Welded Transfer Steel Truss Splices 2

[abusementpark]

I have some ~65' long x 6' deep heavy transfer trusses using wide-flange members. The trusses support building columns which have the load of multiple building stories. The truss diaogonal and vertical member connections are all full-pen moment connections (for stiffness and achieving the most clear space between members for MEP routing). I was hoping they could be shipped and installed in one piece, but the fabricator is saying they can't and they will need to break into three pieces.

I plan on making them do full-penetration welded field splices with ultrasonic testing. Is there anything else that can/should be done for quality insurance and quality control in the field?

 

[BridgeSmith]

It seems strange that they can't ship a 65' x 6' truss. We have 120'+ long by 8' deep bridge girder sections shipped without a problem.

Anyway, we'd never specify a full-pen welded connection to be completed in the field. If WF sections requires field assembly, it's done with a bolted splice with fully tensioned high strength bolts.

 

[RontheRedneck]

BridgeSmith, I once worked at a wood truss plant that wouldn't build anything over 12' tall. They were in a tight spot, and shipping stuff out of the area is what caused the restriction. It was not limitations in manufacturing equipment or trailers.

Abusementpark, I gotta believe there are other fabricators who can build the things you need in one piece. If checking around is an option, I would recommend it.

 

[HTURKAK]

I do not have any idea why the fabricator is saying this. If this is the only option, i would prefer high strength bolted splice .( the splice plates welded to one side , bolted with the other side.
Acc. to just my past experience.



He is like a man building a house, who dug deep and laid the foundation on the rock. And when the flood arose, the stream beat vehemently against that house, and could not shake it, for it was founded on the rock..

Luke 6:48

 

[KootK]

I plan on making them do full-penetration welded field splices with ultrasonic testing. Is there anything else that can/should be done for quality insurance and quality control in the field?

I also like a bolted splice but, until you say otherwise, I'll assume that you have your reasons.

Beyond testing the weld quality of the joint, I think that it would be prudent to take some measures to protect your own interests a bit more as follows:

a) Define the allowable, lateral straightness of the truss after it has been field spliced if you care. That can affect bracing requirements and aesthetics and you don't want some, zillion pound compression chords meeting up with a 10 lateral degree kink or anything.

b) Clearly define the contractual obligations should the field welding result in failures requiring rework since that is something that may well come to pass. That the contractor should pay for the rework seems obvious enough but, more than pointing fingers, I'd like to get the contractor thinking about this ahead of time so they can manage their risk ahead of time. I'd also follow this up with an informal conversation with the contractor.

Most of us have "been there" with something like this where the field welding fails QC but the costs to fix that are so ridonculous that they result in great pressure on the engineer to let it ride.

 

[BridgeSmith]

...i would prefer high strength bolted splice .( the splice plates welded to one side , bolted with the other side.
Acc. to just my past experience.

Interesting. I guess it would keep the splice plates where they're supposed to be. Our bolted field splices for bridge girders are always bolted on both sides. Of course, our splices have inside and outside plates for the flanges and web splices on both sides. I would foresee major difficultly in getting the splice together, trying to slip the bolted-side flanges and web between the splice plates. Or did yours have only outside flange plates?

 

[HTURKAK]

I agree with you that bolted splices bolted on both sides for bridge girders is the common practice ( plate girders ,hot rolled ..).In this case , the discussion is for truss and the OP did not provide a lot of details here .I guessed that , the top and bottom chords composed of WF's with vertical orientation and considered outside flange plates is OK. The OP may clarify the details to get better responds.



He is like a man building a house, who dug deep and laid the foundation on the rock. And when the flood arose, the stream beat vehemently against that house, and could not shake it, for it was founded on the rock..

Luke 6:48

 

[DrZoidberWoop]

It's pretty common to break the trusses into thirds (or more segments) and assemble on site. Normally, 2 segments can be shop assemblies and the remaining segment is "knock down" or site assembled. It may not be the case for you, considering the short height of your truss. The erector and fabricator will have discussed their preferred delivery and assembly strategy, and that is what they're presenting to you.

As far as QA/QC, you need to ensure the weld joints are inspected for compliance w/ AWS D1.1 4.17.5 (2020) (if applicable), which requires weld runoff tabs and explicitly forbids the use of end damns at the splice. Also, AISC 360 Table N5.4-1 "Inspection Tasks Prior to Welding", which covers the alignment and preparation details of the joint in question.

As a side note, a 65'x6' truss is a cost-savings nightmare, due to the welding and joint preparation. One of the first things we (steel fabricator in eastern US) would do is the write an RFI proposing to replace the truss w/ a plate girder w/ web penetrations for MEP. It saves everyone a TON of money because material is the cheapest part of the value engineering, labor and arc time is the most expensive.

 

[BridgeSmith]

As a side note, a 65'x6' truss is a cost-savings nightmare, due to the welding and joint preparation. One of the first things we (steel fabricator in eastern US) would do is the write an RFI proposing to replace the truss w/ a plate girder w/ web penetrations for MEP. It saves everyone a TON of money because material is the cheapest part of the value engineering, labor and arc time is the most expensive.

I agree. If you have the option, a plate girder would be far more economical. The only trusses we do are purely for aesthetic reasons, typically to replicate a historic truss we're replacing. We don't even include transverse web stiffeners anymore, except for very rare circumstances.

 

[abusementpark]

Abusementpark, I gotta believe there are other fabricators who can build the things you need in one piece. If checking around is an option, I would recommend it.

Not sure what industry you are in, but in the commercial world, the GC picks the fabricator, not the engineer. I could dig in my heels and say no splice just because I'd said so, but they are offering to do the full-pen welded field splices that should work.

 

[abusementpark]

a) Define the allowable, lateral straightness of the truss after it has been field spliced if you care. That can affect bracing requirements and aesthetics and you don't want some, zillion pound compression chords meeting up with a 10 lateral degree kink or anything.

I agree that this is a concern. I am going to require that bolted alignment plates be provided at splices that can also be used as backing for the welds. However, I am not sure how to define and measure this in the field. How would you approach this? Obviously, they will need to meet all AISC tolerance requirements.

 

[KootK]

How would you approach this?

I think that I'd chase down the AISC tolerance limits that would normally apply to a truss and then - baring a good reason to do otherwise -- use those. Or, if you feel that relaxations can be accommodated, work those out up front to help make everyone's lives a little easier.

Obviously, they will need to meet all AISC tolerance requirements.

I feel that the trouble there is the "obviously". Obvious to you and I? You bet. My real worry is that the truss won't meet the AISC tolerance requirements and then there will be great pressure on you to relax those requirements, which you can do as the EOR. The magic EOR wand is often our friend but, then, sometimes our enemy in this regard.

I don't know that this needs to be any fancier than making it abundantly clear that there will be no relaxation of whatever tolerance requirements you specify. Basically, if somebody shits the bed on this, I'd not want that fecal matter to wind up in my bed. Plenty of fecal matter has found its way to my bed in the past and I've not enjoyed that experience.

These things are awkward because, regardless of the contractual relationships involved, it is almost always the case that our contractors are, to some degree, our clients. Our businesses depend on our reputations for being easy to work with and producing cost effective solutions. In large measure, that perception comes from how happy our contractors are with our work. It's a system rife with potential for abuse of course.

I know of a major firm that puts asterisks on their drawings for things that are "super important". I get the impetus for that but, then, it kind of gives that the impression that QC for the rest of the job can be allowed to slide. And perhaps it can.

 

[CANPRO]

Thoughts/questions from the perspective of a fabricator:
[ul]
[li]As noted above, 65' isn't that long. If breaking it into two pieces still isn't sufficiently short then surely there are more delivery problems with this project. Is this a length issue or a weight issue? Must be a weight issue.[/li]
[li]How heavy is this truss? Should be easily shippable up to 65,000 lbs. Does this weight more than 1000 lbs/ft?[/li]
[li]Assuming it is a weight issue and not a length issue, the cross sections involved must be significant. And so must be the CJP welds to join the sections.[/li]
[li]This will not be cheap so the fabricator/erector must have a good reason to do it. Are they looking to charge extra for this work?[/li]
[li]Assuming we're talking very heavy sections, the CJP welds will be many passes. They won't be flipping this huge truss to maintain ideal weld access so welds will be happening in all positions.[/li]
[li]A web member will need to be spliced in as well. What does that weld detail look like and will the fit-up of that web member be any different than if it was built in the shop? You might end up with a lot of restraint in that web member with it being welded into a now rigid structure both ends. Quick math on this: [/li]

[li]if the truss 6' deep and webs are at 45°, the web length is about 102" node to node.[/li]
[li]If the weld shrinking is 1/16" (this isn't a ridiculous number if you're bridging a large gap due to design/fit-up) then you have a total strain of 0.000614 and a resulting stress of about 18 ksi. That is assuming the ends of the webs are perfectly restrained[/li]
[li]It would be almost impossible to put an exact number on the resulting stresses, but it is helpful to look at the ball park of potential resulting stresses. They can be significant in high restraint joints and should be mitigated through proper sequencing where possible.[/li]
​

​

[li]It would be reasonable to ask the following:[/li]

[li]What weld process do you intend to use?[/li]
[li]Can you produce approved weld procedures for the work? Do the procedures cover all positions required to complete the weld?[/li]
[li]Can you produce a connection detail, including any rat holes (access holes) that maybe required for continuity of backing bars.[/li]
[li]How do you intend to maintain alignment and fabrication tolerances during this splicing process?[/li]
[li]Based on your experience with welding structures like this, do you consider this is a high restrain joint?[/li]

​

[li]It would be reasonable to require the following:[/li]

[li]3rd party inspectors present for the entire welding operation. To KootK's point about poor field weld quality; if you have thick sections and inclusions or defects found within the weld, the excavation and repair of those welds will be significant. If there are issues with weld quality its in everyone's best interest to catch it early.[/li]
[li]A written procedure of the splicing process including means of securing the work and maintaining alignment, documentation on weld procedures and welder qualifications, splice details (including specific weld details) stamped by an engineer.[/li]
​

[/ul]
Are you able to share any details of the truss?