Calculating Max PSI of NPT threads

[jetinsystems]

Hello

I got a Recieved a question from a customer that I can't answer, He wanted to know how we calculate how much psi a npt thread would take. For instance we make pipe and fittings out of stainless steel that are rated for 15,000 psi working pressure. I can answer the pipe thickness rating rating question and what we rate the pipe for in max psi but not how the threads are rated for the pressure. I was hoping there would be a formula for the variable pipe thread sizes and different materials.

 

[CESSNA1]

JETINSYSTEMS: What does he mean. Is he looking for the axial capability of the threads or the radial capability? If it is axial it is a simple tension calculation using the root diameter of the thread. A radial calcualtion is different and I have never seen it, nor has anyone ever asked me for it. Why does he want it? You could use the outside diameter of the pipe and subtact the root diameter to get the minimum wall thickness then calcu;ate the stress using that.

Regards
Dave

 

[TBP]

He'll likely need to know what piping code he's working with. Codes like ASME B31.1 and B31.3 have the formulas in them. He should be aware that there is an element of "judgement call" regarding corrosion/erosion allowances. There are all different materials & grades of pipe & fittings, manufacturing tolerances, etc I don't think anyone can make a blanket statement to the effect that "Pipe threads are good up to XXX PSIG".

 

[rconner]

I was just curious (perhaps for anyone intimately familiar with same) how the "formulas" TBP talks about account for all manner of sealants and "pipe dopes" that may be (used) out there?

 

[bcd]

jetinsystems,

The ability to calculating strength limits is useful for design purposes. Basically estimating appropriate dimensions. However, at the end of the day, the real proof is by testing. Simply run a burst test and tell the customer that this is how you establish the rating. I doubt anyone could predict with certainty the location and mode of failure for an NPT joint without exhaustive analysis. There are so many factors and variables. How far above yield to you have to go and for how long until the material finally gives-up. Are there any side oads on the NPT connection? How tight are the threads (preload), etc., etc.

bcd

 

[jetinsystems]

I appreciate all of the input from the community.

I had originally offered the customer the axial calculation simple using the root diameter of the thread. The same one that Cessna1 offered. The customer is working at a Dupont plant and the Dupont engineer on that site wants it proven on paper that the threads will seal at 10,000 psi, not the yield of the material to burst or the strength of the threads. I have never been asked anything like this. Nomally simple teflon tape or loctite PST Pipe dope. Side loads are rare or supported but Frequently there is pulsation or some harmonic vibrations due to the triplex pump action, not excessive. Sealing id not normally a problem.

In addition as I have thought about this and explored some of the other NPT topics addressed on this site the variables are too many to consider when mating various Mfg's products due to different tolerances, materials, depth of threading, IE total qty of engaged threads at what percentage of thread cut. TBP touched on this on his reply to the post. If anyone has seen the sealing capabilities of NPT addressed please let me know. Yesterday I offered our certification of test and burst pressures that have been tested in the past and they refused it, wanting the mathematical proof that the threads would seal. We are not doing anything that is not an industry standard. We will probably just supply the certification and leave it at that, but it is an interesting question since it is true that as you increase the size of the thread above 1"npt that it does become harder to get the threads to be drip free. The question here is less that 1" so I don't feel it is an issue but an interesting question nontheless.


I prefer a metal to metal seal such as an autoclave,NPT is a lousy thead, but one that I don't think we will see go away.

 

[proinwv2]

I will add some confusion to the issue. A major problem with pipe threads is that they are often not threaded to specification. That is they are not checked with thread gages to insure proper dimensions and length of engagement. Also, there is the possibility of actual surface damage to the threads before assembly. Sealants are undoubtely useful.

Further, be aware, that for manufactured valves and fittings, while probably tested by the manufacturer, they may or may not have been tested using engaged threads. Sometimes they are clamped in vise like test fixtures and the threads are not actually checked.

As previously stated, nothing takes the place of a test. When possible, a lab burst test is certainly useful, but certainly not always possible. In the case of field installed joints, an appropriate test pressure must be used based on safety in the field considering all the elements to be tested. Pressure testing must be properly planned, conducted and supervised.

Personally, I have no love for pipe threads, especially in larger sizes and would avoid them if at all possible. But that is my personal nonsense and I realize it is not practical in many cases.

Paul Ostand

http://www.ostand.com