Reviewing steel connection calculation package 3

[Structgeek]

I would like to get some advice on reviewing steel connection calculation package. Its always a huge document of about 300-500 pages, so I would like to know what to look for in those document to make the review process efficient.

 

[KootK]

In theory, I believe that we're legally obligated to check every damn thing such that we could pretty much stamp the calcs ourselves if we had to. In practice, nobody has time for that.

I pretty much only check things that allow me to verify that the delegated designer properly understands my intent for their designs. A non-exhaustive list of such things would include:

1) Their loads match my loads including more nuanced situations such as pass through loads at braced frame connections etc.

2) Their assumptions regarding joint boundary conditions match my assumptions regarding joint boundary conditions. Restrained, pinned, accommodating movement, etc.

3) Their connection modelling choices appear valid to me given my understanding of how steel connections ought to be designed with respect to stiffness, load path etc.

4) I'm seeing some, meaningful level attention paid to the failure modes that I expect to be checked in a given connection (Whitmore spread, plate buckling, block shear...).

5) I answer their questions clearly if any have been asked.

I feel that this strategy gives pretty good value for money with respect to my time and my ensuring a responsibly designed structure.

I've submitted a lot of precast concrete calc packages as the delegated engineer. In my experience, what I've outlined above will be a fair bit more than most EOR's will do. The biggest issue that I've had with calc review over the years is that it usually gets delegated to someone at the EOR's office who is fairly junior. That often results in:

1) Lots of foolish comments that just waste everyone's time and;

2) The absence of more sophisticated comments that probably would benefit the project in meaningful ways.

 

[DrZoidberWoop]

Unfortunately, there are not many shortcuts. If you want to immediately get a good reputation, do everything in your power to avoid checking the R&R (Revise and Resubmit) box on your return approval. R&R will push the schedule by delaying the detailer's ability to produce shop drawings for approval, which will send the project managers into conniptions. Try your best to check the "Make Corrections Noted" box or something similar. That way, the person producing the sketches and calcs can make the minor corrections you noted and shoot it quickly over to their detailers. You can add a note to your list to revisit a particular issue at the shop drawing review phase.

 

[Aesur]

Lately even when I use the "reviewed and noted" checkbox on our shop drawing stamp I tend to get a resubmittal, which I don't want to, nor have the budget to check again. As KootK said, there just isn't time nor fee to review these submittals like the contractor would want. I'm not sure if we are legally required to review everything as it is my understanding reviewing shop drawings (where the EOR designs the connections) is a courtesy to help the contractor by trying to catch the items you deem more important (ie drags, special connections etc) but the responsibility of the shop drawings is on the steel fab and contractor as you provided a complete structural set that technically could be built off of without shop drawings, which are used more for shop fabrication.

I'm starting to add verbiage that we include one review and will bill for additional reviews because these 800+ page submittals can take days sometimes and CA fees are way too little.

A quick overview of how I approach these are as follows:
Deferred Submittals (typically trusses, i-joists etc): Review loading, check for general conformance to the layout, check special conditions, look at utilizations and for any special notes/questions they have.

Shop Drawings (items you designed, but they made fabrication drawings): Check special conditions and check member sizes, but not the cut sheets for each piece and not every connection, spot check those only.

Delegated designs: I don't delegate many designs, but I would assume I would use sim to deferred submittals.

One thing I would note however is, if you have a younger engineer who you want to give some training to, have them go through each page of a large package, I believe shop drawings are a good tool for training, sim to drafting.

 

[271828]

KootK has a great list. Do something about like that.

I submit these packages. I'd recommend focusing on potential sources of confusion.

Example: Some EORs will require that some connections are SC. It's often difficult to figure out which connections this applies to. Look at the submitted calcs to make sure they have the right ones SC.

Example: Braced frame connection. The BF elevation will typically only show the brace axial force. On the plan, an adjacent beam might have an axial load that needs to transfer into that connection. Also, either on the elevation or plan, the braced frame beam might have an axial load. If you start drawing FBDs of the joint, typically those forces don't go together, so the connection designer must do some guessing and try to figure out the worst case or cases. The beam-to-column connection is usually the critical element here -- make sure that is being designed for the force you want.

Example: If there is any mixing of material properties, make sure they have the right ones in the calcs. For example, I've seen a couple of jobs lately that required some of the HSS to be A1085. On one of them, it took an R&R and a couple of emails to get us on the same page; it wasn't as simple as the note on the structural drawings would indicate. LOL

One more: Sometimes the connection engineer is a bit disconnected from the shop drawing process. The drawings typically get submitted without notifying the CE that's about to happen, so there might not be any review. Make sure critical elements on the drawings match what's in the calcs, or are more conservative.

 

[jerseyshore]

A lot of great points above and I agree with them all.

I make sure to always check moment connections specifically. There has been a big trend recently to add the shear and moment values to steel projects and have those connections designed by others. So I always go to the MC's and make sure they work with the forces I put on the plans. Always make a note for "no paint at MC", stuff like that.

If I have a weird connection, say a larger W shape framing into a smaller beam, check that one out in the package. Stuff like that.

 

[human909]

My work is 95% steel so this is something I do all the time. Though the work flow I deal differs from others here so my comments might now be relevant.

I generally do most of my checks of steel fabrication detailing at the 3D model level. I would normally only get into the details when there are abnormal and critical welds needing to be noted on drawings.

Most of the time I'm reviewing key connections which are generally moment connections, strut connections and perusing over aspects of a few niche areas to make sure the detailer hasn't done anything bizarre.

I do member and connection design so I'm not checking others calcs.

 

[structSU10]

This article discusses reviewed delegated design submittals a bit:

https://www.csiresources.org/blogs/...drawings-and-submittalsdelegated-design-submi

It talks about limited review and described what the AIA and EJCDC outline. Not super specific on what is reasonable but its some good things to keep in mind.

 

[Brad805]

I find this discussion very interesting. In Canada there are no calc submissions, and I am not sure some of the connection designers I run into on small projects do much beyond a cursory review. I tried doing connection design for a supplier, but they do not include fees to figure out all the reactions correctly or check any dimensions. I moved on.

 

[KootK]

I tried doing connection design for a supplier, but they do not include fees to figure out all the reactions correctly or check any dimensions. I moved on.

#MeToo. Had to design 10,000 connections at a $0.02 a piece for a stadium and then was expected to sign off on the installation sight unseen after the fireproofing had been installed. I think that you can make a go of it if you're built for that kind of work (spreadsheets, standards, offshore help). Stepping into it as an EOR generalist is pretty tough though.

I have had some success smaller projects where I don't do the drawings but, rather, just crunch some numbers and prepare some detail sketches for others to turn in to drawings. I've quite enjoyed that actually when the relationships are fun.

 

[KootK]

In theory, I believe that we're legally obligated to check every damn thing such that we could pretty much stamp the calcs ourselves if we had to.

Anybody wanna disagree with me on that? I've always thought it bizarre that it seems so difficult to split stream responsibility as one would wish in these situations. I get that there's inevitable overlap and that lawyers take a shotgun approach to lawsuits but still... can one really not split things up contractually in a way that offers the desired protection?

When an EOR delegates design, I feel as though they should be able to delegate all of the associated responsibility save one aspect: whether or not they defined the design criteria adequately for the sub (usually they don't).

 

[structSU10]

In that article I posted it talks about that - trying to separate the liability and provide limited reviews of those documents so the delegated designer maintains responsibility. I haven't gone as far as the article describes myself yet though.

I have seen court cases though that still came down on the EOR as they should have known or commented the delegated design may have been improper and done a more thorough review. I think that was in a webinar I saw so not sure I can track down the source though...

 

[KootK]

@structSU10: thanks for that and for the linked information that you shared above. I've not had time to review it but plan to in the near future.

 

[KootK]

...trying to...provide limited reviews of those documents so the delegated designer maintains responsibility.

Although that just seems foolish and ridiculous to me.

Good contracts and good delegated engineering mostly come down to having the right people do the right things based on the specialized knowledge possessed by the various parties. EOR's being deliberately less helpful surely is not the path to improved public safety.

 

[human909]

As somebody who works in a locality with the definitive EOR approach I find this discussion interesting.

In some ways a single EOR makes crystal clear who is responsible, which in some cases is good. When multiple parties are interacting sometimes nobody takes ultimate responsibility.
True responsibility slips through the cracks and engineering mistakes or disasters occurr.

On the flipside the reality is that most projects a single EOR cannot realistically check everything. So they are relying on others to do their job properly. So why shouldn't those delegated to do their specialty job be responsible for their work?

(I'm just posing the question for the sake of discussion rather than taking a clear stance.)

Although that just seems foolish and ridiculous to me.

Good contracts and good delegated engineering mostly come down to having the right people do the right things based on the specialized knowledge possessed by the various parties. EOR's being deliberately less helpful surely is not the path to improved public safety.

Agreed. Though in my world if I get external consultants to perform engineering work they don't expect me to review their engineering designs. They might expect me to review the suitability to the project but actual review of the calculations is not expected.

That said good consultants will sit down at the table together if necessary and hammer out solutions. Ultimate responsibility rests on the head contractor and cascades down be that to the primary structural engineers or other engineering consultants.

 

[dauwerda]

Quote (KootK)
In theory, I believe that we're legally obligated to check every damn thing such that we could pretty much stamp the calcs ourselves if we had to.

Anybody wanna disagree with me on that?

As an EOR, when you spec a proprietary item (anywhere from a Simpson hanger to Castconnex HSS Seismic connections), I think it is fairly clear that the performance/reliability of that item is the responsibility of the providing company. It is the EOR's responsibility to make sure they specified the correct parameters to select the correct item.

To me, delegated connection design (and other delegated designs) would fall into a similar category. It is the EOR's responsibility to provide the correct information to the designer, but the designer would be responsible for the connections (meeting the requirements given, If the EOR gave bad inputs/requirements, or even unclear ones that would be on the EOR).
The EOR should be doing a cursory review making sure the correct inputs were used and looking for anything that might have been interpreted incorrectly. In my experience this rarely happens (especially with large firms), the new EIT gets to review it and provides all sorts of comments and questions that aren't actually pertinent to the design, so now the delegated designer is getting tasked with training the EIT rather than the firm that should be doing it, and any real issues that might be there aren't being found. (If you sense some bitterness in that last statement, that's because there is some).

As someone who has been an EOR on many industrial projects and now provides a niche delegated design service for a fabricator, I know that I certainly don't expect the EOR to be an expert in the specialty design. To that point, I try to create a very well laid out calc package showing all of the major steps and inputs/outputs, which are then followed by all the nitty gritty details. When I get an actual EOR reviewing it I usually get comments about how easy it was to follow, along with an approval and clear review of the major steps/inputs (yellow highlighter). When I get junior EIT's reviewing it, I usually get a bunch of comments/questions in the nitty gritty detail sections and a revise and resubmit. Experienced engineers tend to know they aren't and cant be an expert in everything, but they do know what to look for when reviewing things they are specifying. Inexperienced engineers simply don't understand any of that and when they start commenting on minute details, they really start to cloud responsibility lines.

Clearly experts in specific areas exist for a reason, and it's not so that they can pass blame onto an EOR that is not an expert in that area.

 

[DrZoidberWoop]

When I get junior EIT's reviewing it, I usually get a bunch of comments/questions in the nitty gritty detail sections and a revise and resubmit. Experienced engineers tend to know they aren't and cant be an expert in everything, but they do know what to look for when reviewing things they are specifying. Inexperienced engineers simply don't understand any of that and when they start commenting on minute details, they really start to cloud responsibility lines.

This has been exactly my experience as well. Whenever I get a really whacky or silly comment, I try to find the person's LinkedIN. More often than not, it's a fresh EIT. Although I have had the opposite, w/ senior level guys saying, "well, I've done it this way for 25 years!" And if anyone here works for LERA or WMATA, you should not be allowed to comment. JK. Sort of.

 

[phamENG]

I think they have the misguided notion that, if there's an issue, that protects them. "Look, judge, I told them I didn't do my job. It's right there. Surely I can't be sued."