Welding Wide Flange Splices 4

[sedonas]

When you weld two wide flange together by cutting them square... would it make sense to further strengthen the web by putting cover plates over the web? Or over the flange?The purpose of flange is for tension and compression.. the web is supposed to be only for shear? Which to focus more in welding.. But if you weld cover plates on the flange.. won't it weaken the flange by having many welded regions.. how do you weld your splices (that involves no bolted connection)?

 

[sedonas]

Theoretically what's the biggest gap one can join/weld without using backing and using backing? Does it have to do with dynamics of molten metal and filling it and rate of propagation of heat to neighboring molecules? What dictate it in principle?

Also kingnero said above that "6011 is not to be used for structural welding". I'm confused.. what electrode then do you use for structural welding for A36 steel (not A992)? He doesn't give alternative. Why can't the 6011 be used for structural welding? We use it mainly in our country, so I need to know why before starting initiative to ban it.

 

[gtaw]

Generally, if backing isn't used, the root opening and root face equal the diameter of the electrode. Can it be wider? Yes, but it becomes more difficult to achieve complete joint penetration without burn through. Backing makes it easier to achieve CJP without burn through and it requires less skill on the welder's part.

As for the use of non low hydrogen electrode; ASTM A36 contains relatively low carbon and has a low probability for delayed cold cracking if the thickness is limited to 3/4 inch or less. As the base metal thickness increases, as joint restrain increases, and as the percentage of alloying constituents increase (increased carbon equivalency) the greater the potential for delayed cold cracking and an increased need to utilize low hydrogen welding practices.

The cellulose in the flux covering of E6010, E6011, E7010, etc. has an aggressive arc that drives penetration. It is the hydrogen that is produced by the decomposition of the flux covering that gives the EXX10 electrodes deep penetrating characteristics. But there are limits to everything. The ability to penetrate is not without limits.

At best, welding is a compromise. One must weight the advantages against the disadvantages to arrive with a solution that meets the demands of the project. There is no such thing as a perfect weld. All weld, by nature, contain discontinuities of one type or another. When the discontinuity is severe enough that the weld is not acceptable, it becomes a defect. One must consider the application to determine when the discontinuity is harmful to the performance of the weld. Here in the US and in other industrialized countries, they have developed and adopted welding standards that provide guidance to the engineer and the owner as to what practices promote welds that will provide the service and perform as required. Several people have recommended reviewing the requirements of AWS D1.1 Structural Welding Code/Steel. That is good advice. The AWS D1.1 structural welding code is a cook book of how to produce welds that will meet most structural applications. It is based on 100 years of "lessons learned". You can't go wrong by following AWS D1.1.


Best regards - Al

 

[kingnero]

Also kingnero said above that "6011 is not to be used for structural welding". I'm confused.. what electrode then do you use for structural welding for A36 steel (not A992)? He doesn't give alternative. Why can't the 6011 be used for structural welding?

See the post of gtaw, 30 Oct 15 01:56
See also the other posts, where it is said to get the relevant codes. although maybe not applicable in your country, the principles are valid.
See also the post where I suggested to do the test. What was the outcome?

He doesn't have a grinder.

This is actually a good thing, so he hasn't got the mean to dress up the welds afterwards.

 

[sedonas]

Generally, if backing isn't used, the root opening and root face equal the diameter of the electrode. Can it be wider? Yes, but it becomes more difficult to achieve complete joint penetration without burn through. Backing makes it easier to achieve CJP without burn through and it requires less skill on the welder's part.

Is it a rule that the root opening can't be smaller than the electrode? Won't the filler metal flows into the smaller gap in molten stage? For a given thickness of metal, is there a gap to thickness ratio or percentage of some kind between gap and thickness of the metal to be butt welded? Yes I got the AWS D1.1 documents but couldn't find this rule of thumb dtails. The following is the sample left of the project.

It's cut by acetylene. We plan to use bolted connections and to ensure full moment connection.. we would also weld it and make the cover plate as the backing.

This is actually a good thing, so he hasn't got the mean to dress up the welds afterwards.

Why is it a good thing? They didn't dress up the welds afterwards but just paint them with epoxy primer.

About the test. Well.. the old contractor won't even talk to me.. much less spend time on any test because we paid him full already. In our place, welders and contractors have zero liability.. we don't even know their full names. So we need to use hybrid bolted and welded connection and oversee every step of the critical repair.

 

[gtaw]

Mixing bolts and welds in the same connection isn't a good idea. The bolts will slip under load, thus the load is transferred into the welds. If each, the bolts and the welds, are not capable of transmitting the entire load, failure can occur. Once the bolts slip, the load is transferred to the weld. If the weld isn't capable of transferring the entire load, they will fail. Once the welds fail, the load is transferred back into the bolts, which if they cannot sustain the entire load, well, we don't want the worst to happen do we? Even slip critical bolts are expected to slip slightly when the faying surfaces are painted unless the paint system has been tested and demonstrates the ability to transfer the loads without slippage.

CJP groove welds without backing are not prequalified per AWS D1.1. Thus, the figures in clause 3 depicting the prequalified groove details show either welds that are backed, welded from both sides with a back gouge operation, or partial joint penetration. Any CJP groove weld made without backing or welded from both sides with a back gouge operation must be qualified by testing to demonstrate the contractor and the welders have the skills necessary to do it successfully.

The dimensions I offered are rules of thumb developed over forty years of welding experience. Call it tribal culture if you will.

Best regards - Al

 

[sedonas]

Mixing bolts and welds in the same connection isn't a good idea. The bolts will slip under load, thus the load is transferred into the welds. If each, the bolts and the welds, are not capable of transmitting the entire load, failure can occur. Once the bolts slip, the load is transferred to the weld. If the weld isn't capable of transferring the entire load, they will fail. Once the welds fail, the load is transferred back into the bolts, which if they cannot sustain the entire load, well, we don't want the worst to happen do we? Even slip critical bolts are expected to slip slightly when the faying surfaces are painted unless the paint system has been tested and demonstrates the ability to transfer the loads without slippage.

Is there by chance the cover plate bolts need to be tightened using torques wrench? In our country. Our contractors don't own any torques wrench.. they just use manual wrench to tighten bolts in base plates, and weld the top of the bolt and nut so they won't come apart. How about the bolts and nuts in the cover plates.. it's okay to weld them too so they would be fixed in place?

According to other structural engineers. It's so rare for contractor to create splice in wide flange by using bolts and nuts.. they said it's so expensive.. it's easier to just weld it they said.

So in the event even bolted options would be out of reach.. then may have to invite international welder to repair the moment critical splice.

 

[Ron]

Unless I completely missed it when I went back and re-read the threads, gtaw did not say that E6011 should not be used for structural welding. He did say, and correctly so, that E6011 should not be used for welding when a low hydrogen electrode such as E7018 should be used for A992 steel.

While E6011 has many good attributes and properties, I believe the point gtaw was making was that the electrode must be matched to both the base metal and the application. He knows his stuff! Go back and read all of his posts...then follow his recommendations.

 

[sedonas]

Mixing bolts and welds in the same connection isn't a good idea. The bolts will slip under load, thus the load is transferred into the welds. If each, the bolts and the welds, are not capable of transmitting the entire load, failure can occur. Once the bolts slip, the load is transferred to the weld. If the weld isn't capable of transferring the entire load, they will fail. Once the welds fail, the load is transferred back into the bolts, which if they cannot sustain the entire load, well, we don't want the worst to happen do we? Even slip critical bolts are expected to slip slightly when the faying surfaces are painted unless the paint system has been tested and demonstrates the ability to transfer the loads without slippage.

I'll buy a torque wrench if contractor doesn't have it. I prefer the splice bolted.. as you said.. bridge has bolted splice. In your experience.. what happens to the weld when you grind the surface flat.. because if I'll put the cover plates and bolts.. I'll have to grind the bottom of flange flat.. would this procedure weaken any weld even in normal welds?

Also in cover plates and bolts.. should there be gap in the splice? I'm asking because if we would acetylene away the old weld to put cover plates.. it would be messy.. so instead we would put the plates without removing the old welds.. the plates designed would be peer reviewed by local team to ensure it can mobilize the full moment capacity of the beam. I'm more confident with bolted connection because of lack of any skilled welders.

So many thanks for the assistance.

 

[gtaw]

The welds in your photographs are of questionable utility regardless. Grinding them flush with the surface of the flanges will not degrade the function of the welds. If anything, there may be some unexpected surprises waiting for you once the welds are ground flush.

High strength bolts are tightened to provide the clamping force needed to ensure the connections cannot slip. If the bolts are not fully tightened and if they do slip, they will go into bearing, i.e., the bolt will fetch up against the side of the drilled hole. This is an acceptable condition if there is no cyclic loading that would cause the bolts to move or slam back and forth in the bolt holes. It is important that the bolt holes be drilled rather than torch cut or drilled oversized. To function properly, all the bolts have to be in bearing, i.e., the shank of the bolt against the side of the drilled hole. Initially, one bolt will bear against the side of the drilled hole. The base metal will go into yielding as it is over loaded and the next bolt will go into bearing and so forth until all the bolts are bearing against the side of their respective bolt hole.

High strength bolts installed in unpainted connections and fully tighten to a prescribed preload depends on the clamping force and the resulting friction to transfer the load from one member to the next. That's why we used to call them friction connections. Now they call them slip critical connections meaning the clamping force should be sufficient to keep the bolts from slipping into bearing. In the US, we specify slip critical connections if there are load reversals so the bolts will not slam back and forth as the load changes. It is important to ensure the bolts are installed and tightened properly if the connections are specified as slip critical. I usually use either A325 or A490 bolts and the turn of nut method of tightening these connections. Both ASTM A325 and ASTM A490 bolts are high strength quenched and tempered steel. DO NOT WELD THESE BOLT TYPES.

Bearing connections are typically used in structures that are not subject to cyclic loads and where there is no chance there will be load reversals. The installation and tightening is not as critical as long as the bolt holes are not over sized. ASTM A307 bolts are fine for bearing connections, but their "low strength" will require more bolts when compared to A325 or A490 bolts of the same diameter.

The AISC Steel Construction Manual is a good reference for bolted connections.

This has been an interesting thread. I am glad you took the time to bring it to our attention. We sometimes forget the operating conditions of localities other than our own.

Best regards - Al

 

[sedonas]

Thanks for the tips.. what would happen if you weld the A325 bolts? This is what the contractor did after installing elsewhere. they weld the bolt to the nut and baseplate to ensure they are locked. Now I'm losing confidence in cover plates bolted connections because we don't have the machine wrench that can do the work. Hand works may not be enough.

This is the welding machine commonly used in our country (it costs about $150). Is it like it in your place or is there something missing in the following?

Now I must look for repair welder of expert nature. I will first discuss on a theoretical level to see if he pass. The old welders told me they only focus on the web in I-beam because they consider the flanges as not taking any important functions.. they don't understand about compression and tension zones.. although I may print some to show them one final time if they can comprehend it. Then I'll ask the new welder about backing bars and bevel and how big it is.. to test if he knows gtaw world.. if he passes but not using right equipment.. may have to let him bring the equipment from abroad too.

How long usually to fix per splice.. because there are 4 splices.. I wonder if he can do it in 2 hours or need 2 days.. because need to estimate my budget for him, the worse if need to get international welder because have to accommodate him to hotel etc. In neighboring hong kong, china, Singapore.. do you think there is even one competent welder or do I need to hire one as far as the US or even Russia. We don't have any welding code in our country. I wonder if Hong Kong, Singpoare have welding codes and how they differ to that of US AWS D1.1?

 

[sedonas]

Gtaw.. in addition to the above.. may I please know the following.. these will be the last questions.. thanks a zillion to you and others

1. How many times can one reweld and reweld the same portion. For example. If the new welder didn't make it right.. and we grind/acetylene the bad welded flange again and then the third welder failed again, and we regrind.. reweld.. is there a limit to the repair cycle?

2. When grinding.. can one grind just a portion of the welded filler and assume the filler (if it is of good weld) is as good as the original metal or do you have to remove all the welded filler part when doing reweld? After 5 reweld jobs by different welders team.. I wonder if the gap can become bigger or it would just stay constant as the new team simply remove the welded filler part..

Many thanks!

 

[gtaw]

Based on the photographs you provided, the existing weld should be removed entirely and replaced. The welder should not be expected to be an engineer, metallurgist, and welder. The welder may have the skills needed to deposit acceptable weld, but he will likely need direction as to what needs to be done to correct the existing connection.

The root opening will become larger with each successive repair if the entire weld is excavated as it should be. The welder can deposit weld to build up the groove face to reduce the size of the root opening. The welding should initiate on one member to build up the groove face first. Backing should be used to provide the welder with the best opportunity to make an acceptable weld. I would limit the maximum build up the thickness of the flange, 10 mm in your case.

However, looking at the welds in your photographs, it would be easier to remove the affected section and replace it with new a new section of WF.

Done properly, the weld should be as strong as the carbon steel that comprises the WF beam. The attached sketch depicts what I would consider to be a reasonable approach to correct the problems depicted in your photograph. In the long run, it will save time. It can be modified to be bolted if it is easier than finding a competent welder.

As for the welding machine in your photograph; it is fine if the welder uses E6011 because it appears the output is AC only. E6010 electrode is intended to be used with DCEP (direct current, electrode positive). If the welders attempted to use E6010 with AC output, it would explain some of the problems they experienced.

Best regards - Al

 

[Agent666]

Sedonas, in answer to your earlier question on what would happen to the bolts if welded which hasn't been responded to:-

I believe two things happen (in simple terms) when welding quenched and tempered steel
1 - Risk of hydrogen embrittlement - hydrogen atoms enter the metal, with the risk that the steel in the bolts become brittle (as opposed to being ductile)
2 - The heating and slow cooling 'undoes' the hardening that happens during the quenching process - lowering the strength of the affected steel.

As gtaw noted, never ever weld high strength bolts, or never reuse previously tensioned bolts.

If there are bolts that have been welded they should be replaced asap.

 

[sedonas]

By the way. If 10mm thick flange splice were not beveled Vee shaped at both sides or no backer used and there is a slight gap (not entirely touching) say a 3mm gap (below size of electrode). Wont the weld filler be able to pass thru the 4mm gap? The contractor is saying there is a slight gap where the 6011 molten filler were able to move thru it. What can you say about this minimal gap dynamics?

 

[gtaw]

I think we've pretty well covered that a couple of days ago.

Best regards - Al

 

[sedonas]

I think we've pretty well covered that a couple of days ago.

When you mentioned this a couple of days ago "Generally, if backing isn't used, the root opening and root face equal the diameter of the electrode. Can it be wider? Yes, but it becomes more difficult to achieve complete joint penetration without burn through. Backing makes it easier to achieve CJP without burn through and it requires less skill on the welder's part."

I'm talking about not wider, but narrower than the diameter of the electrode. Perhaps your experience says that root opening and root face equal the diameter of the electrode in order for the arc produce by the welding rod to get into the gap.. but if it's narrower, maybe some arc can also get inside the gap.. perhaps enough to put the middle in molten stage?

In welding, the welding rod is what produce the molten arc, if the electrode is bigger, the arc get bigger and it is dependent on the size of the electrode? or is it not dependent on the size? For example.. if you don't use filler electrode but just use wires in the holder.. would it also have arc, and how big would this arc be? This is to understand the relationship between the arc and the electrode size in scientific details.. not just engineering. It's not in the AWS D1.1 which is an engineering manual. Thank you!

 

[kingnero]

What did the practical test say when presenting this scenario?

 

[gtaw]

OK, I see what you are driving toward. If the root opening is insufficient, the weld may not penetrate to the rot. The weld will increase the welding current to achieve more penetration, but at the expense of the probability of burn through. The weld and the arc are harder to control.



Best regards - Al

 

[sedonas]

In splicing the wide flange.. the welder uses acetylene to cut it. Since its not staight cut and the edges are rough and uneven and they dont own any grinder. Dats how some 2mm gaps were produced with some in contact some not.. you said 2 messages prior about replacing whole sections of it. Wont it be harder since now one has to make 2 splices?. and they would exceed the L/4 locations we were told to splice. So better options would be opening the gap right in the weled portions and either use backing or Vee shaped bevel on both sides.

By the way when you mentioned more possibility of burn through with larger gap without backing. Were you referring to the gap not being able to fill up? Because burn through is supposed to happen when its thin sheet and you burn through it.. does this term also refer to unfilled sections? Is the term burn through accurate here..

 

[gtaw]

Burn through is whenever the welder melts through the base metal and leaves an unfilled hole.

My suggestion that it would be easier to remove the entire compromised section of beam is based on what I saw in your photographs, specifically October 28 at 1:58. Attempting to fixing what is depicted by that photograph is nothing short of opening a nasty can of worms. It would take a magician to fix that, not just a welder.

The cost of a grinder to properly prepare the joint and to allow the welder to remove questionable weld is a lot less expensive that trying to repair the connection we've been discussing for the last several days. There are a few basic tools needed to do any job properly. A mason needs a trowel, a carpenter needs a hammer, and a welder needs a grinder. The mason can substitute a stick for the trowel, but the job is going to suffer. The carpenter can substitute a rock for the hammer and the welder can substitute a hand file for the grinder. At some point one must think about the cost benefit of having the proper tools to do the job.



Best regards - Al