Mixing ASD and LRFD

[SteelPE]

I am currently looking at a project to help out one of my customers who is a fabricator.

They have a project that they say is "simple" but things in this business are never simple. The building they need some assistance with is a new building designed by what I would consider a world class engineering company. My client needs me to design some of the connections for the building. In going through the plans to define my scope of work I noticed that they have specific notes on the drawing that say shear reactions are given at the working stress level and that moment reactions are given at a ultimate strength level. It is up to the connection design engineer to run the equations through some formulas in order to combine them together.

Am I behind the times in how buildings are designed nowadays? What would be the purpose to list reactions on a project it two different forms? This seems like they are purposely introducing confusion into the equations.

 

[JAE]

They are suddenly not "world class" anymore.

Seems very strange to me. I can think of no reason other than perhaps the moments are the basic seismic moments (1.0E) and everything else is gravity done with ASD.

 

[Ron]

Agree with JAE...odd to mix them up.

In delegated engineering design that I've done, I've generally followed the method used by the EOR for the structure...and sometimes it is dictated that supplementary designs be done in a particular manner.

 

[SteelPE]

In this particular instance I need to design some moment connections that have what I would consider high moments. I need am going to need web doublers or increase the column size. In order to properly check the columns I need to figure the axial load on the column. I can sum the vertical reactions but now I have to run them through his equations to get values that are similar. On top of that, he wants the moment shear reactions to support 2Mu/L+R. Again, I have to run them through his equation. Not difficult.... just another step that could add some error.

I don't even know if I can meet the fabricators schedule on this project. They always try to cram these things down my throat.

 

[nutte]

I see this often, indeed from world class firms on world class projects. I can't say why it occurs, other than to speculate that you have so many people working on the job, and a lack of quality oversight bringing all the groups together, that things like this slip through. It could be a simple mistake in which note was put on the drawing, or it could be a larger bad practice of mixing the two methodologies.

Either way, you're asking for trouble if you try to figure out what they want and move forward. If you guess wrong, you'll wind up redoing the work, delaying the schedule, and incurring more cost that might not be recoverable. If it gets to that point, steel could already be fabricated, and nobody will be happy.

I suggest you sumit an RFI to the engineer of record clarifying exactly what the requirements are. It should be either ASD or LRFD, not a mixture of the two.

 

[BAretired]

If shear reactions are given at working stress level, how can you convert to factored loads unless you know the breakdown of dead and live load?

BA

 

[SteelPE]

BA they state in their design documents that the shear reactions are working stress and that moments are ultimate strength. They go on to say that to convert Ultimate to Working multiply loads by 0.8 or to convert Working to Ultimate multiply loads by 1.5. All the loads are probably way overblown anyway so precision is probably not really necessary.

It's not a large project. Just two stories. I can't see multiple people working on this project at the same time. I am very disappointed in the engineering company with the way this is being presented. Out of school, I wasn't up to "snuff" when I interviewed with the company. Oh well.

 

[JAE]

If there is significant seismic effects, you'd also need to know if the moments contain the required overstrength factors, etc.?

 

[slickdeals]

Please tell me you are not reviewing TT drawings.

 

[SteelPE]

slick,

No, no where near as large as TT..... but people would know them if I gave their name so we will leave them out of it.

 

[slickdeals]

No worries, the reason I asked is because they had a reputation for delegating their connection design work, lately it appears they are trending away from it.

Good luck figuring it out. Hopefully the EOR gives you some good direction.

 

[JAE]

TT?

 

[dhengr]

Yah... That’s TT, LLC..., Tits and Tush, Lot’s and Lot’sa, Convolution. They are well known for not wanting you to see what’s actually there. So, they mix it up just to keep things interesting. On the surface, they appear to be well detailed, but when you dig a little deeper (peal back the first few layers) you do need an RFI (redirection of factual inconsistencies). Usually there is much more there than you would really like to see, and it’s all in the wrong places. Their upper cantilevers are usually augmented and their lower levels are over weighted to the rear so their C.G. is too far back. But, at first glance this is all deceptively concealed in all the foundational underpinnings, adjustment factors and resistance factors which are foisted on us these days.

 

[connectegr]

I use to see this all the time. 10 years ago the explanation was that fabricator throw money at factored loads and think ASD service loads require less expensive connections (smaller forces). So they would design the building using LRFD then convert the forces to ASD on the contract drawings. The confusion would be when they provided general information for shear reactions, tables or %UDL, and left the bracing elevations or moment connections as factored loads. I don't see combination of ASD and LRFD forces much any more.

With a conversion factor it does not require any additional connection design time.

Now with moment frames %gravity and %lateral moment information may eliminate some column web doublers.

http://www.FerrellEngineering.com

 

[dozer]

This post just proves to me once again of the folly of having two design methodologies.