Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Modified NPT Thread

Status
Not open for further replies.

sopperman89

Mechanical
Apr 6, 2023
3
Hello,

I have been with my current company for less than a year and I am trying to fix an issue we have been noticing with a valve we purchase with a modified NPT thread. The L1 dimension for a 1/8-27 NPT was modified to 0.0839" (image attached). Only the female side was modified. The stem that is used has a regular 1/8-27 NPT thread. These stems are coming loose in the field. I have never modified a thread before like this, is it possible to modify the male thread to get a better/tighter joint?

Any help is appreciated,
Steve
 
 https://files.engineering.com/getfile.aspx?folder=ff705331-5e67-4074-961d-95efa3621225&file=Modified_Thread.PNG
Replies continue below

Recommended for you

L1 is the hand tight depth of the gauge plug (male). Changing the male pipe thread diameter would change the L1 depth.

For 1/8NPT it appears that the standard L1 is 0.162" so yours is about half. You could re-tap the hole to get a standard L1. More threads engaged plus a locking thread sealant may help with your loosening problem.

However, one must always ask first, why the shallow L1 in the first place?
 
Your drawing shows the L3 condition being where the female threads end without any thread runout beyond that, and the fact that you have a nonstandard L1 condition means a standard plug gage won't tell you if the bore is under or over distance. Realize that the L1 dimension for male NPT fittings is gaged by a ring gage that allows an error of +/- 1 full turn. So, a male fitting with a longer dimension (near + 1 turn) will bottom at the end of the L1+L3 dimension (end of threads) when wrenched down "tight" - possibly more if the female bore is running under depth. So, roughly half of your fittings (if they were built to standard) will be engaged only on the endmost thread, allowing the fitting to wobble in the tapered bore. You need to have your female threads run at least a turn, and likely 2 turns, past the L3 condition to ensure a proper mate to standard male fittings.

This probably doesn't apply to you but, we see a lot of cheap male fittings made overseas that have over-long ends that won't pass the ring gage test. Also, a lot of mechanical firms doing pipe fitting in the field, put their junior guy on the threading machine, and give him a coupler or two for gaging (proper ring gages are expensive, and will wear out in time on a job using steel pipe). The newbie is going to pass a lot of male threads made that will engage the fitting, and not catch if they are over length.
 
@TugboatEng Short story behind the modified thread is that this is used on an oxygen valve that sees at most 3000psi when the tank is full, I have not been able to find an oxygen safe threadlocker rated to this pressure. The modified thread was required because back when this was being designed they decided to chose a valve that our manufacturer already made and sold (much cheaper than using a different valve) but the valve wall thickness was not thick enough for a full NPT thread. Beyond that there is no one left here who knows how or why the engineer chose this design.

Would modifying the male fitting help with thread engagement? Possibly change the L1 dimension to match the female? Also, is there a good guide I can use when designing threaded connections or is this really just about having the experience?
 
X-Pando pipe thread compound is claimed to be good for Oxygen. Also says it is rate to up to 5,000 psi. Maybe look into this option. I have no relation to the manufacturer so have no skin in the game.

We have used this before when having troubles with other pipe joint compounds or need something for high temperature. It expands as it cures so it helps to fill-in any voids. It is messy to work with and cures pretty quick once you make a batch.
 
At least it saved money.

Per the standard there should be 0.1014 inch beyond the L1 depth maximum engagement, and this calls for 0.1111 (a whopping 0.10 excess) but that depends on the thread being fully formed at that depth and there being nothing on the mating part extending beyond that - so knowing what the problem is requires a drawing of the mating part.
 
...or at least an inspection report (ring gage) of the mating part...
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor