Help. Screw preload problem with sandwiched parts 1

[Chase123456]

Hi,

I have a screw preload problem during assembly on one of the products at my workplace.

Basically, production workers have been applying a set torque of 0.82 Nm (which is a typical M3 torque for our products) and realising that this value of torque causes damage to the thermal pad (made from rubber insulated electrical fiberglass) and the nylon bushing.

This isn't a typical preload problem as there a few things being sandwiched between the screw when torqued.

My question is:

How can i determine a lower torque value which will be suitable to ensure good thread engagement and contact between the thermal pad and thermistor sensor but also ensure that the pad doesn't become damaged.

Can someone guide me on how i could determine this?. I know it'll probably involve calculating the compressive stresses due to the axial loading but i'd greatly appreciate if someone could help out with a worked example and the final torque value.

My assumption is that the critical weak point in this arrangement is the nylon bushing which deforms and then damages the insulator pad.

Please see attached schematic to help visualise this as well as drawings for the parts and damage seen during assembly.

Help would be highly appreciated. Thanks.

 

[3DDave]

This is usually solved by adding a spring with a low spring rate and then controlling the amount of spring compression rather than using a very stiff screw and a measurement with a large output range. Even with good control of torque one can get an actual compression range of 50% to 150% of the desired value. In addition you have differential thermal expansion to contend with as the nylon will expand and contract at a rate far greater than the steel screw. Note too that nylon is susceptible to becoming brittle if it is not kept hydrated.

 

[CWB1]

Gaskets often have the same issue with over-compression. You need the correct amount of both pad compression and thread stretch. Increase surface area of the stack on the pad and/or decrease thread size. Its a standard BJA but the tolerance stack analysis becomes more critical with additional parts bc of the added complexity/variability.

 

[TugboatEng]

Can you use a sleeve to control the amount of compression on the rubber?

 

[SWComposites]

you can faf around for a long time trying to calculate a torque, but there are going to be so many assumptions that its doubtful you will get a useful result.
just get some spare parts and run some experiments with various torque levels and sort out a torque range that works.

 

[mfgenggear]

Change material of the bushing to aluminum or brass. Increase the compressive strength.
Or chose a plastic with better compressive strength.

 

[TugboatEng]

Glass or carbon fiber reinforcement for the plastic will likely be necessary

 

[3DDave]

It's an electrical insulator, so carbon, brass, aluminum aren't going to help. The shouldered bushing is intended as a compression stop.