Improving pipe threads on LDPE injection molded component 1

[99ishvr4]

Hello,

We recently designed some LDPE components that 1/4" NPT pipe threads. One component with internal threads and what component with external threads. We had molds made and received initial samples.

When tightening components together, the thread engagement feels sloppy. Before the components are fully tightened together, I can feel a small amount of play if I push/pull on the components. We compared this to existing off-the shelf products that use 1/4" NPT pipe threads, and this issue is not present in those. If we screw our male threads into the existing product female thread, and vice versa, the sloppiness improves, but the threads are still not mated as well as when using both existing products.

I am used to working with metal threads where we would check the threads with a plug gauge or ring gauge. Would this also be applicable when checking plastic threads? It seems that if the threads did not conform, the gauges would just deform them into complying, without me being able to validate the inspection.

I did examine the male threaded components with a microscope and overlayed the ideal thread profile in green. Our nozzle is top picture, existing off-the shelf nozzle is bottom picture.

As you can see, the crests and roots of our threads do not conform to the standard and I need to figure out why. Is this a machining issue in the mold, an issue with the injection process, or both? How should I go about correcting this issue with the mold maker, and what is the best method to inspect these threads upon arrival, other than the microscope?

I appreciate any help on this matter.

 

[3DDave]

Were this my problem I would have an FEA done to determine the deflection due to a uniform contact pressure on the threads, whatever pressure seemed useful and then invert that distortion from the nominal thread geometry. If the thread moved in 0.001 under contact pressure, then the thread should be molded 0.001 outward.

Plastic has such a low stiffness compared to metal the interference fit will produce much different installation feel. If you want the feel to be similar, you need to compensate.

 

[btrueblood]

Axial play or lateral (perpendicular to the thread axis)? Not sure why you are concerned about it - the real purpose of pipe threads is to seal fluids, presumably that is why you are using them? Do the threads seal okay bare, and/or if sealant is applied?

 

[99ishvr4]

btrueblood, It has both axial play and lateral play. I believe that when the two components are fully tightened, that only 1 or 2 threads are actually engaging. In testing, when we put the system under pressure, the male threaded component blew off of the female threaded component. There is also a feel aspect involved, the operators that will be using these daily want a confident feeling when they are tightening the threads.

 

[btrueblood]

Ok, back to the subject of axial and radial play - all threaded joints have play in them, there is a purposeful gap to allow the joint to be assembled. The play is worse in a tapered pipe joint, at least at the start of engagement, because you have a greater amount of gap in the joint. This "slop" goes away as the joint is tightened. NPT joints don't necessarily seal on their own, and usually require sealant to fill slight gaps between thread crests and roots.

As far as thread engagement - have you measured how many turns of the part are made before the joint is "hand tight", or used NPT gages to ensure the thread form and taper are correct? A good NPT nipple will have about 3 full threads engagement at "hand tight".

Now you are pointing out a new problem, of the joints failing. Can you show a photo of the failed components? Is the end of the nipple left inside the female part (male nipple fails) or are the two components completely separated (threads failing in shear)? What pressure did they fail at and how long was the pressure applied? When it comes to polyethylene pipe nipples, I can't find any that are rated for pressure...

 

[99ishvr4]

btrueblood,

I should have stated that it is "more play than desirable". The existing products we are comparing it to feel much more secure as they are being tightened. No sealant will be used to seal the threads, as they are supposed to be quickly interchanged. Using thread gauges is one of my questions, is it acceptable to use standard steel NPT thread gauges on a plastic LDPE component? I feel that if the thread was out of spec, that the steel gauge would just deform the plastic and there would be no way of knowing if it was in spec.

I do not have pictures of the failure, but I do know that the component blew out under pressure because of poor thread engagement. It did not break into two pieces and leave threads behind. They are used up to 100 psi. I do not have the rest of the info as I am new to this project, which has almost no documentation to speak of.

 

[3DDave]

LDPE seems like such an odd material for this. An acetal would be what I would expect. Poor thread engagement may be that the parts aren't stiff enough to withstand the pressure

If the taper and thread form are OK then it is down to the material being too soft for the application, regardless of the thread form.

In a quick Google search I see LDPE only in very low pressure irrigation and push-on fittings; mostly the material is used for the water lines.

In contrast there are acetal threaded pipe fittings with NPT threads.

 

[Compositepro]

NPT threads are not designed to seal without a sealant.

 

[99ishvr4]

3DDave,

I don't think the material is the issue because products like this already exist and have existed for a long time:

example:

https://www.adhesivedispensing.net/TS616_Nozzle_6_x_1_16_pk_10_p/ts616.htm

I believe it is most certainly the thread form that is the issue, and I am trying to figure out the best way to inspect/gauge the thread form, or to actually correct it in the first place. Right now, our molder is saying that the cnc'ed the thread form in the mold, but I am not confident the thread form is to spec, so I need to find a way to help them check it and correct it.

 

[3DDave]

The beauty of tapered threads is that when you turn them they all engage unless you specifically have a stop to prevent them from engaging. It's not a thread form problem. I expect if you duplicated that thread into mating metal parts it would survive just fine.

Pictures of the failed items, the drawings of the items, the test pressure the failure occurred at, the temperature of the items during the test, the material used to perform the pressurization - that's the minimum for more than a passing diagnosis.

 

[Tuckabag]

How many threads are left exposed when screwed in on your new part versus the "off the shelf" part?
The bottom thread looks quite short for an NPT thread, first two threads or so are missing, this may be by design for clearance in the mating part?
Any chance you are bottoming out the tip or the first thread of your new part when it is installed?

 

[RobWard]

Sort of on-topic(ish):
A elderly toolmaker friend of mine told me of when he was a specialist witness in a law suit...
Over time, a very large multi-national company making vessels had worked with their toolmakers to tweak the threads on the lids, supplied by a third party, to get the best possible seal.
Enter a new quality manager who spotted that the threads no longer exactly matched the original spec and ordered the tooling be altered so that future parts were to the drawing dimensions.
Well then, predictably, the lids started leaking, the multi-national sued the lid supplier and the lawsuit was, I'm told, for $$$ millions.
He told me this tale as a lesson that the parts need to perform, not just match the orginal design as drawn up by some guy in an office somewhere. It has made me speak to the end user and customer a lot over the following years to make sure the end result was correct. I know that when we do threads we end up making a lot of trial fits until everyone is happy.
Yes, in that case it should have been documented properly and procedures followed, but this was back in the 70's or 80's I think.
Just thought that might be of interest.

"I love deadlines. I love the whooshing noise they make as they go past." Douglas Adams

 

[Arnaud Divialle]

Hi 99ishvr4,

If the thread are molded, did the Mold designer account for LDPE shrinkage during the mold design?

From your images, it looks like most of the threads are reasonably acceptable, except for the tip. It might be a short shot (lack of plastic to fill the last threads)