/lbs, my torque driver is in in/lbs but the guys on the other side of the pond want it to be lb/ins which is probably more correct.

Craig
Click to expand...



What the hell is this? Torque is defined as the product of force times lenght, not as a division!


Best regards!
 
MaccAu

MaccAu

Member
Joined 2009
JonSnell Electronic said:
Hand tight is the norm. Depends on the nut/bolt type. If steel more torque than brass but if silicon heat sink compound is used, hand tight is fine.
Have a look at the chart;
https://mdmetric.com/tech/Standard_tightening_torque.pdf

Hope that helps.
Click to expand...

Hello Jon
Thank you for the reply and help. I did it by hand carefully each time checking continuity to make sure there was no break in the silica pad

Elvee said:
Since this one is fiber-glass reinforced, it has no practical limit by itself, but the object you want to cool certainly has, especially if it is partly or totally made of plastic (remember though that even full-metal parts also have a limit)
Click to expand...

Thanks Elvee, i remembered not too hard on the plastic of the transistor, but a nice firm contact with heatsink

scottjoplin said:
Use these to allow for expansion and contraction Belleville washer - Wikipedia
Click to expand...

Thanks Scott, that looks good, the screw had a sort of washer like that

multi-volti said:
Old Motorola (obviously old) semiconductor databooks used to have app notes discussing things like heatsink mounting surface flatness (25-50 micro-inch iirc) and hardware torque for different packages. 6-32 screws on a TO-220 package probably had only a single-digit-to-low-double-digit torque (#-inch) need (wild guess 6-12 oz.-inches?). The flatness and grease or silpad assure no air gap so there is good heat transfer/low thermal resistance. Tensile strength probably relates to limits on the screws...shouldn’t be approaching any failure limits...finding semiconductor mounting requirements would probably be more meaningful than limits associated with failure. The heatsink flatness probably assured the bulk of the work for good heat transfer. There was probably no pressure spec for thermal compound used before pads because the flatness made it happen.

I also remember hole diameter recommendations for mounting screws to avoid deforming the semiconductor surface into the hole, causing potential die damage inside the device package.
Click to expand...

Thanks Murry? Multivoltage and Craig

Picks below actually no screws for the thyristors but a clamp.

I had trouble so i added a thin layer of wakefield 120-2 grease - very thin coating to thyristor so i could stick the silipad in place but none on the heatsink itself.

The back pair of thyristors are nice and flat, the other pair a bit of lift at the top of the pad but the sides look flat, if I were to tighten more i could damage the plastic of the the thyristor?

An externally hosted image should be here but it was not working when we last tested it.



other bank of thyristors is prefect

An externally hosted image should be here but it was not working when we last tested it.


Thanks
Sam
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom