1/4-20 'thread forming' screw

[bernardg]

Hello All,

We manufacture the dc motors. And for attaching the end frames and the shell, we plan to replace the bolt & nut assembly with the self-threading screws (thread forming).

We use 2pcs of 1/4-20 'thread forming' screw to drive it into a blind hole in the casting. (The thread-forling screws (casebolts) are made of case hardened steel. They are driven into the blind hole of the zinc casting).

We apply 150 in-lbs input torque to the screws (casebolts) during the final assembly.

Before doing a random vibration testing on the motors, our customer wants us to place the motors in the environmental chamber and make the motor see extremes of temperature (+85C to -40C) for 3 days. The relaxation that happens in the vicinity of the thread-forming screw during the 3-day thermal cycle is so severe that the backout torque drops substantially to as low as 20 in-lbs. Because of the drastic reduction in the clamping force... we failed the vibration test.

Instead of 1/4-20 'thread forming' screws... we used the 1/4-28 'thread forming' screws. But this didn't help us a lot.

We did not verify if 'using a wicking loctite on the casebolt threads' would sustain the retention (clamping) strength. That is what we are planning on experimenting next.

Have anyone come across this same scenario? I need your expertise to figure out what we can do to adress this issue. Please help me out here.

Thanks,
Bernie

 

[CoryPad]

Do you know if you develop actual preload with 150 in-lb torque? Perhaps you use most of the torque to form the threads and there is no preload developed to clamp the joint.

Do you know the creep behavior of your zinc casting alloy? It seems like you have relaxation of the preload (both axial and radial) and that is causing the low "backout torque". By the way, loosening torque is not a good indicator of joint quality.

Thread adhesive may help, so you may want to investigate using it. Both Loctite and 3M make good products.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[alexit]

My experience with Loctite and self-tapping screws is bad. I think that since zinc is changing size from temperature greater than the steel you are needing a belleville washer sized to maintain clamping force.

 

[bernardg]

Thanks for your prompt replies.

Other data that might help you to understand this better...
On 2 motors, I torqued the casebolts to 125 in-lbs... and I allowed those motors to sit for 1 week. Then, i measured the backout torque. It measured 55-57 in-lbs.

on 2 motors, I torqued the casebolts to 150 in-lbs... and on 2 motors, I torqued the casebolts to 175 in-lbs. I placed those motors in the environmental chamber for 3 days and cycled it hot-cold-hot-cold. Then measured the backout torque on all for motors. They all measured between 35 to 40 in-lbs.

Cory, The driving torque for the casebolts (to form the threads) measures 45-50 in-lbs. I have placed the 4 motors that have loctite in the casebolt threads are in the encironmental chamber. I will vibration test these this Friday.

Alexit, The belleville washers sounds like an excellent idea.I am building 4 sample motors, with the belleville washers underneath the casebolt head. This should sustain some amount of clamping force.

So do you have other suggestions on this? Anything at all... I will experiment those options and choose the one(s) that works... and will share the results with you.

 

[diamondjim]

I think I would try using very heavy
washers under the head say 3/8 thick.
I think the greater clamping length
might help. Washer od greater than
the head diameter as well.

 

[NickE]

Have you solved for clamping force for 1/4"-20 torqued to 100in-lbs? Finding a belleville that can significantly increase your clamp load (and allow you to use a 1/4-20) is going to be tough.

 

[diamondjim]

The greater clamping length may do
as much as the belleville washer.
It is worth the try. You could even
experiment with adding the belleville
washers to the thick washers.

 

[vcebrad]

I'm assuming you are replacing the bolt/nut combo with the thread-former as a cost savings project? You may want to watch the costs of assembly A vs. assembly B as you start adding in more part numbers. The part and labor costs may appear low, but it's one more p/n to order, carry in inventory, inspect, run out of, etc.

 

[Pressed]

You could test a set of Nord-Loc washers. They make them down to an ID of 4mm but. I would check to see if there is enough contact to your bolt underhead, followed by a test.

_______________________________________
Feeling frisky.........

http://www.tailofthedragon.com

 

[Screwman]

What is your failure torque for the thread forming screws in the joint. Try a seating toruqe of about 75% of that failure torque. It may help.

I have seen this problem before on stator bolts where the laminations had creep during bake out. It is a tricky one. One thing that I would check to see is what kind of retained torque you have with the machine screws. Generally, if a joint can provide enough thread engagement to have bolt breakage as a failure mode; if it was sucessful with machine screws it will be successful with a thread former. And generally speaking, vibration loosening isn't a common failure mode with thread formers since the formed thread has no internal clearance to allow relative nut movement.
Make sure that you are using licensed Taptite screws, the other ones can have spotty heat treat and dimensionals. Also check on your draft angle, it shouldn't be any more than 1.5 degrees included. Hole size should be about .221 - .225.
The Nord Locs will certainly work but they will eat up all of your savings from going thread forming.

Good luck.

Dick