Bolt Failure Problems 1

[Matador]

I work at a pulp mill in Saskatchewan Canada and we have been struggling with downtime due to bolt failures.

Case Study

The bolts being used are Gr 8, 5/8" X 2" Lg (CSH*C58-2P, UNC) and we use lock nuts. (LNS9C58P, UNC)

(1) - The bolts secure the pans to a chain links and the conveyor is used to remove hot smelt from a recovery boiler. The bolts can be exposed to hot molten metal at the recovery boiler spout and during winter months -25 C outside temperatures at the conveyor discharge. Cooling water is also sprayed on the bottom of the conveyor pans.

(2) - In the past our mtce people tried Gr 12 bolts thinking it was a strength problem. The bolts failed right after they were put in service. Now they leave the Gr 8 bolts loose, by using a locknut and damage the threads to help prevent them from backing off. The downtime problem still exists because the nuts eventually come loose and the conveyor pans fall off.

Questions:
(a) - Can I change from a UNC thread to try and keep the bolts from backing off?
(b) - Does anyone have experience using diiferent types of nuts to ensure they will not come loose and back off?
(c) - Is the idea of leaving the connection loose a good one. Maybe bolt with different metallurgy would stand the expansion and contraction without failing.

Any help is appreciated.

 

[oldrunner]

Try using Anco nuts. We use them in the tower industry. I believe they only can be put on one time and can't be backed off.

 

[butelja]

Try the corrosion engineering forum. You may be experiencing liquid metal embrittlement. Have you tried tack welding the nuts on low strength (A307) fasteners instead of using hight strength fasteners? If the problem is corrosion related, then higher strength bolts will in general have more problems.

 

[WKTaylor]

Matador...

I am surprised that ONLY your bolts are failing!!!?

What is the condition of the Conveyor system?? IF OK, then what is the conveyor material? I suspect the conveyor is made to match the environment

Liquidmetal embrittlement (maybe from the fasteners' own coating IE: cadmium, Zinc, etc) is likely a factor, as noted by butelja.

Chemical attack, thermal shock and wear are also "obvious" problems.

I suspect that unplated [BARE] aerospace bolts (alloy A286, MP35, Inconel, etc...and maybe Monel... MAY relieve or solve this problem! Your company may be benefit by replacing "cheap steel fasteners with "expensive/exotic aerospace fasteners" that would be MUCH more durable in the brutal environment You indicated!!!! Part cost VS operational down-time could be improved dramatically.

NOTE: be cautious that the bolt alloy is compatible with the conveyor alloy... IE: not gall-prone and that dissimilar material corrosion [environmental conditions] be evaluated.

NOTE: be cautious regarding the hole-tolerances of the conveyor system... they may be "shot"... IE: worn "sloppy-loose".... in this case even great fastener-systems may fail due to chatter/vibration!

Note: suitable NUTs & Washers WILL BE REQUIRED in-order to match the environmental capability of the [noted] bolt alloys.

Some engineering eval/testing with these various aerospace alloys ... and/or assistance from a metallurgical consultant... is advised!

Regards, Wil Taylor

 

[Matador]

The conveyor pans are standard material. Mild steel, not cast.

I contacted Webster chain and they recommended a locknut called security locknut which they use on the vibrating conveyors they manufacture. The supplier is sending me a few to try.

Anyone have experience with this type of locknut?

 

[Ron]

Look at the metallurgical compatibility between the nut and bolt as well as thermal fatigue in the bolts. You are thermal cycling heavily and this could be a contributing factor.

 

[RHK]

Have you tried using castellated nuts, with a transverse hole in the end of the bolt, and a split pin through to hold the nut just that tight?

Alternatively, if you can tighten the nuts and still make your pans function correctly, the most reliable bolted joint is one with the bolt preloaded to yield. This can be achieved in the field by (a) tightening the nut as tight as you can, and then (b) turning it another 1/3rd to 1/2 a turn! (Codes call this "snug tight" and "turn of nut".)

Hope this helps. Russell Keays

 

[WKTaylor]

Re: RHK's Suggestion 4 Mar:

Obviously in a conveyor joint, rotational movement is required. You would need to use smaller diameter bolts and "clamp-up" bushings... to get a tightly fastened joint with rotational ability. The Idea of castellated nuts/cotter-pins as an added "safety-factor" with this type joint is excellent.

NOTE: due to the severity of environmental exposure, I still believe more durable alloys are mandatory/essential requirement here! Regards, Wil Taylor

 

[varunmahajan]

If your bolt is subjected to high temperature try using A286 material as suggested by wktaylor and use high temperature thread locking compound.

 

[MEisme]

You might investigate the use of Huck fasteners; concentric ring type bolt with hydraulically swaged collar taking the place of the usual nut. These have been used on the rake attachments of chain conveyors on chip pile reclaimer booms at paper mills. They are good for addressing potential fatigue problems and torquing of threaded fasteners.

 

[okeng]

When you say "bolt failure" do you mean that the bolts are breaking or that the nuts are falling off?

I occasionally see applications where we have high loads and very short bolts. Often we can improve bolt life by using a longer bolt and a spacer. You may not have room for that, of course.

I'm trying to remember the theory behind that--I remember having to re-prove it every time I do it but it never quite sticks in my mind. I has to do with having a longer "spring" as a bolt and seems to work where we have cyclic or shock loading. One common place I see other people do this is on the front wheel lug bolts on Deere combine harvesters.

I think the best nut for this kind of application is the "deflected thread" style.

There are also some miracle-cure nuts and lockwashers that might actually work. One that looks interesting is from Nord-lock. I haven't used them myself but it might be worth trying.

You might also evaluate seriously if it's practical to tack-weld the nuts. Usually is isn't, but it's worth asking the maint. guys anyway--sometimes you find that removal requires a torch for some other reason (rust) and they are happy to weld them together. Kind of a farmer/oilfield approach but very effective.

Another problem that I see fairly often is when the components are soft (like A36 or 1020) there isn't enough bearing area under the head of the bolt and the material yields in compression and makes the joint loose. I use hardened flat washers on a lot of things and it really helps keep them from working loose.

Yet another thing to look at is some way to take the load off the bolts. Can you weld a cleat or lug on the back of the pan so the bolt isn't carrying the shear load?

Good luck.

 

[IFRs]

If loose bolts are OK, why not use clevis pins?

 

[chicopee]

I agree with the castellated nuts secured w/ cotter pins.
The holes thru the bolts for the cotter pins should not weakened the fasteners either.

 

[Matador]

Yes the bolts were breaking. One solution tried was to install a shime under the nut for tightening then remove it and leave a gap for thermal expansion. This worked but the nuts would tend get loose, come off and the bolts fall out.

I'm curious about the spring theory to justify longer bolts with a spacer. It's worth a try but probably a tough sell to our maintenance people without some sound engineering logic for backup.

 

[evelrod]

I like the ANCO nuts as the little wire that locks the nut in place will ABSOLUTELY not allow the nut to come off. Indeed, it cannot be removed without great effort and damage to the bolt threads. It also locks in place even if the fastener assm is left purposely loose (alternately drill and "cotter" key the fastener). I also like the longer bolt/sleeve assm. Never used it in this application (I worked on several conveyor systems at the Southwest Pulp and Paper mill in northern AZ) but it sounds interesting.

Rod

 

[Sparweb]

Another problem with the bolts you were using may be the UNC thread. Try UNF threaded bolts. Aerospace bolts as mentioned above are always UNF.
The size of the bolt hole in the pan may also be a factor. Is it a close tolerance hole, or is it really big now that the joint has buggered up several times?

STF

 

[Matador]

The conveyor chain & pans have been replaced several times over the last 10 years. There is a tab on the chain link every few pitches which gets bolted to the pan.

Last year we started a program where we replace the chain & attachments every year during our annual shutdown. It hasn't been a total success but we don't have as many repeatitive failures.

 

[OzarkMTBr]

I have used double nuts with a longer bolt before to achieve a slip connection. With the double nuts you can reach the pretensioning to stop the loosening nut and allow the gap too.

With a loose fit though, you could also see bending and/or shear stresses depending on how the bolt is oriented.

 

[hydrodude]

Why not drill the holes larger and use a larger diameter bolt. You may have to beef up where the bolts go through to maintain the strength you need

 

[deanc]

Try a ASTM A320.

Contact the Industrail Fasteners Institute,Cleveland,Ohio
for their excellent handbook.