Screw failure causes

[gersen]

Hello,
recently I had the a problem with a bolted flange. After a repair some of the screw head pop out (see the picture below).
Is it possible that the failure is due to a too high torque during bolting-up?
All the damaged screw broke at the base of the head.
The screw are Socket Head Cap M24 class 12.9, they bolted them with 2 (don't know why) split-lock washer DIN 127 B.

 

[hydtools]

I had hex socket head screws pop heads due to hydrogen embrittlement. The vendor did not bake the screws after treatment.

Ted

 

[gersen]

I had hex socket head screws pop heads due to hydrogen embrittlement. The vendor did not bake the screws after treatment.

Maybe is something like this because they used screws from 2 supplier and all the damaged one are from the same one.

 

[3DDave]

I would look to see if there was any filet between the head and the body; I would also send it to a materials lab to check for hydrogen embrittlement. Either problem could be supplier specific. I would not expect torque to be a primary cause because there should be enough stress concentration at the threads to cause a failure in the threads before the head fails.

Here's a related article that looks interesting for the problem:

https://www.assemblymag.com/articles/94240-why-threaded-fasteners-fail

 

[drawoh]

gersen,

We are missing a lot of information here. My quick estimate is that an M24X3 Grade[ ]12.9 screw can take 1200N.m or 900lb.ft of torque. What did you specify and what did they apply?

The M24 screw torqued to 1200N.m exerts around 250kN (57kips) force. Your split washers are being squished flat. They might as well be flat washers.

What else are people doing with your screws?

--
JHG

 

[BrianE22]

Certainly some corrosion present. How long were they in the field?

 

[chez311]

After a repair some of the screw head pop out (see the picture below)

Can you specify at what point? As in during/within seconds of tightening or sometime after (minutes, hours, days)? The state of corrosion suggests they spent some time in the field, or maybe were stored improperly prior to analysis.

they bolted them with 2 (don't know why) split-lock washer DIN 127 B.

Also FYI split lock washers are known to do next to nothing to prevent and possibly accelerate self loosening of fasteners. This is a legacy design that has been phased out of most applications. If one is doing nothing or detrimental, for sure two is not necessary.

There are many resources which discuss this at length but check out this short summary/plot from bolt science:

http://www.boltscience.com/pages/helicalspringwashers.htm

 

[gersen]

Here more informations:
Total Number of screws used: 96
Number of screws from supplier A: 59 - damaged 0
Number of screws from supplier B: 37 - damaged 21

they bolted up using a pneumatic tool.
I don't know for sure the torque used, because the man who bolted-up had little experiece and he could have used the wrong settings either 1200 Nm or 1700 Nm.
The failure didn't appear on the spot, the material was shipped to the client and stayed few days on the ground outside when he noticed that some head had popped out.
The material remains there a couple of weeks, this explain the corrosion spots.

Here are other picture:

 

[hydtools]

In my case, the heads popped off hours and days later. Product was on the shop floor when surprisingly screw heads were ricocheting off the walls and ceiling.

Ted

 

[SnTMan]

I'd think very unlikely due to over-torque. I'd think a manufacturing defect is much more likely.

A) All failures from Supplier B
B) Bolt-up procedures presumably the same for all fasteners

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[drawoh]

SnTMan,

I agree that if the broken screws all are from one vendor, it probably is bad screws. On the other hand, I see M24X3 Grade[ ]12.9 screws failing at 1700N.m torque. The installation torques are worth investigating. Perhaps manufacturer[ ]A exceeds screw specifications more than manufacturer[ ]B!

--
JHG

 

[chez311]

I would agree that the failure rate certainly points to a defect with supplier B's product. I would have these analyzed by a material lab.

I also agree with drawoh that your installation torque and procedures likely needs to be re-evaluated. Not knowing what torque setting was used as well as the installation of multiple washers tells me there isn't a whole lot of control over the process. If it wasn't done already, torque testing should probably be conducted on this joint to verify that the settings specified are correct for the fastener/joint combination.

Also speaking of procedure evaluation, if there was an approved supplier change (or multiple source approval) without proper testing/evaluation of the new supplier's product (or at least review of whatever conformance/certification testing the supplier performs) then a 57% failure rate should be enough to convince you that this is necessary. If it was completed, then perhaps some conformance testing of a regular interval is in order.

 

[gersen]

I also agree with drawoh that your installation torque and procedures likely needs to be re-evaluated. Not knowing what torque setting was used as well as the installation of multiple washers tells me there isn't a whole lot of control over the process.

I am aware of this problem an I'm trying tomake changes. It is difficult to change things in a 35 years old workshop, you wouldn't imagine the things I see...


Reviewing the process of bolting up, I'm pretty sure they used the right setting with the right torque.
The flanges were also tested, so that ideally every screws got an additional 50 KN load during test.
If they were bolted with a torque of 1700 Nm the preload would have been very close to the yield limit and I had a failure during test.
The thread of every screws is visually perfect.

 

[chicopee]

My question is on what type of installation is the pipe flange serving? Also have the broken cap screws examined by a metallurgic lab. Are the cap screws occasionally removed and reinstalled? These long neck cap screws should be able to take cyclic loads unless these cap screws came from places such as China. Next time you buy screws such as the ones you are using, it would be a good idea to get certs. Not much of a fillet between the heads and neck from those pictures shown.

 

[gersen]

My bad, the flanges are the head of some hydraulic cylinder they repaired so they are subjected to cyclical load during work. The screws were new new, the cylinders arrived with hex head screws.
Wie use quite a lot of scews from M3 to M56. We have few different supplier all hardware wholesaler (not sure if this is called this way in english).
This is the first time we had such a problem.
I have removed from the warehouse all the remainig screws of this kind. I will also make a warning to the supplier.

 

[hydtools]

Our purchasing dept. found an alternate supplier who had socket head capscrews without manufacturer-applied black oxide finish and the alternate supplier had black oxide done to satisfy our original specifications. We subsequently added specific manufacturers to the capscrew specifications so that purchasing could not be permitted to be so helpful.

Ted

 

[RolMec]

DIN 127 washers are way too soft for a screw quality 12.9 and respective torqueing, imo a = 1 hardened washer needs to be put under that head.
BTW DIN 127 has been withdrawn, as the securing functionality of that split-lock design does not work to the required degree.




Roland Heilmann

 

[Artisi]

Not my area of expertise, but would suggest first step is proper analysis of the bolts/material before anything else otherwise you are guessing.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Tmoose]

https://res.cloudinary.com/engineer...47586/tips/IMG_20181119_175743_790_ncq4vh.jpg

Are the heads on those fasteners still (barely) attached, or were the broken pieces stacked together for the photo shoot ?

 

[chicopee]

I would suggest that you research the topic of screws subjected to cyclic loads. There is an accepted (ie tested in labs), screw or bolt design subject to cyclic loads.