I've only ever tig and stick welded SS pipe. Socket welds or open butt. I do know if the procedure is allowed you can mig weld both joints. But the only time I see that done is by the guys in the fab shop who are using positioners. They build spool pieces that get field welded. But if they were chasing leaks from the start then it probably will be an ongoing issue. Could have been from bad setup or bad practice.
I have more questions than answers
Josh
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS
FIRST NAME
*
LAST NAME
*
*
MESSAGE
*
ADDITIONAL DETAILS
Thanks. We have received your request and will respond promptly.
Log In
Are you an
Engineering professional?
Join Eng-Tips Forums!
*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.
Promoting, selling, recruiting, coursework and thesis posting is forbidden.
Eng-Tips Posting Policies
Stainless Steel Bolting Questions
thread725-321096 Forum Search FAQs Links MVPsForum
Search
FAQs
Links
MVPs
(Materials)
(OP)
30 Apr 12 13:06I have A193 Grade 8 (low yield stress) stainless steel studs which may have been stretched. What kind of gauge would detect that and who might have one? I kind of doubt that metallography would capture it, but all opinions and experiences are welcome.
In general, how is torquing of low YS bolting done? Personally I would avoid them altogether by going to a higher YS variety, resizing if necessary.
Perhaps someone could provide some explanation for the myriad variations on Grade 8 bolts in A193.
(Structural)
30 Apr 12 13:29How about buying new ones??
(Materials)
(OP)
30 Apr 12 14:33We will, but that doesn't answer the question.
These bolts were on a pressure-containing flange that did not contain pressure. We don't want that to happen again.
(Mechanical)
30 Apr 12 16:16brimstoner, don't why there are so many varieties of SS bolting. If using 304 I will nearly always use B8CL2, good in small diameters, but stresses go downhill quickly as diamater goes up. As you noted the yields are pretty low. Not much you can do except to avoid yielding the bolting.
I just about never use SS for external bolting unless it is known to be in an external corrosive environment. External bolting is generally not considered to be exposed to the process.
Don't know any way to detect yielding by physical measurement unless overall lengths were know precisely beforehand.
Regards,
Mike
(Materials)
30 Apr 12 20:34Here is an example of a simple stretch gauge that is used for measuring bolts before and after installation/preloading:
(Mechanical)
30 Apr 12 20:54"These bolts were on a pressure-containing flange that did not contain pressure."
The flange leaked? they may have yielded, but If the bolts were not torqued to apply clamping force sufficient to resist the force the joint will open.
(Materials)
(OP)
30 Apr 12 21:58The joint could have failed for any number of reasons. I've only just been called in to help, and that is one of the speculations reported to me. All I know is that no studs were broken.
(Mechanical)
1 May 12 01:18I assume the stretching took place on the threads. Couldn't a direct measurement of the thread pitch give a rough idea of strain?
David
(Mechanical)
1 May 12 01:25Why do you suspect that the studs elongated or stretched rather than the nuts backing off or loosening?
(Mechanical)
1 May 12 08:14I worked for a company that had made a few hundred specialized hydraulic cylinders as part of a spindle drawbar release mechanism.
One day the design changed to require a little extra hydraulic pressure. On the test bench when the higher pressure was reached according to the test tech, the cylinder "farted" and made a big oily mess. A check of a few pressure/area/force//clamp numbers confirmed the lightly torqued bolts' clamp preload was likely exceeded, leaving the stiffness of the bolts (not the stout main housing) left alone to controll the cap position. The name brand US made socket head cap screws stretched elastically (not yielded) a few thousandths of an inch which was enuff to unload the o-ring. The "solution" was to torque the existing bolts tighter than the customary "2 hands on a T-handle allen wrench". Loctite would not have changed what happened at the low torque.
(Mechanical)
4 May 12 11:56Unless you knew the length of the parts before installation, I don't know of a way to check for yielding after use. The amount that they will have yielded will be so small that no thread gage will pick it up.
I agree with others that it is likely that some other factor caused the problem. I would look at the gasket stress to see if it has the correct resilience and I would also look at the tightening sequence; flange are notorious for bolt cross-talk issues.
(Materials)
(OP)
4 May 12 13:16Screwman1,
I agree a thread gage can only hope to pick up deformation that is probably already obvious in other ways. At the size I am looking at, the gage covers only 3 threads anyway. My best hope is to assume the pitch was correct to begin with and try to capture elongation over as much length as possible.
Installation was done by impact wrench, almost certainly uncalibrated, following a procedure no more sophisticated than that used at your local tire shop.
BTW I don't permit my tire installer to use an impact wrench. I have been stuck on the highway with a flat I could not change because the nuts had been so severely reefed.
(Materials)
(OP)
9 May 12 11:01Update:
As expected, my brand new thread gauge did not detect anything. It does not help that in the larger thread sizes the gauge only covers 3-4 threads.
However, I have found one stud to be bent by about 0.050", located at where one of the nuts seated. What does this say about the flange assembly practice (or rather, malpractice)?
(Mechanical)
9 May 12 12:10brimstoner,
I recommend using a thread pitch micrometer. Check the potentially yielded bolt against a new one, preferrably from the same manufacturer. The yielding action would have necked down the thread pitch. It is not usually uniform across all threads, but centered on one area.
(Mechanical)
26 Jun 12 03:12 . Then, apply at least a modicum of "control" to your bolt tightening process. Although torque wrenches are notoriously inaccurate in terms of applied tightening force to bolt stress, perhaps in your situation the relative quantum leap over impact wrenching will be sufficent.This is a common issue. First solution is to throw away the damn impact wrench!. Then, apply at least a modicum of "control" to your bolt tightening process. Although torque wrenches are notoriously inaccurate in terms of applied tightening force to bolt stress, perhaps in your situation the relative quantum leap over impact wrenching will be sufficent.In any critical bolting application where leaks canot be tolerated, even greater control must be applied to the process. A key factor is acheiving the proper bolt stress. This cannot be done by torquing alone since the relationship between torque and bolt "tightness" is highly dubious and capricious. To overcome this, one must measure each fastener's length before tightening and then again after tightening. Any fasteners which are thus found to be too loose or too tight are then torqued or tensioned more (or less) until they're within spec.The above takes the "guess work" out of the torquing process
(Mechanical)
26 Jun 12 14:17Just: With studs it is often not possible to measure stretch because one end of the fastener is not accessible. Torque angle is one alternative, but care must be taken to reduce (not eliminate) the effects of cross talk and interaction. Is a logical tightening pattern and multiple passes to bring the parts up to full load.
(Mechanical)
26 Jun 12 14:53 With studs it is often not possible to measure stretch because one end of the fastener is not accessible"Actually, this isn't the case. In fact, it's as easy to measure the relative elongation of studs as it is bolts. Furthermore, this is true regardless if either is in a blind hole or not. Our Technicians use specialized ultrasonic devices for this purpose. The instruments are effectively glorified "fish finders" working on a pulse-echo principle: A transducer is placed on the top of the fastener, it is pulsed and then "catches" the return echo a few nano seconds later. The time-of-flight is converted into a length measurement (after temperature and stress have been taken into account). Cool system and highly effectiveWe've found that "Torque-angle" is fraught with innacuracies since it depends on the accuracy of the initial torquing step. If the former is 'out', the subsequent angle step is also 'out'
(Mechanical)
27 Jun 12 12:45Hi Justanotherengineer,Do you know the make and model?How much does that ultrasonic device cost?I'm wondering about a few things.Does the end of the threaded end of a stud need to be finished better than these typical bolts ? -Does it need to be calibrated for different sized bolts, or different configurations of bolt heads, or fully threaded bolts?You mentioned stress is taken into account. A bolt is stressed along nearly its entire length. After the first few threads a stud is barely stressed at all. Can you enter a stress gradient?How much does it weigh, and how long does it run on fully charged batteries? Ot does it need to be plugged in?Is it sealed well enough to be used atop a piece of equipment inside a cement factory or a coal fired power plant or in a pumping station where a little water might splash or drip unexpectedly?Are the cables oil resistant and tough enough to be pulled around machinery and occasionally stepped on?Is the case "hardened" so it would withstand a few bumps as it was being slung from a shoulder strap as a technician climbed a ladder? Or, worse, a 3 foot drop?thanks,Dan T
(Mechanical)
27 Jun 12 13:31@ Dan (TMoose)Our Techs use these things in some pretty nasty conditions. However, speaking of "conditions", the Techs are also conditioned to have considerable respect for the devices! At approximately 7-8 K per instrument (including the necessary bits'n'pieces) we don't want them to get banged up. We find that they're fairly-well protected by the nylon working-cases. This also enables them to be slung around the neck. As such, they're quite portable and convenient to use. Based on which part of the globe a particular project might be, our guys simply tuck the units into a pocket in their coveralls when not using the units.The parts that one has to be particularly careful with are the cable connections. Step on them? No way! Sorry, but these are precise instruments. They're not made for knuckle-draggers. But, since even some of our Techs occaisionaly scrape their knuckles, we've made simple mods to our units to accomodate them.Yup: The bolt ends have to be prepped! They must be flat. Otherwise, one can't place the transducer properly. In fact, if one were to try to shoot fasteners with raised markings, one would be a very frustrated Tech indeed!Oh yes, they run on 3 AA batteries.You have to realize that these things require significant expertise to operate properly. We've encountered a number of situations where somebody simply bought one, tried to read the manual and then began to screw up the project. After facing significant frustration, it was obvious (after reviewing their recorded signals) that they "cooked" the numbers (easy to do if the Client or Inspection Agency doesn't know what they're looking for!). Keep in mind: Any monkey can buy a scalpel and read a manual but, would you entrust that ape to take out your appendix?Here's a link: http://www.heviitech.com/Hevii_UT.html
(Materials)
27 Jun 12 13:32The best ultrasonic measurement devices are from Micro Control Inc.:Other options with lesser capabilities are:
(Mechanical)
27 Jun 12 13:49 !! You cannnot use typical UT thickness guages or even UT flaw detectors for this specialised application. Attempts to do so have gotten more than one well-known NDE company into very hot water!The "best"? Perhaps but the "best" unit as described doesn't read bolt stress!! Youuse typical UT thickness guages or even UT flaw detectors for this specialised application. Attempts to do so have gotten more than one well-known NDE company into very hot water!The other two units mentioned are proper units. Just bear in mind that since these are (I'll repeat) precision instruments requiring particularly skilled operators (your standard NDE Techs don't make the grade because this is a very distinct and highly specialized process quite unlike what they're familiar with), make sure that wherever you buy your units the supplier knows what he's talking about. You'll surely run into situations in the field where you'll need some quick on-the-mobile-phone assistance. A simple salesman is not going to be much help to you in those inevitable under-the-gun situations!
(Mechanical)
27 Jun 12 15:29I agree with Cory, the Micro Controls is the best unit. We have used several of them over the years and they are versatile and tough as a brick. You can set them up for pretty much any type of fastener data acquisition that you need. We use a tensile pull to calibrate the load / deflection curve for a specific part, since that is the most accurate way of determining stress in the fastener. At that point the MC unit works like a charm. (if you have a trained operator; I'm still partial to Tq/ Angle for the vast majority of big bolt applications because of historic issues related to operator error with US).
Just one thing to note about using US on studs; you will have to do the end prep yourself if the parts are to one of the recognized specs, since they will have to be supplied by the mfg. with end markings per the spec.
(Materials)
27 Jun 12 16:03Justanothereng,
Compared to an MC900 system, all the other units are glorified ultrasonic thickness gauges. The MC900 system certainly can read bolt stress, from ultrasonic sensors and strain gauges.
(Mechanical)
27 Jun 12 17:35Ahh, I stand to be corrected.
(Mechanical)
22 Jul 12 19:25Throw out your used bolts, install new B8 bolts (preferably strain hardened class 2- and make sure that your supplier really does supply class 2), calculate the assembly torque required staying below 60% of yield if possible, then use controlled torquing following the legacy (star pattern) procedure outlined in ASME PCC-1-2010.
Problem solved.
(Structural)
22 Jul 12 21:23Can a sample be cut, polished and etched for a microphotograph? Looking for Luders Bands. This may show that the material has slipped... doesn't solve the problem, though.Dik
(Materials)
22 Jul 12 22:51I will have to agree and not to agree with the above posts.
The majority of our fasteners in the intermediates area are 304 SS or 316 SS.
We use impact wrenches to tighten all flanges using the parallel face method, Just make sure the faces come together in parallel. The impacts are for Ingersoll Rand and routinely refurbished and calibrated. The mechanics determine this point in time. They also change the torque based on fastener size up to 1" studs. Most of our fasteners are stud bolts Class 2 up to 1 1/4".
We have a design philosophy of many small instead of a few large on all flanges where possible..
Gore-Tex gaskets on Class 125 flanges. spiral wound can only be used on class 300 or above.
A lubricant is used on all studs, normally C-5A copper.
All studs for a job are reassembled in teashop and any problems the stud or nut is tossed, in other words no one tries to seat a bad stud/nut.
The mechanics will try one wrap with the backup wrench if one sticks, and if it doesn't go it is tossed.
No opened end tooling, if we can catch it.
We reuse all that we can, normally a pretty high recovery level.
The last time I looked at some problems the source we in the nut. In the case the lead was off and there were root/tip interference.
There have been various attempts to impart some trying o tightening requirements over the years and all have been a failure. All types of gauges have been tried to no avail. We even bought a Raymond UT bolt extensometer and an instrument by Branson, neither were practical in the field and laboratory work showing such a scatter that there was no practical correlation
There is a process that will reduce the scattered of SS fasteners and I hope to be able to work on it. In the near future, this will be an independent effort on my part. It worked well enough on the first passes that the statistician was able to calculate a correlation coefficient that matched what I me measuring by micrometer.
(Structural)
23 Jul 12 07:07Can you cobble together a device that has a dial gauge and fits over the thread end of the bolt and check the extension from snug to whatever pre-tension you require?
Dik
(Materials)
26 Jul 12 23:21Here is an example of many small bolts in lieu of a few large bolts. In this area we use Gortex Gasket material and by using this we have very good sucess with this approach. All manways, tanks and vessels are bolted this way. The mechanics like this approach as they don't have to drag big tools around. We use C-5A on all studs.. We also allow impact wrenches as the tightening procedure uses the parallel face method.
Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.
Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.
Click Here to join Eng-Tips and talk with other members! Already a Member? Login