Weird bolt crack

[electricpete]

Attached is one of 24 bolts that holds a motor upper bearing bracket to the stator (goes into a tapped hole).

It was found during motor disassembly... nothing else unusual, other bolts are fine.

The externally visible crack starts in the center of the flat face at the "bottom" of the bolt (i.e. the opposite end as the head)... goes almost straight along the threads, and stops just short of the head.

I forgot to check bolt material - will report that tomorrow.

What could cause this type of crack?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[msquared48]

It is interesting that you get this same type of longitudinal crack in wood columns from shrinkage.

I would start by looking at differential cooling during the tooling of the bolt.

I do not know if you can this failure pattern from an overtorquing of the bolt. It seems like the head would fail first in shear, not this scenario.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[mfgenggear]

looks like a heat treat crack.

mfgenggear

 

[Pud]

Looks like bolt was made from seamed bar. The ends of the wire stock should have been cropped to remove any piping. (rolled in scale)

H

http://www.tynevalleyplastics.co.uk

 

[gerhardl]

Such an even crack, not disinterrupted or changed by the windings, indicate in my opinion a throughgoing material fault or weakness, pre-existing before mounting and caused under fabrication of the bolt.

if the bolt is fabricated by drawing out of larger barstock, cut, and then machine-forged to form head and stem, some fault or gliding within or between the two halves of the forging tools could possibly have caused this type of fault.

Other causes may be possible, but I doubdt difference in cooling.

 

[ivymike]

I think Pud & Gerhardl have it. That very straight radial line on the end of the bolt clinches it for me.

 

[TVP]

This is most likely a quench crack, due to the fact that it extends all the way to the center of the fastener, rather than a seam, which is a defect in the raw material (wire/rod/bar prior to forming and thread rolling). Either way, it is definitely a defect, and the fastener should not be used. ISO 6157-1, SAE J1061, and ASTM F 788 all use the same terminology for these defects (quench crack or seam), and specify their acceptance limits, which is 0 for quench cracks and up to 0.3D or 0.20 mm for seam depth.

 

[CoryPad]

TVP is correct, a quench crack is most likely. In addition to scrapping this particular part, you should inspect all parts you have from the same lot to detect any other cracks. You may need to use microscopic visual inspection, liquid penetrant examination, or magnetic particle inspection.

 

[ivymike]

why would you bother with dye-pen on a bunch of bolts? If you're worried about them, wouldn't it be much simpler & cheaper just to replace 'em all?

 

[CoryPad]

Depends if the testing costs/effort/timing are more than the aquisition costs/effort/timing of new fasteners. Also depends on how critical the joint is. What guarantees are there that new parts are crack free? If the risk is small, don't test. If the risk is large, test.

 

[electricpete]

Those are all good comments, thanks. The bolt material is ASTM A193 Grade B7.

We will go ahead and replace all the bolts in this motor and have this one evaluated further.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Cockroach]

This was definitely done during the manufacturing process, post machining. I would be looking up MTRs on this one and following the legacy of the bolt from bar stock through to completion.

I'm tempted to say HTSR, do a metallurgical and look for residue in the crack at various points. If there is an oxide layer, then the crack was done in the furnace at a high temperature. If there is oil or rust, then the crack was present at the time of usage or cracked during usage.

But it looks more metallurgical than mechanical. The line is perfectly straight, sulfide stringer in the iron matrix? Poor quality steel at the foundry? Water quench process during HTSR?

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[rmw]

You may have a patentable item there Pete. We are all familiar with locking nuts that lock by crimping onto the bolt threads, but a bolt that expands into the hole in order to hold tension........

Thanks for sharing. A very unusual event.

rmw

 

[JR97]

Keep in mind some specifications such as Section III of ASME B&PV Code are quite lenient regarding axial flaws such as in your bolt. MT/PT criteria per subsection NB is up to 1.0 inch axial linear indications allowed. No circumferential indications allowed.

I was under impression all in-service stresses are in axial direction therefore, there are no stresses to "grow" crack during use.

 

[MJCronin]

Your question was "What could cause this type of crack?"

I believe the correct answer is "the purchase of third world originated goods"

Which country did they come from ?

Consider re-posting in the forum "welding, bonding and fastener Engineering"

My opinion only

 

[ivymike]

all in-service stresses are in axial direction
depends on the application - for the one described here, it's probably fair to say it's axial only.

For long bolts in rapidly-moving structures, the bolt inertia can (by itself) induce significant bending loads.

 

[btrueblood]

Torque (torsion) loading is part of the normal "service" for a bolt, and would involve pretty severe non-axial loading. I'm surprised the bolt could be tightened without shearing.