Verifying Bolt Pretension

[connectegr]

Any suggestions on how to verify correct bolt pretension 18 months after installation?

Connections are galvanized and slip critical. In some cases with OVS holes. Erection was completed nearly 18 months ago, and an issue has been raised concerning the bolt installation and whether the appropriate pretension was provided. Bolts are 7/8 A325 and 1 1/8" A490. They were installed by turn-of-the-nut, but most of the marks are no longer visible. And some of the bolts in question clearly lack the appropriate pretension, even though the marks look correct.

I am familiar with the arbitration method provided by RCSC. But even in ideal conditions torque is not a good representation of pretension. Exposure to the elements and loss of lubrication will may torque value scattered.

Thanks

http://www.FerrellEngineering.com

 

[rowingengineer]

As you suggest the torque method would be variable for your situation, I put forward that maybe the influencing factors that will scatter the results will only ensure that the torque is higher than the calibrated device (this is a thin argument). I would calibrate a wrench on a few new installations where bolts can be removed/reinstalled without causing harm. I would then proceed to check each bolt for torque v’s calibrated pretension.

An expert is a man who has made all the mistakes which can be made in a very narrow field

 

[connectegr]

If the torque is calibrated on new installations, thus new properly lubricated assemblies, the torque required would be much less than the torque required for the aged/poorly lubricated assemblies. This is my problem.

rowing...
Your suggestion is similar to the RCSC arbitration. But, this is intended for use relatively close to the installation date of the bolts in question.

(It is curious that RCSC lists numerous variables that can impact the consistency of using calibrated wrench method for pretension. But then provides torque as the only method of arbitration in a later chapter.)

http://www.FerrellEngineering.com

 

[paddingtongreen]

"(It is curious that RCSC lists numerous variables that can impact the consistency of using calibrated wrench method for pretension. But then provides torque as the only method of arbitration in a later chapter.)"

That is the major drawback of the short thread lengths on A325 and A490 bolts, I looked into a direct tension installer/tester but it needs enough thread to be able to pull the bolt with full load.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[ToadJones]

Any chance some of the materials, including galv have relaxed a bit in that time thus throwing off a check even further?
I remember reading something in RCSC that mentions this.

 

[connectegr]

RCSC does provide information concerning retightening of galvanized bolted connections. But this should have been done months ago. I don't know if this still applies after 18 months.

http://www.FerrellEngineering.com

 

[rowingengineer]

connectegr,
I'm in Australia so the RCSC arbitration means little to me . If you are concerned about the lubrication factor, you could remove and reinstall the existing bolts calibrating only the existing. Or you could do some testing and see what the differences are on a few different assemblies to get a fug factor.

However if you are looking for a write back to the existing installation for comp, you best choice is to set the lower limit and see how many fail, no one can argue the comp then.


An expert is a man who has made all the mistakes which can be made in a very narrow field

 

[paddingtongreen]

"They were installed by turn-of-the-nut, but most of the marks are no longer visible. And some of the bolts in question clearly lack the appropriate pretension, even though the marks look correct."

Snugging out the unevenness of large gusset plates is sometimes a problem with turn of the nut; at which point do you say they are snug and place the starting marks.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[rscassar]

I didn't think there was a direct relation between applied torque and bolt pretension.

Would it be possible to un-install each bolt one-by-one and re-install new bolts with load indicating devices (either load indicating washers or bolts).

I am actually surprised by how little construction workers are aware of tensioned bolted connections. When I indicate on the drawings that the bolts are to be TF (tensioned friction), it means nothing to them and they install the bolts as they would for snug tight.

 

[connectegr]

I agree that torque and pretension are not directly related. But, torque is the only method I am aware of to review a completed installation. Replacing all of the bolt assemblies is certainly a posibility, but an enormous task. This is 10,000 ton industrial project the erector should have understood the requirements for installation of slip critical connections. These connections are identified on the erection drawings and bolt lists.

http://www.FerrellEngineering.com

 

[desertfox]

hi connectegr

I might have misunderstood the context of your statement that pretension and torque are not directly related, but have a look at this site:-

http://www.exprobase.com/Default.aspx?page=841

desertfox

 

[ToadJones]

Pad-
Read the first few paragraphs of the attached:

 

[connectegr]

desertfox...
Your reference is based on an ideal installation condition. I agree that if all other variables can be controlled and repeated that torque may provide an acceptable method of verifying pretension. I am familiar with other construction applications which prefer the use of torque values to establish pretension, and use torque controlled bolts to control the maximum pretension.

But, for steel construction torque is not representative of pretension. There are several variables that vary from one installation to the next. Lubrication, degree of contact between connected plies, repeatability of the field equipment, etc.

This said, torque has been the only suggested method to verify completed installations.

http://www.FerrellEngineering.com

 

[paddingtongreen]

Toad, thanks. Actually, I wrote something like that as a guide, not as well written, back in the early seventies where I was the de facto connections expert for a large company (the nominal expert, a PhD would send people to me for help).
What I was getting at was that snugging has to be repeated, like when you put a wheel back on your car, you snug the first nut, then the third, fifth, second, fourth...and when you get back to the first, it is loose, so you go round again before final tightening. It seems the more bolts, the more times you have to go around.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[connectegr]

RCSC provides guidelines for sequencing of tightening to assure that all bolts are correctly pretensioned. This sequence is regardless of the method used.

http://www.FerrellEngineering.com

 

[paddingtongreen]

connectegr,I know that RCSC covers the sequencing, but it comes down to judgment as to how many circuits are needed, and money affects that judgment so steps are missed. That may be why some of your bolts are showing loose.

The only thing I can come up with, is, after you have reinstalled the obvious loose ones, instead of the torque wrench method, back off the nuts lubricate if necessary,and reinstall by turn of the nut using a different color marks.

That doesn't work for the galvanized ones, that needs more thought, if it is even doable, the zinc is subject to galling.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[desertfox]

hi connectegr

Thanks for your response I agree there are a number of varibles that affect the preload.
Here are some accuracies given for various methods:-


Operator "Feel" +/- 35%
Torque Wrench +/- 25%
Angle Torquing (Turn of nut) +/- 15%
Load Indicating Washer +/- 10%
Measuring Bolt elongation +/- 5%
Hydraulic Bolt pretension +/- (1% to 10%)
Strain Gauges / Ultrasonics +/- 1%
Superbolt multi-jackbolt tensioners +/- 5%

Now I notice in one post it mentions backing the nuts off before retightening, if its possible you could measure the contraction of the bolt before retightening and therefore estimate the original preload but i guess that won't be easy.

desertfox

 

[ToadJones]

How much can one do with these bolts before they are considered to be "re-used"?
RCSC does not allow the reuse of galv bolts, right?
I know "touching up" or "retightening" of adjacent bolts is allowed but how much further can you go?

 

[connectegr]

RCSC does recognize that hot-dip galvanized coatings may require re-pretensioning of the bolts. This condition is not considered re-use. But, I believe this is expected to occur during the erection process or relatively close to installation. And re-use of galvanized bolts is not allowed. They are considered assemblies, so the entire assembly must be replaced (bolt, nut, and washer).

In my present condition, I cannot verify if the bolts were ever tension properly. 18 months after completed erection, I cannot determine the cause. At this time I am suggesting that all loose bolt assemblies be replaced.

I read an article concerning erection and bolting in windmills. I read that in many cases the bolted connections require inspection on 6 month intervals and re-pretensioning is required. I believe these connections are "pretensioned", but not "slip-critical". Thus the pretension must be maintained, but the resistance to slip is not an issue.

http://www.FerrellEngineering.com

 

[paddingtongreen]

I agree, connectegr, the adjustment is intended to be within the construction period, before turn over. As to replacing the loose ones, you might find that other adjacent bolts have become loose when the new ones are fully torqued.

Michael.
Timing has a lot to do with the outcome of a rain dance.