How to determine the NPT fitting's engagement

  1. As we know, the male NPT fitting can screw into a female NPT thread. this process could be discribe like this: tighting the fitting for several turns with hand, and then tighting the fitthing for several turns by the wretch. Here I have a question: what is the tolerance of the engagement of the coupling?how can we control the screwed depth? by torque or anything?
  2. jcsd
  3. Ranger Mike

    Ranger Mike 1,675
    Science Advisor
    Gold Member

    NTP is National Taped Pipe numbers refer to inside diameter of the piece of heavy wall plumbers pipe that was originally designed to receive the male thread. the threads are so designed that the male thread is an interference fit into the female counterpart and this interference that keeps the fluid from leaking. These threads are not designed to maintain tension and are only for connection. Typical pressure maintained is under 500 psi, use teflon tape or pipe dope to seal up the threads as manufactuirnf imperfections on thee low quality threads are huge.
  4. S_Happens

    S_Happens 299
    Gold Member

    AFAIK there is no standard for connecting NPT (we try to go with a no leak standard). With all the applications within the pipe and what type of thread sealant is used (and how liberally it is applied) I don't believe there will be a torque that could consistantly correspond to a certain depth.

    I'm not a pipefitter so if there are a set of industry standards or site specific standards that we use I'm not up to snuff on them. The only standards I know of are when/where it can be used, which if not in water service is only for temporary, low pressure use.

    Is this simply for curiosity?
  5. I thought the OP was trying to figure out how deep the threaded male pipe would go into the female fitting, so that they could determine how long to cut the pipe (which I presume is only one of several pieces of pipe to be fabricated).

    I have wondered about this before - seems like there must be some uniformity or else the pipefitters would have to cut & thread each piece sequentially at the jobsite if they need to build the system with any degree of precision. Maybe we have a pipefitter on line who can shed some light on this.

    If this wasn't the intent of the OP please excuse me, I'm not trying to hijack your thread.
  6. Q_Goest

    Q_Goest 2,989
    Science Advisor
    Homework Helper
    Gold Member

    Machinists use a thread gauge to verify the threads are cut to the correct profile. There's a ring gauge to check male pipe threads and a plug gauge to check female pipe threads.
  7. FredGarvin

    FredGarvin 5,084
    Science Advisor

    According to MIL-P-7105, which takes it's cues from specs like the ANSI form, the thread depth form variation is plus/minus one turn from the basic diameter, i.e. the pitch. In practice, Q_Goest is right that gauges are used to make sure manufacturing is within those tolerances of the entire form.

    The control on assembly is to screw in until hand tight and then three full turns after (wrench take up). There are prepared tables that list all of these dimensions for each thread size.
    Last edited: Sep 30, 2009
  8. S_Happens

    S_Happens 299
    Gold Member

    Which sounds fine and dandy, but is hardly practical in the real world. Pipe runs are hardly ever straight (numerous fittings), hand tight is subjective, etc.

    I think the general nature of those guidelines is itself an answer to the OPs question regarding whether torque could be used to determine depth. Most certainly the pitch can be used, but I had assumed (always bad to do) that it was self-evident and only addressed determining depth through indirect means such as torque.

    Actually, in my experience in the chemical industry that is exactly what they do. We only use screwed pipe in certain applications, which aren't high precision. Most everything is manufactured on site and assembled so that fittings are in the proper direction and there are no detectable leaks. Almost exclusively it is the direction of a fitting or valve that determines how many turns are finally used.
  9. FredGarvin

    FredGarvin 5,084
    Science Advisor

    It's not unrealistic at all. I use pipe threads all the time. A pipe run does not need to be straight, just the hole/portion that is being threaded. The thread form is governed by the dimensions and hand tight engagement is one of those dimensions. It's not that difficult to hit within ± 1 turn. In my experience, the only thing that seems difficult to do is for the machinists to utilize the proper tap depth when tapping holes. If I had a dollar for every time they didn't run the tap down far enough..
  10. S_Happens

    S_Happens 299
    Gold Member

    I never said it was unrealistic, simply impractical. Basically I was pointing out how general the guideline is, when it seemed to be referred to quite specifically. Hand tight is not quantitative, and not every application is a fresh, clean, undoped set of threads.

    I don't think we are looking at this any differently. Your added tolerance of +/- 1 turn puts you in the same boat as my description of lining up the fitting to accomodate a certain direction. I didn't specify to be within 3 turns after hand tight, but the norm in my situation is a minimum of 3 turns.

    My point about pipe runs not being straight simply refers to having to use fittings (90s, 45s, etc). If you follow the guidelines and simply go hand tight, then 3 turns, there is no telling which direction the fitting will be facing (compounded by each successive fitting). If you simply keep within 1 turn and place the fitting in the direction it needs to be, then you did exactly as I already described instead of explicitly following the guidelines. Either way, neither the guidelines, nor a certain amount of torque can be used to calculate the depth.

    I think the OPs question was answered, and a very general guideline nitpicked.
  11. Fred, where (what kind of systems) do you like to use threaded joints? My field experience is in power plants, and the only place I remember seeing threaded pipe is in the compressed air systems. I thought the reason for that was they wanted to use galvanized pipe (for corrosion resistance without the cost of going to stainless steel). The pipefitters I talked to said it's faster & easier to assemble a socket welded system compared to threaded. But I never did see them install threaded pipe so I just don't know how it goes - if an elbow (say) tightens up before it is oriented correctly, do they have to unscrew it, and cut more threads on the pipe?

    I'm not criticizing any previous posts, I'm just trying to find out how the work is actually done...
  12. FredGarvin

    FredGarvin 5,084
    Science Advisor

    I work with pipe installations for a lot of different materials, i.e. air, fuel, oil, etc... in our facility when required for test programs. I prefer to not use threaded pipe for connections, but I use NPT threads for some fittings and flanges when required. The situation I use them most is when welding is not a real option, for example in a tight location inside a building where access is limited.

    The pipe fitters I have worked with are as varied as machinists. Everyone seems to have different ways of doing things. I'm not saying my way is best at all. I just simply wanted to let you know that they are somewhat controllable in my experiences. Someone putting in a pipeline out in the middle of nowhere or in a large chemical plant may have different experiences.
  13. we use 1/8 -3/8 NPT often on our products. 2-3 turns minimum. Use only quality teflon tape (the thick stuff) When taping the fitting always leave the last thread un wrapped, and 2 wraps max. if stainless or steel is used, ALWAY use a lube in it. we run npt up to 16,000 psi on a consistant basis, but its not your typical Home Depot brass fitting. Many people screw up and assume x size fitting is good for y pressure and it seals for a while only to fail later. before you ever screw something together, ALWAYs find out the manufacturers suggested working pressure unless you want to have the ER (or coroner) removing the fitting later

  14. thanks you all.
    I have no internet access these days.
    we need to use two NPT fittings in our products. in order to do the tolerance stack-up, the NPT engagement tolerance is needed.
  15. gmax137,
    you are right. I want to know the how deep the male thread dose go into the female thead.
  16. maybe my description is not clear.
    this NPT fitting was used to seal high pressure refregrant. it is screwed into the female pipe thread on a box, and the box is full with high pressure refregrant, about 500psi.
    typically, in our products, the torque was given for the NPT fittings. but I don't know whether it is right. according to different material, the torque should not be the same value.
  17. FredGarvin

    FredGarvin 5,084
    Science Advisor

    Have you seen this table? This is reprinted from Machinery's Handbook.

    Attached Files:

  18. I an pretty sure, if you are working with solid works software, swedgelok has a fittings library download, which has that included

  19. FredGarvin
    Thanks very much.
    I've read the table that you attached. What made me confused is that threads depth tolerance. You know, the tightening process is sepreated into two steps - the hand tight and the wretch tight. According to the Machinary's handbook, the max allowable variation in the commercial product is one turn large or small from the basic dimensions. Does this mean that the total tolerance is +/-1 pitch including hand tight and wretch tight?
  20. thanks dr dodge,
    I will check it. We use the swedgelok fittings in our products too.
  21. FredGarvin

    FredGarvin 5,084
    Science Advisor

    Swagelok as well as Parker have all their fittings in 2D and 3D formats. I use them all the time.

    Yes. The tolerance is +-1 for the combination of hand tight and wrench take up.
Know someone interested in this topic? Share this thead via email, Google+, Twitter, or Facebook

Have something to add?