It depends on the type of materials you are joining and the type of stainless steel. For example, austenitic stainless steel (Type 304, 316) that is in a solution annealed condition can locally yield under tensile stress because of the relatively low yield strength. The micro yielding at stress risers, like threads, can relax the preload over time and cause leaks.
Stainless is far tougher on threading dies and other tooling than carbon steel, so the surface finishes of the threads are often worse because people use dull tooling. That goes for both the people making the fittings and the people threading the pipe to connect them.
Stainless has roughly twice the linear coefficient of thermal expansion when compared with carbon steel. If you use a thread sealant which isn't compliant enough to continue to seal despite movement from thermal expansion and thermal cycling, leaks will result. Use a good quality paste-type anaerobic pipe thread sealant in addition to or instead of teflon tape and most of your problems will go away.
Stainless is easier to "gall" than carbon steel, hence people don't pull stainless steel fittings up to the same extent during tightening. A tighter joint is less reliant on the thread sealant to prevent leakage.
Then there's the quality of the 150# cast stainless steel threaded fittings on the market. Most of the ones on the local market here are from China, and the quality is hit-and-miss. We've seen some with huge angular tolerances on the threading, such that concentric reducers and 90's aren't (co-axial or 90 degrees that is), and branches of tees go off at wierd angles etc. Of course if you compare these to 3000# forged steel fittings, there's no comparison in quality. But comparing these fittings to 150# malleable iron fittings, the quality is similar(ly poor).
But the biggest single problem with threaded stainless steel as a fitting system is the union fittings. 150# or 3000#, none of them seal reliably to the limit of the threaded pipe, even if you lap the parts to one another before fit-up. Carbon steel unions are much more reliable, especially if they come with a brass seat.
Threading is a flexible, inexpensive and reliable way to join small diameter pipe, but it gets a bad reputation because people don't understand how the system works and hence do it poorly. If you never forget that an NPT joint is a spiral path connecting the interior of the pipe to the exterior world, and the only thing between you and what's in that pipe is the thread sealant you choose, you will be less likely to go wrong in using this method regardless what material the pipe and fittings are made from.
Great post, moltenmetal, although I take exception to one thing. At least one anaerobic pipe sealant maker (Loctite) says that using teflon tape WITH their dope interferes with a proper cure. Here was their response to me when I asked about using tape with their 567 anaerobic dope in 2003:
"The pipe sealant 567 already has PTFE in it to aid as a lubricant, the Teflon tape is nothing more than a thread jammer. It can interfere with the 567 curing properly because the tape may create a barrier from the metal. Anaerobic products cure in the presence of metal."
I asked them this question after seeing many contractors using the Teflon tape with the expensive anaerobics. I wanted to make sure that they weren't doing more harm than good, and apparently they were.
By the way, I've found that the 567 Loctite does an excellent job on threaded stainless when subjected to temperature swings of 200F.
As well, many people fail to read the instructions for various dopes & tapes. There are very often limitations on pipe size and/or temperatures. Most are limited to 2" pipe and under. Loctite makes a wide range of sealants, but the vast majority - if not all - run out of gas at 300*F meaning that they're unsuitable for steam service over about 50 PSIG. We've even hit some dopes that SAY they're good for, say, 400*F steam service but in fact, are not. The nice light blue colour that came out of the can cooked-out to beige in about 10 minutes once the 353*F steam hit it. It frequently leaked, and was ALWAYS a treat to get apart since it virtually vanished, except for a light residue that assisted in galling. We only tried it on a couple of jobs, and it was horrible even on factory sch 80 A106B nipples and 3000# FS fittings. We're using up the last of it on some compressed air piping.
Our experience has been that the presence of steam makes the service much tougher on the anaerobics. Non-steam services to 400 F are frequently no problem, but steam services beyond 300 F can be troublesome especially for larger sized joints.
For temperatures beyond these limits, our next option is grafoil tape, top-dressed with graphite flakes in heavy oil (UCAR markets a brand of "thread sealant" which pretty much is just that. It works well for small, carefully-threaded joints.
After that, all we're left with is Xpand-o. Very leak-resistant, but sometimes you wish you'd just given up and called the welder over. If you use that stuff, make sure you don't thread the components up to full wrench tightness or else there will be insufficient sealant left in the joint to expand and seal properly.
As to the point about using a combination of tape and anaerobic paste: our experience has been that joints with both seal more reliably long term than joints with either tape or paste alone. We theorize that the tape offers better ultimate galling resistance than the few PTFE particles in the anaerobic paste alone can offer, plus it acts as a creep-resistant large void filler. The paste acts merely to provide elasticity in the joint to compensate for dimensional changes due to thermal cycling. But properly applied, paste alone is far better than tape alone, any day. The only time we use tape alone is when the service and the paste aren't compatible with one another.
TBP is dead right: some people wrongly assume that materials like 567 are basically PTFE in paste form, but nothing could be farther from the truth. Get the compatibility chart!
As to the presence of PTFE tape reducing the ability of the anaerobics to "cure"- there's plenty of exposed metal on the female side of the joint! And the stuff may "cure", but it never sets up hard- which is exactly why it works!
Thanks for all the replies. Great information. Makes me think about threaded joints in a different way.
Has anyone had any experience with the combination of Lotite 55 and Leak Lock? OK so far during testing(no leaks!). It sure does extrude a lot of material while tightening the fitting.
We have been having the same problems with leaking at SS threaded joints on fittings larger than 2" in general. Nearly all of the 3" diameter nipples and fittings are leaking especially the unions. Many time is seems that the SS threads sieze and you just cant crank it any tighter even with a 48" pipe wrench. We have tried a number of combinations of dopes, pastes, tapes and other sealants. We ended up welding several of the locations. Looking to go to Victaulic type grooved fittings. But I may try the Loctite 567 even though we do not have the high temperature issues. Is was unclear what the general concensus was on using the anaerobic sealant (#567) combined with the teflon tape? One "unique" issue we have is that the threaded fittings are installed on an HDPE pipeline so they are subject to the thermal expansion/contraction of the pipe. Any additional thoughts are welcome. We are looking into several ideas such as expansion joints and/or loops to tame the HDPE.
HDPE doesn't bond to anything, so I wouldn't expect much from the Loctite. Maybe you could heat/spin a flange on the HDPE and use a gasketed flanged joint with steel backup flanges. And Belleville washers on the bolts.
Threading should be limited strictly to 2" and smaller. For low pressure services larger than 2", especially when dealing with dissimilar materials, Victaulic grooved is the way to go. Heavy-walled HDPE can be grooved successfully.
I don't think that Henkel Loctite 567 is recommended for plastic fittings of any kind. It may be as another poster previously stated, that the material relies on metal exposure to catalyze curing. But some of the anaerobic materials induce cracking in plastic fittings, so you should read the vendor's literature carefully. Loctite makes a sealant called "No More Leaks" which definitely is suitable for use on plastic fittings.
Softness in the parent material. Stainless threads are prone to galling, material transfer and pickup. They are typically reused only once or twice.
Try using a thread sealant or anti sieze compound. This tends to add lubricity to the joint in addition to sealing capability.
Also, most field technicians tend to overtighten the joint, you know, it's leaking...tighten some more...still leaking...tighten some more....still leaking....give her another turn. I wish I had a nickel everytime this happened! I'd have about two bucks by now.
Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada
We use transition fitting from the HDPE pipe to stainless steel NPT then thread on check valve, ball valve, union, then transition back to HDPE. Although the 3" is a bear to tighten, as mentioned there is a tendency to overtighten. back when I was doing fire sprinkler system design/installs we would sometimes have even 4" diameter threaded pipe. I suppose the galling of the SS threads is the most likely culprit, but with the thermal expansion issues with the HDPE, the grooved approach might be a better alternative anyway as they have some defection tolerance. We might give the loctite 567 a shot anyhow. At pressures below 50 psi, most of the leaks have subsided for now.
