hoop stress in tapered threads-fittings are cracking 1

[hcoles]

Would like some help finding link to equations to calculate the hoop stresses in the female side of NPTF/NPTF 1 1/4" connection. The material is brass. The current best guide from Parker is to use the TFFT. ... for 1 1/4" the TFFT is 1 1/2 turns to 2 1/2 turns. Some fittings may have been over torqued, i.e. the TFFT was not followed. How can I get an idea how much stress was put into the fittings before they failed? The failure was a straight crack running along the outside of the female fitting parallel to the flow direction. I think there should be some idea how much hoop stress is in a brass fitting when assembled e.g. 2 turns from finger tight (TFFT). I'm looking for that type of equation.

 

[MikeHalloran]

Couldn't you just look up the tensile stress of the alloy and call it a day? Your gorillas couldn't have developed any less stress than that, or you wouldn't be here. Since the part failed, they couldn't have developed any more stress than that, either.

If you want to estimate how long a pipe they had to slip over the wrench, then, yes, you'll need to do some math.



Mike Halloran
Pembroke Pines, FL, USA

 

[hcoles]

Mike,

Thanks. We need to ask fitting supplier what the exact alloy of the brass is, we don't know yet. I think we want to get comfortable that if we reinstall new parts to replace cracked parts that we are not near over stressing the material. For example if 2.5 turns past finger tight gives 15kpsi and the brass is good for 55kpsi..then we are feeling that fittings will not crack due to installation. We are not sure how far they were put on 3 turns or 4 turns??? Some large tools were used and the operators, not our company, were given instructions something like get it as tight as you can....

After some more thinking the fitting supplier should know exactly the stress as a funtion of turns past finger tight so we will ask that tomorrow.

Appriciate your interest....

-Henry

 

[hcoles]

Mike,
I just had an idea..
I might be able to estimate the hoop stress because it is basically an interference of two hoops...
I know the interference addition per turn because the threads are tapered I think 3/4" per foot or something like that...

Does that make sense?

-Henry

 

[hcoles]

In NPTF fittings there is some thread crushing and "bending" over while the TFFT is being done... so maybe that idea is not so good after all.
-h

 

[MikeHalloran]

Small women especially seem to take "as tight as you can get it" as a personal challenge, but they're pretty good about following directions.

You need to give the ultimate assemblers a specific procedure, e.g. Parker's,

...

or use push-on fittings,

...

or cut their wrenches in half.





Mike Halloran
Pembroke Pines, FL, USA

 

[deanc]

Are you using Teflon Tape? Bad news...if so. A little pipe dope is best.

Check your ANSI spec. for this fitting. 2 turns past hand tight is a good place to start. If your threads are to spec.you should have 2~3 showing when tight.

 

[FOETS]

Why not just set up a test to establish the various torque settings.
Take a sample of fittings to failure. Try a sample by going with the manufacturers recommendation. Try some samples with various thread dressings to see the difference. You will then have a feel for what the torque should be,and be able to come up with a standard; which will probably be far better than trying to calculate as you still seem to have some unknown variables affecting the outcome. Manufacturing tolerance and quality control on this type of fitting probably has a fair range so calculation would only be an estimate anyhow.
By knowing the failure torque you know the factor of safety.
The hardest part is then training your gorilla's to torque them correctly and making sure they do it.

Tighten till you feel the threads strip then half a turn back to lock it ...