cracks in malleable iron threaded 90° elbow 2

[Eduardo1982]

Hey guys. In a construction project we were facing some problems with some malleable iron threaded 90° elbows. During some pneumatic tests two elbows failed (I attach some picturees). We think that the problem is bad practise during construction by the contractor (due to excessive wrenchs marks). However the client is asking for a deeply root cause analysis with emphasys in the material properties. We got some quality certificates of these material but are too basic (most of them just say that mechanical properties met the requirements of ASTM A197. 90° elbows are class 150 and correspond to ASME B16.3 fittings). I was thinking to send the failed fittings to a laboratory for metalography test (probably to check microestructure if everything is OK) but I am not sure if it could be enough.. I would like to know if with this failed fitting can I get some tensile test.. I need some advice of you guys that maybe have much wide experience in these fields!!
Any help will be helpfull.

 

[ornerynorsk]

Have seen this before and assumed it was from massive over-tightening. Good lord some of those monkeys don't know when to quit turning the wrench. It's either that or they forget the pipe dope and figure they can "make" it seal by giving it an extra half-turn. Curious as to whether you found any tape or compound in the joint when it was taken apart?

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[Eduardo1982]

Yes I found tape in both threads. However the point is that my client does not believe that the contractor could have the fault. That is why i am here asking for some advice, so I can test failed material.
Oh I forgot to mention that we had problem with a T, too. Maybe it is the reason why my client thinks that the material is poor quality.

 

[MintJulep]

You can insert images directly into the body of your post by using the "Upload Image" feature.

Interesting that the wrench marks are on the end that didn't fail.

What size is the fitting?

 

[EdStainless]

Eduardo,
If they want to pay to have someone tell them that these are typical fittings and they were over tightened then go ahead.
These are tapered threads, if you keep turning they will eventually split.
If you have some other fittings around you could just over tighten some as a demonstration.

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

 

[ornerynorsk]

Yes, understood, thank you. Do you have any new fittings from the same lot that you could benchtest yourself to gain some foreknowledge before tensile and metallurgical analysis? Short of the supplier being able to provide better certs or documentation, labwork is probably the only way to prove or disprove that the fittings meet spec. Torque measurements to split a fitting are not necessarily a reliable indicator as there are too many factors in tapered threads. As a sidenote, Teflon tape is a wonderful lubricant in allowing overtightening to be more easily accomplished.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[Compositepro]

Once you have the elbow on one pipe you do not need a wrench on the elbow to install pipe onto the other end. The wrench is applied to the pipe and not the elbow.

 

[metengr]

The blame game. Best to have the lab perform an analysis of the material to eliminate any other possibility because it could get ugly, fast.

 

[ornerynorsk]

Also, some perspective in the scope of things would be helpful. Judging by the woodgrain pattern of the table this is lying on, it does not look like a large fitting. Class 150 is a relatively cheap fitting. Is it cheaper just to spend the money to replace them instead of spending a couple thousand $$ in labwork? Are there only a handful of fittings or are there 10,000 that need replacing?

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[Eduardo1982]

Thanks guys.. I did not know that I can insert images over my notes.
Just to clarify a little more about the issue. Diameter of elbow is 1/2". I just got this elbow and one T too (both failed during pneumatic test). Around 500 fittings among 90° elbows and Ts have been already installed in lines (service air and seal water lines). It is kind of difficult to replace all these 500 fittings for new ones (due to access: some of these are installed 10 meters high) and the client is seriously concerned to this material..
The client issued a non conformity report to us and I contacted the supplier of these material.. they dont answer and just one seller guy said that can replace the material for new ones (but they can not cover the cost to replace all the lines nor to pay the manpower hours of work needed). So my boss told me to try to find out a solution and make all needed tests to show that the material meet the standars (ASME B16.3).

 

[arunmrao]

The elbow is a sand casting with several surface defects seen . It instils a low level of confidence in the product. As was rightly pointed out by others ,it may be a cause of over torquing .

In one of the pictures I could see the colour of the fracture surface to be gray,which should be ok. A metallurgical analysis might give a better picture about cause of failure.



"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[btrueblood]

Root cause? Over-tightening due to a threaded pipe section that was not to spec. I.e. they let the new kid run the threading machine, and he ran the threads too deep/long, so that when the fitters went to make up the joint, they had to wrench it "extra tight" to get it to seal properly. When you do this, then end of the pipe runs out into the bottom of the threads, and the wedging action of the taper can drastically exceed the normal forces on the pipe wall. Get a 1/2 NPT ring gauge, and check all of the nipples/pipe ends to see if the failed fittings came off of defective (overlong) threads. Make the contractor come back and replace the defective pipe sections.

 

[Eduardo1982]

Yes.. I supouse the same.. over torque but the client is concerned due to material. So now I need to do some tests on the material and to check if these meet the ASME B16.3 specs. It is kind of difficult because according to ASTM A197, the tensile strenght is the only mechanical property specified in the standard. But I can not get (with my two pieces) the specimen to do tensile test. I was thinking to try to do hardness test in both pieces and do a correlation with the tensile strenght. In adittion to see the microstructure and to check if everything goes according to this kind of foundry.

 

[mrfailure]

The longitudinally-oriented cracking indicates hoop stress exceeded tensile stress either as a single event or under cyclic loading. That is more characteristic of overpressurization on the inside of the fitting. Overtightening will most typically either induce circumferential cracking at the base of the threads or angled cracking from the resolved shear stress.

I agree with your client that you should have an independent metallurgical lab perform an appropriate failure analysis that will include both metallographic analysis and tensile testing so you can identify whether material quality is a major issue.

 

[ornerynorsk]

"Overtightening will most typically either induce circumferential cracking at the base of the threads or angled cracking from the resolved shear stress."

Hmmm, I've seen a lot of "popped" fittings in my day, and the ones that get over-tightened look exactly like the example shown. An analysis is definitely in order with the client claiming what they are. Contractors will always blame "insert any reason or excuse here" rather than admitting that they did something wrong.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[racookpe1978]

The clientappears to have made his mind up already (wrong material was chosen), perhaps because that way it was not their favored customer who caused the failure (by over-tightening a screwed joint).

The pipe wrench scars on the other joint of the elbow show excessive force was used while turning. Also, clamping that end of the elbow might have been the "resistance" torque holding the pipe_1 & elbow as the new pipe_2 was twisted and over-torqued.

 

[Compositepro]

I agree that this crack is entirely consistent with what you would see from over-tightening. The crack is widest at at the end of the elbow where it is thickest and the only load at that point is from the pipe threads and not internal pressure. Perhaps not everyone is aware that pipe threads are tapered and that over-tightening expands the outer fitting until it bursts. It is a very common problem, particularly with cheap, thin-walled fittings.

 

[mrfailure]

There is a hoop stress component to tightening that adds to the hoop stress from internal pressure from the pneumatic tests in terms of overall tensile component of hoop stress that is required to drive that longitudinal crack. At the same time, I think a lot of tea leaf reading is going into interpreting the meaning of the wrench marks - they do not necessarily mean overtightening. There are a range of potential root causes for this longitudinal crack - you would need a lab investigation to figure out the correct one and/or eliminate others.

 

[WKTaylor]

REf ASME B1.20.3 DRYSEAL Pipe Threads...

Can we assume 'Dry-seal taper-pipe threads' are on the male component [externally threaded pipe]?

What 'dry seal pipe threads' are on these female parts [elbows/tees]... (a) internal tapered threads or (b) internal straight-cylindrical threads?



Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[ornerynorsk]

Do not think these are dryseal (NPTF), just common NPT.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.