WE HAVE RECIPROCATRING COMPRESSORS WITH CAST HIGH AND LOW PRESSURE CYLINDERS.THE THREADED JOINTS IN THEM ARE GIVING WAY AFTER TWO TO THREE TIGHTENING SEQUENCES.NORMALLY THE THREADS ARE GETTING SHEARED. IS IT A COMMON PHENOMENON WITYH CAST IRON? ARE ANY OF YOU ALSO EXPERIENCING SIMILAR PROBLEMS?WE ARE RUNNING HIGHER SIZE TAPS AND AGAIN USING THE CYLINDER. BUT THAT IS A RECURRING PROBLEM. ANY SOLUTIONS?

Common practice is to use a chart for selecting torque which is based on the material characteristics of the bolt material (not the mating hole threads). Specifically, most charts are set upt to give a stress in the bolt threads of some fraction (50%-75%) of the yield stress of the bolt material.

This approach is very common but pretty much ignores the characteristics of the piece that has the mating hole... and that's asking for trouble when you're using a weak piece such as cast iron.

So consider going at the low end of the band or your torque chart, or doing a more detailed analysis of the application to see if you can use a lower torque (of course OEM may have the best info).

This message has been edited. Last edited by: electricpete,

I would check with the manufacturer of this machine. It sounds like poor casting. But also could be from over tightening. Is this pipe threads? Or std. threads? Now adays the casting are not as good as they used to be. Also if you have machinist available they could probably tell you if the casting is bad or what kind of cast you have.

I was wrong when I said above the standard torque charts are based on yield of the bolt thread. The standard torque charts are instead based on yield of the bolt tensile cross sectional area (the root). The tensile force applied to the bolt root area can be determined by bolt diameter, threads per inch, lubrication conditions and torque. But that prload force when transmitted through the threads is spread out among the threads in a manner that depends on length of thread engagement. So the bottom line, bolt cross-sectional tensile stress (for a given bolt size, lubrication, torque) does not depend on length of thread engagement, but thread shear stress does.

Machinery's Handbook 27th ed pages 1510 and 1511 has some formulas that can be applied to estimate minimum thread engagement distance so that the bolt would break (tensile) before the housing threads would strip (shear). They indicate that it is preferable to design so the threads wouldn't strip until after the bolt breaks.
Very similar formula's are given here:
http://www.roymech.co.uk/Useful_Tables/Screws/Thread_Calcs.html

There is also a thread stripping calculator available for download here (but I haven't tried it):
http://www.boltscience.com/pages/download.htm

One last note threads should be coarse. Fine threads in cast iron are susceptible to stripping, per Machinery's Handbook.

When I worked as a machinist for a company that did work for defense contractors, many.. and I mean many parts made of aluminum, cast, etc would be specificed with helicoil inserts in all tapped holes.

http://www.emhart.com/products/helicoil.asp

Threads per inch (TPI) would are the same as standard, tap and drill are larger.
Drill and Tap , install helicoil, break off and retrieve drive in clip with magnet and your good to go.
High Strenth helicoils that will outlast tapped hole of many metals.

This message has been edited. Last edited by: MikeH,

Generally a tapped hole is 80% full thread and usually only two threads are engaged to carry most all load.

However, with the helicoil you get 100% full distribution over the full length of the helicoil and the helicoil is stronger than the original thread.

[QUOTE]Originally posted by Sam Pickens:
Generally a tapped hole is 80% full thread and usually only two threads are engaged to carry most all load.
QUOTE]

In general, the minimum recommended length of engagement is 1 to 1.5 times the
major diameter of the setscrew for threading in brass, cast iron, and aluminum; and 0.75 to 1 times the diameter for use in steel and other materials of comparable hardness.

http://homepages.cae.wisc.edu/~me232/tolerance_info/fastener_handout.pdf

What Sam is getting at with the '2 threads' comment is an irritating little fact that few know about and those who do wish wasn't true. When male and female threads are mated in tension the stretch of the male and compression of the female (in the case of a nut) is such that the vast majority of the tension is carried by only two threads. It doesn't matter of you have 20 feet of tapped hole with full engagement, the rest of it won't be getting appreciable tension unless the outermost two threads fail.
What I don't understand about Sam's message is how the helicoil magically makes this all better.

Hullo Akhtar,
Have you checked the vendor's manual or spoken to the vendor of the Compressor? I would suggest that before we redesign the threads and their torque values, it would be useful to know what the designer has to say about these failures. Understanding why exactly these failures take place is the key to finding the right solution.

I request you to use lower case as well in your text.

I don't know what kind of magic the heli-coils use, but I can say from experience they do in fact work.

Another thing I've learned from experience in racing... If you are having problems pulling threads out of a casting, try switching to a stud and nut combo (if possible) instead of a bolt. You get more even and consistant torque when using the stud and nut. Because the nut is turning on the stud while torqueing instead of the bolt threads trying to turn in the casting while torqueing. Plus, while you are torqueing a threaded bolt, if it's really long, the bolt is trying to twist (torrsional) while the threads overcome friction, causing torque fluctuations. This is less likely to happen with a stud and nut.

One more thing I've learned from racing... What kind of heat are the bolts and castings getting up too? More importantly, is it equal? Reason I ask is, as an example, on the VW 1600 air cooled engine, the head studs are steel and the case (block) and heads are aluminum. At one point in history, they (VW) installed a bigger (10mm instead of 8mm) head stud and they had problems pulling threads out of the engine case under racing conditions. Turns out the components had different thermal expansion rates, causing the effective torque on the head studs to increase beyond the failure point of the threads when at or above normal opperating temperature. In other words, make sure you have a normal heat range and that things are equal.

One other thing... Make certain that your toque wrench (or device) is calibrated. This is something very important that is often overlooked.

Bill,
I support your views on the helicoil design. Your suggested 'best practices' are also apt and useful.
But I urge all the readers to step back a bit whenever an unusual problem appears. Get a good understanding of why the item failed before looking for a solution, however attractive and fast that may appear. Stronger or superior widgets are not necessarily the right way to go, because they sometimes cause other parts to fail, sort of just getting up the food chain.
The RCA process requires us to gain such an understanding, using factual evidence to support our theory of the failure machanism.
Solutions usually appear quite quickly once we can visualize the sequence of events leading to the failure. Usually, too many solutions appear, and we have to decide which one is most suitable.