Bolts Loosening in Transport 1

[pbc825]

I just received a phone call from a client who had a lifting/tailing beam nut loosen in transport. The part being transported is a cylindrical shell of ~24' diameter and length. The shell has lifting lugs on top and a tailing beam on the bottom. The tailing beam is left on for transport and bolted to the bottom of the shell with 2 bolts at each contact point (one bolt/side of the beam web). The shell has 1.5" carbon steel plates on the bottom of it, and we opted to have the plates drilled and taped for 7/8" diameter A325 bolts installed using turn-of-nut. Threads were engaged to about 1". The bolts have one washer each and were torqued using the turn-of-nut method.

The part was traveling about 1000 km to site and this was noticed about 900 km in. The new (longer) bolts have been used to reinstall with a lock washer. I've done some cursory research

Some questions for you:

1) Were we naive in thinking torquing using turn-of-nut would not vibrate loose for this transportation?
2) Is there a specific torque (say to 80% of yield capacity) that may have provided some benefit over-turn-of-nut?
3) Are there others out there that use a different installation method, type of bolt, or locking washer for transportation scenarios?
4) My cursory research has pointed to Nord-Lock washers as a securing method for vibratory bolted scenarios. Are there others out there? Is it standard industry practice to use a product like this?

Thank you in advance for any advice.

 

[SwinnyGG]

1) Were we naive in thinking torquing using turn-of-nut would not vibrate loose for this transportation?

The depends- on what specific turn-of-nut method you used to tighten these fasteners.

2) Is there a specific torque (say to 80% of yield capacity) that may have provided some benefit over-turn-of-nut?

Without knowing how much bolt preload was induced by your turn-of-nut method, it is impossible to say if some torque value would have been more or less resistant to vibration loosening.

3) Are there others out there that use a different installation method, type of bolt, or locking washer for transportation scenarios?
4) My cursory research has pointed to Nord-Lock washers as a securing method for vibratory bolted scenarios. Are there others out there? Is it standard industry practice to use a product like this?

I suspect that if you were to poll engineers designing assemblies for similar applications, you'd get as many answers as you would poll responses.

Anecdotally, I have experience with Nord-Lock products, and have found them to be excellent. They are head and shoulders above any other preload loss prevention solution that comes in the form of a washer or set of washers.

There are still lots of other options, however. For really extreme cases, you may want to consider safety wire. For REALLY extreme cases, you may want to consider designing your assembly to use fasteners which simply don't loosen due to rotation- i.e. HuckBolt or similar products.

There are too many possible solutions to this problem to name them all- but the first defense against fastener loosening is to make sure your fasteners are getting enough preload; without knowing how much preload these fasteners were subjected to, we can't help you determine which direction to go.

 

[moltenmetal]

I can tell you that I have zero confidence in the effectiveness or even basic usefulness of the split spring lock washer.

Nordlock products work but are expensive- useful for a handful of critical applications but not on every fastener. Pre-assembly and post-assembly threadlocker adhesive products also work, especially in services such as transport.

 

[pbc825]

Thank you all.

jgKRI, my understanding is that turn-of-nut is meant to yield the fastener. This should generate "high" preload (close to the yield strength of the bolt).

We've acted as the engineering team for a number of pieces of large equipment like this in the past, and this is the first time it's ever come up. Our protocol moving forward will be to peen (or otherwise damage) threads on the backside of nuts (that don't require removal following transport), use threadlock adhesive on those that do, and use Nord-Lock washers in instances where the bolts are in drilled/tapped holes. This is not a standard, but I recommend this to those who read this thread in the future.

 

[SwinnyGG]

jgKRI, my understanding is that turn-of-nut is meant to yield the fastener. This should generate "high" preload (close to the yield strength of the bolt).

Well.. the turn-of-nut method CAN be used to preload a bolt into the inelastic range. But so can torque methods, if correctly applied.

What I meant above is this- if you told us that the fastener had been tightened with a torque wrench, you likely would have told us the torque value.

By just stating that turn-of-nut was used, without describing the scheme, no one can tell you if the bolt had sufficient preload for the way it was loaded, or not.

If you're the one specifying the installation method of these fasteners, just saying 'install using turn-of-nut method' or something similar is not nearly descriptive enough for you to expect reliable and consistent assembly of your parts.

 

[pbc825]

Fair point. The specification included turn of nut in accordance with the CSA S-16. Which should be descript enough for the application. For the length of bolt, I believe the procedure was snug tight +1/2 turn.

 

[dhengr]

Pbc825:
It would really be interesting and helpful to see some photos and good/complete details on that connection, and how it works, we can’t see it from here. 600 miles is a long way for some beam hanging out off a couple bolts (cantilevered off a couple bolts at one flange) at each end to be vibrating down the road, whatever road conditions existed. 1" of bolt engagement in a tapped hole is pretty minimal for this type of application. Why not a hole through the 1.5" thk. pl. and a through bolt? Furthermore, the tapped threads in the stl. pl. probably did not have the same strength as an A325 bolt, so they would control the bolt tension. It most likely isn’t so much the torque or preload on the bolt that matters, but rather that you tighten the joint up and then prevent the bolt from rotating. I haven’t ever heard too many good things about any kind of lock washers for these kinds of conditions. I might also want to do something in the way of braces to keep the tailing beam from flopping in the breeze, as a cantilever.

 

[Tmoose]

As always, pictures of your arrangement would be helpful. I'm most interested in Details of all the beam attachment points to the shell, and some indication of the nature of the loading the tailing beam is subjected too during transport.

Is this shell a size and construction you ship frequently, and over distances greater than 1000 km, or is there something new or unusual about this one?

What is the verified preparation of the mating/faying surfaces between the shell and beam ?
I envision the surfaces on the belly of a large round painted component could have a thick irregular build up of paint.

https://encrypted-tbn0.gstatic.com/...tQ6ck6iIjsnRdWe1q3KKPjwoXKw_kHvrM6lU7ihucImwg

http://p2.la-img.com/406/9227/1848891_1_l.jpg

http://www.secondchancegarage.com/a...cessive-orange-peel-and-major-run-in-door.jpg

Bolting a structural component to a surface with a thick coat of paint is looking for trouble, in the form of fastener loosening.
Anything but smooth, flat clean metal faying surfaces are going to be subject to loosening, even with fasteners that are adequately torqued at installation.

 

[dik]

I have great confidence in turn of nut, if properly done... how long was the bolt grip? I've never encountered one that loosened. Was the installation snug tight, or slip critical? Are you sure it was properly installed? For critical applications, I would normally use a thread locker like Locktite Red.

Dik

 

[pbc825]

That's appreciated dhengr. I would love to add photographs and all the details, but I'm unfortunately bound by a robust confidentiality agreement that prevents me from publishing all of the best details. Also, I don't have any photographs. I'll do my best to fill in some gaps.

The beam is W section 12" tall and 12" wide. The beam is approximately 26" long with a 18" cantilever. The 1.5" plate rests on the underside of a HSS tube that stiffens the cylinder. The plate was part of the EOR's design. Zero load during transport. Just along for the ride.

Turn of nut was witnessed by an end user inspector and client inspector.

I have a theory that some cutting fluid was left in the hole. And paint between the beam flange and 1.5" plate is to blame. Paint would tend to make the bolt loose preload in time. Vibration finished the job aided by the lubricating cutting fluid. Nord-Lock washers would have helped.

 

[EdStainless]

We used to build equipment in OK, and one test that we did on new designs was to ship to Long Beach CA, then Casper WY, then Edmonton, back to Long Beach, and then return to the plant in OK. If the equipment and new packing all survived that trip we would consider moving forward.
Transport vibrations are much more severe than most people realize.
Having seen so much equipment destroyed in shipping I am not the least bit surprised by your issues.
Clean mating surfaces (no paint or oil), through bolts (maybe with nuts on the back), and either Nord-lock or locktite are all useful steps.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[tbuelna]

As noted, probably best to approach this bolted connection as if it were slip critical, including prep/inspection of the faying surfaces to ensure the proper friction coefficient.

If possible, it would be better to use through holes in both flanges with a bolt and locknut, instead of installing the bolt into a tapped hole. If threads are necessary in the flange, you might consider installing a self-locking thread insert (like a heavy duty Keensert) in the tapped flange holes to keep the bolts secure.

 

[dvd]

What size tap drill was used? Who specified it?

 

[pbc825]

3/4"-10UNC-2B. We specified the drill and tap

 

[btrueblood]

Regarding safety wire, there is a lot of debate about what is usefulness is. Some maintain it will prevent fastener preload loss, others, myself included say it won't (the torque resistance of a piece of soft 1/16" diameter twisted wire strand is really 150 ft-lb?) and its real function is to prevent loss of a broken screw head where such may cause damage (jet engine compressor wheels come to mind).

 

[dvd]

My question

What size tap drill was used? Who specified it?

was about what size of hole was drilled prior to tapping? Did you specify the drill size or just specify 3/4"-10-UNC-2B threads? Were the threads and minor diameter inspected? If either threads or minor diameter were improper you could loose tension.