Flanged vs Captive Washer Fasteners (Torque Retention) 2

[DeBruce]

All,

I am currently looking at replacing fasteners on an automotive product and have been testing a number of alternatives.

My tests seem to show that fasteners with a captive washer perform worse than flanged fasteners with regards to torque retention. Does anybody have any advice or can point me to any external studies with regards to this?

Also with regards to plastic components, what are the best methods of prevention torque loss over time. Because the bolt or stud cannot be stressed, I assume locking fasteners must be used (e.g. Nyloc). Is this the case (in mass production) or are there other alternatives?

Many Thanks,

Rob

 

[MintJulep]

For starters, torque is not the goal when tightening a fastener, tension is.

So, look at either tension or elongation when making your decision, not torque.

 

[TVP]

Rob,

Have you investigated any fasteners with different thread profiles from regular machine threads? There are several thread forms that were developed specifically for polymers and other low strength materials such as Plastite (Reminc), PT (Ejot), etc.

 

[tbuelna]

MintJulep,

I believe the correct term is strain (not elongation).

Rob,

The most common locking device employed in mass production is good 'ol Loctite compound. It works very reliably as long as it is applied under the proper conditions. However, it is not reuseable.

As for clamping up on a plastic component, it is standard practice to put a metallic bushing in the hole to prevent crushing the plasic material and to avoid the loss of fastener preload due to plastic creep or cold flow.

Good Luck,
Terry

 

[CoryPad]

Rob,

What is your test for torque retention?

The only real difference between a flanged fastener and a captured washer one is the friction coefficient with the mating surface and the lack or presence of torsional stresses on the mating component.

Can you describe your joint? Plastite or PT screws are appropriate for screwing into plastic bosses, and tubular metallic inserts are appropriate for plastic parts in the clamped zone.

That being said, there are many joints (like windshield washer fluid reservoir to body attachment) that are just a low strength polyethylene tab with a slot or oversized hole that is attached with a steel screw. These are the types of joints that have excessive stress relaxation, and can cause problems with inspection and staying tight over time.

Regards,

Cory

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

 

[DeBruce]

All,

Thanks for the responses. The joint comprises a weld stud used to locate a glass-filled thermoplastic component with no metal insert, which is then secured with a Nut.

I am investigating Torque retention because it is the easiest method of testing tension retention on actual components in life simulation testing.

Unfortunately I am limited to selecting nuts which mate with the existing weld studs since it will not be practical to change the weld studs. In addition it will not be practical to add any type of compound to the joint - this type of process is not practical on a high speed mass-production line due to variability.

My test for torque retention is to use a torque guage to examine the wind-up torque required to move the fastener after it has been subjected to temperature cycling. This is more reliable and repeatable than measuring back-off torque.

I am trying to justify whether a Nyloc or Crimped Nut should be used (i.e. with a prevailing torque feature) or whether the standard Hex Flange nut should be retained and possibly the torque increased to compensate for the torque relaxation evident.



With regards to captive washers, I believe the effective bearing area is smaller on a captive washer type fastener because the hex head spins against the washer, whereas with a flanged nut the bearing surface consists of the area under the flange itself. This gives increased friction and hence less susceptibility to torque relaxation. Thoughts...?

 

[Tmoose]

Fastener relaxation due to material creep/embedment requires no nut rotation. I think that makes heroic measures on my part to keep nuts from moving misguided, at least on metallic structures. Far better to have long stretchy fasteners so a few 0.001s of an inch embedment result in minimal clamping loss. Belleville type washers can provide similar effects in light duty applications.

If my torque retention (loss?) is due to creep or embedment of the highly suspicous glass-filled thermoplastic material, I'd expect the biggest surface at the glass-filled thermoplastic face to cause the smallest creep.

I'd be using my testing to achieve understanding, not hypothesizing about what to test.

 

[patprimmer]

Initial thermoplastic creep will reduce the clamping force to a greater extent than will be evident by the torque retention, as the material being clamped will flow away from the clamped area, but will also conform more closely to the surface irregularities in the bearing area. This increases break away torque or so called sticksion, despite the clamping force being reduced.

Increasing initial clamp increases initial creep rate and most of the gain is quickly lost.

You cannot apply "metals" thinking to plastics.

Thermoplastics are simply super cooled liquids with extremely high viscosity, but they are liquids like glass and lead and do flow with time.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[DeBruce]

So I would interpret the information above as indicating that because of plastic creep, tension relaxation will always be evident in fasteners on plastic components.

Therefore it follows that a fastener should be selected which features a prevailing torque feature in order to prevent vibration from further loosinging the fastener over the life of the product. Genius!

 

[patprimmer]

Yes. Unless you have a device to correct for the creep, like a spring between the part and the nut, or a crush tube to bear the load. A crush tube is the method used to retain plastic inlet manifolds on engine cylinder heads.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Tmoose]

"Therefore it follows that a fastener should be selected which features a prevailing torque feature in order to prevent vibration from further loosinging the fastener over the life of the product. "

Prevailing torque would help make the assembly failsafe in terms of not falling off for a while. If part of the assembly's function >>>>DEPENDS<<< on clamping force (sealing, accurate location, protecting the fastener from fatigue, probably lots of others) then loss of clamp force via creep or embedment must be compensated for by some means or another.

I spend some time defining a part's function to clear my mind and be armed to gently head off discussions that sometimes veer away from function towards dreamy techno-philosphy.

 

[bvanhiel]

patprimmer,

Glass isn't a liquid. Google will pull up a hundred articles explaining why this oft repeated tidbit isn't true. Never heard anyone else claim that lead=liquid, so I can't debunk that one.

Doesn't make anything else you said less true. The tiny amounts of creep exhibited in metals are nothing like those exhibited in plastics.

I don't think I saw it mentioned, but there are fasteners available with threadlocking compound pre-applied.

-b

 

[Tmoose]

I think the comparison to lead and liquids may be based on lead and its alloys' strong tendency to respond to stress by recrystallization very quickly at room temperatures, resulting in almost no stress (stress relief). Loading over ~ 200 psi simply causes lead to ooze out of the way.

http://www.key-to-metals.com/Article88.htm

 

[jbthiel]

Corypad touched on the frictional diffences.
"The only real difference between a flanged fastener and a captured washer one is the friction coefficient with the mating surface and the lack or presence of torsional stresses on the mating component."

Torquing both styles to the same value will NOT result in the same amount of clamp load.

I recently compared the torque value of two different 20mm DIN style lug nuts to achieve the same stud elongation. One had a captive washer and the other was flanged. The flanged style required ~50% more torque due to the increased friction.