Welding High Strength Fasteners

Date 02-21-2003 12:35

I would think it would not be much different that tacking a nut to the inside of a tube that would be centered over the hole. We run a bolt thru to make sure the nut has not distorted from the heat. Usually a small tack is all that is necessary to keep the bolt/nut from turning. I guess your question is really for what welding procedure to use? I guess I'm now cunfused by what it is you are asking for. 1) Are you asking for a procedure to weld the fasteners? 2) Or are you asking about a procedure for field welding the joint in place after erection? 3) Is the bolts only for a temporary connection until the erector can field weld it or will they be the permanent connection? Being that there are only two bolts figured into this connection, it is either a temporary connection or a very small end reaction.
John Wright

By jwright650 (*****)

Date 02-21-2003 12:45

After thought,
The Skidmore-Wilhelm devise may prove your connection. Mock up the connection and the fastener assemblies and prove thay you can achieve the proper pretension in the bolts on the structure, rather than proving the fillet welds that it took to hold the fastener from rotating. 3/4"dia. A325 bolts should have about 29kips of tension to be properly tensioned.

Or am I missing the boat with this one?
DGXL, I need your assistance,
John Wright

By MBSims

Date 02-21-2003 13:52

John,

We did similar qualifications to early editions of D1.1 using A307 Gr. A nuts welded to the back of embed plates. We had to qualify the maximum size single-pass fillet size and the minumum size multi-pass fillet size. The test plates were 6" wide x 12" long x 1" thick A36 plate with (3) 1" nuts fillet welded on each plate. The nuts were evenly spaced 3" apart along the center of the plate. You could also put them on seperate 6" x 6" plates if you wanted. We did not drill the bolt holes since their presence did not affect the welding. There was one nut for each test position (1F, 2F, 3F, 4F) and one for each fillet size (1 for the max single-pass size, another for the min. multi-pass size) for a toal of 8 nuts. We welded the full circumference since that is how we welded them in the field. You may only want to weld across a couple of flats on the nut if that is all you plan to do in the field. If you're only tack welding, that should be sufficient. The testing consisted of visual examination of the welds, then cut a cross-section through each nut in 2 directions (90 degrees apart), polish with a 3M buffing wheel, etch with nitric acid, then macroexamination per the procedure qualification macroexamination paragraph in D1.1.

Marty

By MBSims

Date 02-21-2003 14:14

I should probably also mention that A194 Gr. 2H nuts have 0.40% "minimum" carbon. Their properties are developed by heat treating (quenched and tempered). A563 Gr. C nuts have the same heat treatment, but have a max. carbon content of 0.58%. If you weld them, you are affecting the heat treatment. You should also consider microstructural examination to determine if the weld is creating an undesirable microstructure and obtaining raw bar stock used to make the nuts and welding some coupons for tensile and bend testing.

Marty

By GRoberts (***)

Date 02-21-2003 15:13

Instead of tack welding the bolt, if this is going to cause issues, you could try welding a channel shaped piece over the top of each bolt made from the same material group as the base metal. It would keep each bolt captive and prevent it from turning.

By jwright650 (*****)

Date 02-21-2003 15:52

I like your idea, where do you get your hands on this item?
John Wright

By jwright650 (*****)

Date 02-21-2003 15:53

I assume they are mfg'd somewhere?
John Wright

By GRoberts (***)

Date 02-21-2003 17:29

I don't know of anyone who manufactures them, but an easy way would be to have a long channel bent up on a press break, and then saw them to length as required. You would have to have a different size channel made for each size bolt of course.

By jwright650 (*****)

Date 02-21-2003 17:31

GR,
Think there is a market for one if we make some?
John Wright

By GRoberts (***)

Date 02-21-2003 21:56

I don't ever remember bieing in a situation where they might be required, but that doesn't mean other people wouldn't be. I would think talking to a few structural steel inspectors about what they have seen people do in hard to access areas, and how often it occurs would be a good way to get an idea. The hardest part would be getting enough people to know about it that you would sell enought of them to make it wothwhile. I don't really know how you would go about that.

By DGXL

Date 02-21-2003 17:57

Interesting post.

We have in the past rejected any HSB (A325) that has been welded or tack weld. The building depts. in these parts would hang you if they knew A325's were being welded or modified in any manner. (I have seen some unusual ideas by erectors over the years.) There are documented cases of high strength fasteners failing after being subjected to heat.

The concept is to develop and verify tension of the fastener. As John Wright mentioned, a Skidmore would prove the connection, it is how the faster is verified for tension. It wll not verify weld strength. The required tension should be approximately 70% of the minimum tensile of the fastener. John T is also correct in that D1.1, 4.11 would not be applicable because the bolt head or nut welded to the plate is not a T-joint, it is a lap joint (not speaking about the connection itself).

If they were 307 fasteners like MB talked about I would say go for it, but there not. What about using an adhesive on the nut prior to erection? Loc-tite and other companies make some serious adhesives. We have used this method several times and it was sucessful each time provided no one messes with the adhesive until it completely cures.

You also need to make sure you don't compromise the faying surfaces of the connection, or whatever you do will be in vain. This is probably a slip critical connection if their using HSB's. Is this a Class A, B or C connection type?

p.s. John Wright: I got 28 kips for that 3/4" A325, what table did you get that number from?

By jwright650 (*****)

Date 02-21-2003 18:20

The figure I gave has a 5% increase above the table you have for an A325 bolt, that you need to prove on the Skidmore gage during pre-installation verification per the RCSC, July 23, 2000.

We use A325 "TC" (Tension Control "F1852") Bolts that are tightened from one side using a gun that holds the spline and turns the nut until the correct tension is achieved and then it wrings off the spline.

You will still need to find a way to keep the bolt from falling out when the connection is being made up in the field, but it would eliminate the need to hold the bolt head from behind.
John Wright

By jwright650 (*****)

Date 02-21-2003 20:14

Just thought of something else,
If you used a TC bolt, could you tack weld a piece of flat bar(same grade as base metal) across the back of the bolts (do not put any weld on the bolts) to hold them from falling out the back, during erection? The flat bar would serve no purpose other than being a keeper for the bolts.
John Wright

By jwright650 (*****)

Date 02-21-2003 20:41

DGXL,
Section 7.2 RCSC July 23, 2000 says to develope 1.05 times (29.4 kips)of the specified values in Table 8.1.(value in Table 8.1 is 28 kips for an A325 3/4"diam.)
John Wright

By DGXL

Date 02-22-2003 18:33

John,
Thanks again for jogging my noggin, it has been awhile since I've been in the fastener section of the codes. The +5% has always been required in the AISC manual, 9th Edition, page 5-274(2) for calibrated wrench tightening.

By rpoche

Date 02-22-2003 00:28

Here is a post from a Materials Engineer on another Forum responding to welding a HS bolt.

Post Below.

"Depending on what standard you reference, and what strength level is required, high strength steel bolts are manufactured from carbon or alloy steels, with carbon content varying from ~ 0.20 to 0.55. In the quenched and tempered form, these alloys provide high strength and good toughness.

Welding involves extreme heating and localized melting which produces the following:

1. Brittle weld, due to the presence of untempered martensite. The untempered martensite is produced because of the rapid cooling that takes place after welding in thick sections.

2. Overtempering of the heat affected zone (HAZ). Again, depending on the chemical composition and strength level, bolts are tempered at temperatures ~ 300-500 C, and this temperature is easily exceeded in the HAZ, which causes local softening.

3. Severe crack sensitivity, as tempered martensite is quite prone to hydrogen cracking and quench cracking.Depending on what standard you reference, and what strength level is required, high strength steel bolts are manufactured from carbon or alloy steels, with carbon content varying from ~ 0.20 to 0.55. In the quenched and tempered form, these alloys provide high strength and good toughness.

End of Post

We have welded HS bolts, with the Design Engineers approval (signed RFI), but they were always in compression not shear.

Couple of Ideas.

1. Weld piece(s) of cold rolled next to the bolt flat to keep bolt from turning, if using turn of nut method, Then mark and tighten as required

2. If snug tight connection only - impact is acceptable

This post is interesting!!!!

By CHGuilford (****)

Date 02-21-2003 19:05

At first glance, your problem seems simple enough. But, getting deeper into it, I almost shudder at the thought of having to deal with this myself someday.
Like John Wright, I'm not sure if you are trying to develop a WPS for welding the bolts to structural steel or trying to verify adequate strength of the bolt after the head is welded. Also, I'm not sure if the bolt is going to be thru the steel and then tacked or if it's to be welded on the circumference like a stud would be welded if no studgun was available (never mind, your item #5 clarified that).

We all know that it is common to weld nuts or bolts to prevent turning in limited access areas. I wouldn't worry as much about how to weld it other than to record how you do it. I would think you would need to know how strong the joint is after welding, and unless you really soften up the bolt, I doubt it will pull through the hole. A few things come to mind.

1) Test the bolts with a Skidmore and torque wrench to find out the torque required to achieve the Minimum Required Bolt Tension (commonly called the inspection torque). Make your test assemblies and put spacers over the bolt shank so you will tension the bolt when tightening the nut, similar to D1.1 fig 7.3. Torque to the value established in the Skidmore ( or beyond by a reasonable percentage). Use bolts, nuts, and washers from the same lot numbers to be used in production.

2) Test some assemblies to destruction and see where the failure occurs. Again, use spacers to help simulate actual conditions. If failure was in the bolt shank, threads, or nut then you know the welds shouldn't be a problem.

3) Another thought is to run the nut onto the shank (no spacers) and apply the torque value mentioned in #1 to see if the welds will shear or not.

4) Redesign the connection with a thicker plate that can be drilled and tapped.

The fillet weld test from 4.11.2 doesn't determine the strength of the weld, but it might be useful to check for cracking. I would weld a bolt to a plate and saw through the plate and bolt cross section as MBSims said. However I think any cracking would tend to be within the bolt hex surfaces rather than in the plate.

Chet Guilford

By John T. (*)

Date 02-22-2003 16:50

WOW!!
What else can I say, thank you everyone for the prompt and professional responses. I would like to thank each and everyone individually, but that would be like 6 posts. I'm thankful for the chance to hear from each perspective.

Would also like to say one of my original suggestions was to weld a "keeper" along side the head of the bolt, consisting of basically a tab of lighter gauge material bent over the head, or a small angle along one side of the bolt to trap the bolt and prevent it from turning.
But my Engineer shot down this idea. It is definitely the most practical of any solution I see.

Again I thank each of you for the time invested in investigating my problem.

John T.

By MBSims

Date 02-22-2003 17:13

The macroetch is only to verify no cracking and complete fusion exists. The only function of the tack weld for this case is to keep the nut from turning until the joint is tensioned. If only the weld fails, there is no conseqence as long as the nut does not turn before proper tension is obtained. Even if the tack welds cracked, the weld deposit would probably still prevent rotation of the nut. So the main concern should be whether the welding has degraded the heat treatment of the nut enough to cause it to fail in service (or while steel erection is in progress), or does it introduce cracks in the nut that may cause it to fail and lose tension in the joint. If microstructural exams show no cracking or untempered martensite, and tensile + bend testing of welded raw stock (in quenched & tempered condition) produces acceptable reslts, then tack welding should be allowed. Based on the specified carbon contents, there probably will be untempered martensite present in these materials. Proving that is will not degrade the nut properties enough to result in a joint failure may be more effort than it is worth for your application.

Marty

By DGXL

Date 02-22-2003 18:41

To all:
Designers are increasingly prohibiting any weld other than those specified in the area of any moment connection. This would include "keepers" or "tabs" to retain the fastener.

This is due to the many failures after Northridge where welders had tack welds outside of the joint, clips for hanging equipment, cables or tools, etc. Many failures were due to miscellaneous welds/tack welds on or near a connection designed to be rigid. I talked to 2 engineer friends (both SE's) and they mentioned they would never permit any welding of HSB's under any circumstances. Their opinions only.

By jwright650 (*****)

Date 02-24-2003 00:25

DGXL,
How close to the bolt could you weld without detrement to the base metal around which the bolt is acting as a fastener? I don't want to ever suggest something that might end in a disaster. We use lots of TC bolts and there are times we need to "keep" them from falling out during erection. I was totally against it, but recently, I had an engineer insist on welding the heads of those bolts to the underside of cap plates on Wide flange columns (for future expansion of the structure at a later date). I was afraid of what could happen after altering the use of those bolts. I am afraid our company has now assumed all responsiblity for those bolts and relieved the bolt mfg of any liablity should failure result from the tack welding. I have stated my opinion clearly and loudly, but the fabrication continued as the engineer suggested.
Any thoughts?
John Wright

By DGXL

Date 02-24-2003 01:01

Hey John:
I would retain any RFI, transmittal, etc. in which the EOR provided instruction about weld the HSB's. That is what you will use as a defense if there are any failures. I have a few "modified" HSB's which I use for my classes as "how not to install HSB's." The "safe distance" for any welds near a rigid connection would be determined by the EOR, I could not give you any number. It is all based on calcs.

Welds are also being prohibited in the diaphram area adjacent to any moment frame connection. This includes deck welds, temp. attachments, studs, power driven fasteners, etc. You'll find this in FEMA 353 as well.

I saw some photos a few years ago of SMRF's with cracks in the WM, BM or both eminating from tack welds in the vicinity of the connection. Some of the engineers have the actual specimens with the flaws.

As far as welding HSB's, it is pretty much prohibited as I mentioned here in seismic zone 4.