Proper Threaded Pipe Joints

I am not a mechanical engineer, machinist or plumber. So I cannot comment on the nature of this failure or speculate on the cause. What I can say though is that you need to hire a formal forensic analysis to support your claim. If your legal council does not have somebody in mind, contact a local engineering school. We cannot examine the failed components. So, any advice you get here will be colored by your description of what you believe is relevant.

Now that I've obliterated any value to my comments, some ideas come to my mind.

• You might be the victim of a counterfeit piece made to look like your product.
• Your product may have been machined after it left your factory.
• The assembly was not using the correct pieces. Possibly a competitor's nipple or other part was used in the assembly.

Good Luck

Did it break (as per report) during tightening or later during operation.

Did you find the operating condition, especially vibration, dampers, stresses and supports in the final installation? Any of these could create the extra stress that can lead to the cracks. (Also the materials for male and female portions and their thermal expansion- if the temperatures are significant)

• The following ASME code is the most used for pipe thread: ASME B1.20.1 Pipe Threads General Purpose (Inch).

• In addition to the following ASME codes and standards related to screw thread:

ASME B1.1 Unified Inch Screw Threads

ASME B1.2 Gages and Gaging for Unified Inch Screw Threads

ASME B1.3M Screw Thread Gaging Systems for Dimensional Acceptability – Inch and Metric Screw Threads

ASME B1.5 ACME Screw Threads

ASME B1.7 Nomenclature, Definitions, and Letter Symbols for Screw Threads

ASME B1.12 Class 5 Interference-Fit Thread

I assume this was in drinking water service but there is a difference between surface water derived drinking water and well water. Surface water is sterilized with chloramines whereas well water is sterilized with chlorine which is shorter acting. If Chlorine is used for surface water which has more "organics" (fish poop), it can generate H2S , which will attack copper causing it to crack due to sulfide stress cracking. Overtightening would enhance the effect because it would increase the stress in the joint. There could also have been a metallurgical or machining flaw in that particular part.

This you may consider comments of a journeyman machinist.

When tightening a properly designed and produced pipe fitting there shall be at least 3 to 3- 1/2 threads showing.

Generally when tightening a pipe threaded joint the procedure requires that only 1 - and 1/2 turns are required to complete the connection. Over tightening will result in cracking and leakage.

bwire there are no threads showing on the nipple and that is what concerns me. There also seems to be material issues with the fittings so you have a good scenario for cracking. When you have a product issue in the field, fitters tend to get upset when you suggest possible installation errors. Especially when comments come from factory salespeople. Investigating threaded pipe and fittings has always been a favorite. Thanks for your input.

You're quite welcome, the tapered nature of the thread allows a three thread minimum engagement when one and one-half turns are used to tighten; beyond which the female fitting is crushing the male fitting to a point then the cracking may begin. Especially when the material is bronze or cast iron and to a lesser extent if brass or other maleable materials but continued tightening and something has got to give.

In the feild I have begun a statement with due to my experience the following conditions are of concern and the likely causes are application and or installation errors and just because it's a routine process doesn't preclude ineffectiveness in this instance.

I am probably showing my ignorance here but was the correct sealing used? Whether it was teflon tape or similar I would not know, but if no "correct" sealing media was used by someone, then overtightening is possible when an amateur is at "work"!!

Just a thought!!

Interesting post.

You can over tighten brass real easy and the cracks and leaks might not show up until it is stressed. I have also seen brass fittings with thin spots in the walls.

I have gone back over a year to work that was done to repair leaks and being able to say what caused it would be like flipping a coin .

I still say that if no sealant was used, overtightening was the problem. Its as simple as that.....

The use of proper thread sealants prevents leakage when joints are properly tightened.

When joints are overtightened and cracking occurs a proper thread sealant will have nominal effect against leakage.

You may be missing the point I was trying to make which was if a proper sealant was used, the torque required to seal will be minimal, no damage will occur!!.

BUT, trying to get such a union to seal WITHOUT sealant, will in simple folk induce them to tighten and tighten and tighten and tighten and tighten etc.....

With no good end effect!

Now my statement stands:- "No sealant = overtightening", its as simple as that!!

Andy what you say is 100% true, but there is also the issue that thread sealants reduce the torque required by acting as a thread lubricant so that they can, in the hands of the foolish, also increase the possibility of overtightening because the joint doesn't "feel" tight and since most unskilled people go by "feel" and not actual torque values or pressure testing, this can be a two edged sword.

I agree completely.

My personal method is to assemble fully, then turn (in my case) the water on and if it leaks to gently increase pressure till the leak is stopped. Teflon usually needs a bit more pressure......

I tend to use teflon tape for such junctions.

But big steel pipe I still use the Hemp/flax (?) and a sealing paste.....never had one fail, nor has one needed adjustment either, am I brilliant or what?

The problem is that many are willing to (try to) do jobs that they do not have a clue as how to carry out!!!

This is the rule I adhere to when using pipe fittings; I apply plumbers dope and hand tighten then wrench tighten 1-1/2 turns more and that's it. Additional tightening is required less than than 2% of the time.

Very similar to my personal experiences with regards to sealing pipes, though I have only worked (privately only) on gas and water.

I sealed my gas pipes on my caravan to hold up to 1 BAR (14 Lbs.Sq.Inch?), but the working pressure is only 30 Millibar....the tester (they pump it up to twice working pressure and its only allowed to lose so much % (no idea of how much) over 20 minutes.

After 20 minutes it had lost nothing, he WAS surprised and asked me how I achieved it, told him, "its sealed for 1 bar!"....I don't like gas leaks...... of any sort!!

He nearly fell over, he had NEVER seen a caravan before that did not lose a little gas!!!!

UK Caravans have copper piping for gas, in Germany only steel is allowed (much harder to seal up), I had to replace all the copper before I could go on the road legally....

To the first question: There is no official written standard describing how to assembly a pipe joint. But, there is a Best Practice that is often stated in Pipe Fitting Catalogues of various manufactures. There are also things you can investigate to help prove whether the joint components were manufactured correctly and/or assembled correctly.

To the second question: Lack of exposed threads on the nipple does not necessarly always point to over tightening, though it does in most cases. It could just be a case of improper thread form on the nibble or its mating part related to their assoicated dimensional standards.

Standard Best Practice for Pipe Joint Assembly - It is very important to insure the threads in both parts are thoroughly clean. Any threads that may have become burred or bent between original forming and installation should be straightened or removed by retapping. A quality grade of lubricant should be applied to the thread to reduce friction during assembly. The lubricant reduces friction allowing the two components to be drawn together more closely. Due to the deformation of the threads incurred during assembly, a pipe-thread joint should be assembled slowly to minimize excessive friction. If screwed together too rapidly, excessive heating will occur. If mating components are of dissimilar materials, they will heat and cool at different rates.

Regarding some things you will want to confirm when you investigate:

Thread Form Commonality - There are many different types of pipe threads, most common are listed below. Ensure both components were manufactured to the same standard:

NPT - Tapered

NPS - Straight

NPTF - Tapered for Fuels

NPSF - Straight for Fuels

NPSM - Straight for Mechanical Joints

SAE Straight Thread O-Ring

British Standard - 55-degree Whitworth

Condition of Thread - Even though some deformation has already occurred due to initial assembly, the threads in question can still be inspected. Most leaky joints are caused by issues with the male thread form. Look for wavy threads, shaved threads (generally from recutting, or chasing a damaged thread with a different die than was used to make the original form), and rough or torn threads. Any of these could cause a leaky joint.

Thread Dimensions - Not knowing the diameter or thread type in question, I am unable to post specific dimensions... but there are plenty of tables and charts for each standard that show all relevant dimensions. Unfortunately, newer reference do not contain some very good references so you may have to go back to some older ones for stuff like 'Length of engagement to make a tight joint' or 'T.F.F.T' (Turns From Finger Tight.) Often times, manufactures will take subtle shortcuts to increase throughput and manufacturing speed/efficiency. They will use the next size up drill size for female threads. (For example, the standard screw thread tolerance is 2A/2B... almost no machine shop uses the stated drill for that tolerance and most all go one size up... this change still provides a 50% thread engagement and significantly increases tapping speed and tap/die life. Although a common practice, it is still not good practice as most fastened joint calculations are based on a 75% thread engagement. For pipe threads some common issues are... male threads too short, female threads too shallow or are straight, not tapered. If either of these occur, over tightening is required to give the installer the sensation of a tight joint. Due to length of thread reductions, over tightening significantly shortens the mating cross section.

At the end of the day, there are many things that can cause a leaky joint. My suggestion is that you take both mating components to a skilled Piper Fitter and have them take a look at it. They should be able to provide some valuable input.

JavaHead

Javahead do not forget API Line Pipe threads which are identical to NPT threads with the exception that the female thread is counterbored such that the partial threads on the male when properly engaged are swallowed by the counterbore. This is intended as a check to ensure that the threads are fully engaged.

If the person doing the work is used to working with API LP threads, they may not have realized that the lack of partial threads showing was a problem.