Vibrational loosening, steel bolt in aluminum threads

[PacificSteve]

I have a lighting bracket which is cast of an aluminum. This has been tapped, the thickness is 3/8". The manufacturer uses a galvanized steel bolt. This bolt screws through the aluminum casting and engages a 2" steel pipe, thus holding the fixture in place (or so the theory goes).

The bolt fit is "sloppy", i.e. I dont think the tapping was done with the right lube. The manufacturer recommends 30 Ft-lbs. We break each casting at 30 ft-lbs. Then the manufacturer states use 20 ft-lbs. I don't think either 20 or 30 ft-lbs is enough preload to keep this 25 foot tall lighting pole connection from vibrating itself loose in short order.

Any help on analytic method to prove this point (prior to the now installed lights falling down) will be greatly appreciated.

Thanks
Pacific Steve

 

[deanc]

Try thread inserts "heli-coils".

 

[Lcubed]

It sounds as if the bolt functions as a setscrew, holding the fixture in position by friction, and probably causing substantial tension and bending in the aluminum sleeve. Is this the case, and what size is the bolt?

 

[CoryPad]

Why is the aluminium bracket threaded? Does it fail by compressive yield? Buckling? Do the threads strip? If you can provide more details regarding the joint geometry, we can provide you more help.

 

[smcadman]

Perhaps a combination of using Loctite to keep the thread from vibrating out, and chamfering the end of the bolt to a point to help secure the fixture in position.

Flores

 

[PacificSteve]

Some clarifying information:
The aluminum casting is threaded to allow the bolts to act as set screws. On each casting (there are a few varieties) there are either 4 bolts or 2 bolts.

The Aluminum casting fails by brittle cracking that originates at the threads.

I would say there would be substantial tension and bending in the aluminum. The light fixtures are an array of 4 ea 8 foot long fluorescents -- they catch a lot of wind, and at 25' up the pole will defintitely have movement.

The bolts are 3/8-16. They look like zinc coated in some manner. The threads through the aluminum are quite sloppy, although I dont know of a good way to classify them as 1, 2, or 3 "fit". The aluminum thickness is about 3/8" or about the same as the bolt diameter.

Under these conditions, can someone show a method by which the minimum preload torque can be established?

Big thanks, this is a great forum. Lots of science in fastening, I just wish I knew it already.

 

[CoryPad]

Does the screw have a head? What prevents the casting from becoming detached from the pipe? What is the mass of the casting (and anything connected to it)? This sounds like a substandard design.

A place to start is torque balance:

T_strip = T_input - T_fric

T_fric = 0.58 &[ignore]middot[/ignore]; d₂ &[ignore]middot[/ignore]; &[ignore]mu[ignore]; &[ignore]middot[/ignore]; F

d₂ is the thread pitch diameter
&[ignore]mu[ignore]; is friction coefficient
F is force

Standard joint calculations assume screw tension and joint compression. It seems your joint has the screw in compression, and I don't know how to calculate a force from this. One way might be:

&[ignore]delta[/ignore]; = P &[ignore]middot[/ignore]; &[ignore]theta[/ignore]; / 2 &[ignore]pi[/ignore];

where

&[ignore]delta[/ignore]; is displacement
P is thread pitch (1/16" here)
&[ignore]theta[/ignore]; is rotation angle

But what stiffness, elastic modulus, and length should be used to convert displacement into force? I am stuck here.

 

[PacificSteve]

The casting sits on top of a 2" Nominal pipe size stub up. is place on top of this stub up by the bolts only. The bolts have a standard head, however, the head does not come anywhere near the casting surface. i.e. the head of the bolt is in no way helping to stretch the bolt to preload it.

The casting supports up to 140 lbs of lighting fixtures, which are cantilevered off to one side. The moment arm would be about 18" (guessing). The casting itself is light, under 2 pounds.

I can email pictures, provide you email address, indicate if you wish to receive digital picture in the 2.3M size, or I can convert to smaller.

Thanks!

 

[Lcubed]

PacificSteve, I think it would be helpful if you would give us an email address. I will then send you mine, and you can send a picture. I think I have a feel for the cause of the failure and a possible fix, but a picture is worth.... etc., etc.
Regards
Lcubed

 

[PacificSteve]

solsen@pacific-energy.com

Note the dash after pacific

Thanks

 

[diamondjim]

It sound like a poor design.
If the aluminum is surrounding
the pipe and you have a differential
the aluminum ring would expand and thus
decrease or lessen the given preload.
I do agree that helicoils should be used
& I would increase the size to 3/8 inch
bolts. I also would prefer 3 or 5 so
they are not opposing each other.
How hard is the pipe? If any ice formed
between the pipe and aluminum,
it could expand and loosen the preload.
What kind of temperature extremes are the
parts seeing? Looking forward to other
comments and solutions. Hope no one got
hurt.

 

[IFRs]

My two cents:
1) you could put springs under the bolt heads to maintain some tension on the bolts.
2) you could put a hose clamp aroung hte outside of the heads of the bolts to keep them from turning out.
3) you could drill holes in the inner pipe for the bolts to go through, and use a through bolt.