bigretic

Now what? Can anyone tell me what to go buy at ACE for nuts and bolts to mount the replacement?
Oh, Hi... first post here. 97 Z28 just purchased.

Had a 94 TA (gt 6spd) - 20 years ago... lol. miss that car.

Anyway, help appreciated. Thanks. -d

HioSSilver

iTrader: (8)

Just use some 1/4" bolts washers and nylock nuts.

bigretic

Thanks. Going to attempt it this afternoon.

wssix99

iTrader: (5)

You know about the shbox method, right? (Bolts are no good for the regulator to door mount, so you don't want to touch that.

HioSSilver

iTrader: (8)

What's wrong with bolts?


Have you changed one of these before bigretic?

wssix99

iTrader: (5)

Bolts connecting the motor to the regulator (metal to metal) are fine.


Using them to connect the regulator to the door (metal to composite) is not good and leads to this: https://ls1tech.com/forums/general-m...-included.html

VIP1

Bolts didn't do that. See how only one is a bolt and the others are rivets? The problem is flex and fatigue regardless of bolt vs rivet. Mine had the stock rivets and the holes where started to get slotted (just no where near as bad as that thread). I had to add a reinforcing plate to the door with additional bolts to make the mount more solid.

Just be sure to use nylock nuts or lock washers.

HioSSilver

iTrader: (8)

Hell that had the rivets still in it wssix. I've not had any issues with bolts....other than once i bought a fcar to flip and it had a new window motor and they had come loose and caused the window to jump. Which is why i said nylocks.

bigretic

No, I have not done one, but I have done lots of other door work on an array of cars/trucks. I went to ace and grabbed some 1/4 inch bolts with washers and stop nuts/nylocks. I was simply using this video as a guide.
I did not get around to it yesterday, as I instead wound up removing and cleaning the egr and changing some old vacuum lines to resolve a P0400 exhaust recirculation code.

wssix99

iTrader: (5)

This is the method that you want to use and is the same thing as the shbox method. (It doesn't disturb the regulator mounting.)


Bolts do do that: (see bearing failure) http://structuralmechanics.com/about...gn-guidelines/

The problem is movement and bearing of the fastener on the composite. (That amplifies the fatigue on the composite.) The stock rivets could still tear, but the chances of them being loose and moving are much much less than a bolt. If we all keep the cars long enough, these doors will probably all tear out. Personally, I'm counting my days...

The stock rivets have an aluminum jacket, which deforms to fill the entire hole. A bolt (or even a steel jacket rivet) won't do that and will increase those bad fatigue stresses on the composite because they can move in the hole.

A nylock nut will help cut that movement over a normal nut or regular lockwasher, but one would still be relying on the clamping force of that bolt/nut/washer to relieve the bearing stresses on the composite.

A long time ago, I had a paper that talked about the steel mandrel and aluminum jacket rivets and their application here, but I don't know where that went. I suspect this paper below would have a lot of good information on using bolts in this application and their limits, but I'm not up for springing the money for it.

http://papers.sae.org/940623/

I have seen guidelines for SMC that bolt holes should be less than the thickness of the material and using bolts here would exceed that guideline.


If you are talking about not having any issues with bolts securing the regulator the door, then you don't have issues until you do. If you are talking about bolts securing the motor to the regulator, they are fine and nylocks are the way to go since normal nuts can vibrate loose and cause the window skipping.

VIP1

Given the forces/loads involved here....

* Use a bolt the size of the hole (to limit potential movement).
* Use wide washers to spread the load to prevent pull-out.
* Use nylocks / locknuts / lock washers
* Use thread locker.
---(Alternatively, use double nuts).

That will hold sufficiently and not pull through or loosen up.
The real reason rivets were used here is for speed on the assembly line.

It still comes down to load & fatigue and how to spread the load and mitigate/limit the fatigue.

I didn't see where in that article it said that a rivet was better than a bolt. It sounds like your argument/concern is regarding the bolt loosening up and not the method of mechanical fastening. If the bolt is sufficiently tightened and has an additional mechanical (lock washer/nylock/etc) or chemical (thread locker) means then that should be fine. As for spreading the load to prevent pull-out, having a wide washer would have helped the rivet as well.

If the holes are slotted, then you'll have to make new ones. My holes were beginning to slot so I drilled new holes and made a reinforcement plate to sandwich it all together.

https://ls1tech.com/forums/appearanc...ml#post4147168

Something like this mspaint drawing:
(the solid white bit is roughly what I made out of flat aluminum 1/8" thick -- probably less, I forget -- bar from Home Depot).

wssix99

iTrader: (5)

Yea, that is the real problem. I understand that the SMC panels are weak when the fasteners move in the holes and are prone to tearing up.


No doubt, but I think there is an engineering case for the rivets here, also.


Correct. Bolts or rivets can be used. They just need to be designed for.

I'm not an expert in composite fastening, but I understand there are general guidelines on hole size to panel thickness, etc. I did see one guideline that said that, if using bolts, the diameter of the hole should not be wider than the panel thickness. In this case, we'd be outside of that envelope.

If the bolts are clamped down with enough torque to limit movement, they should be fine. I'm not sure what that torque value would be, though. The nice feature about the stock rivets is that they are steel mandrel with an aluminum jacket and the aluminum will fill up the void of the hole to keep the thing from moving. (In addition to any clamping forces it would develop.)