PEMB Modifications 1

[hemiv]

Hi all.

I've been asked to provide a design allowing a PEMB owner to remove X bracing for new doors. I've done this before with chevron bracing and new beam, but this time the openings are going to be quite large and I don't have enough room for that.

So I'm going with a new portal frame. They bay in question is 27'7" wide and the eave height is 15'. Overall 80'x80' building. The bracing is visible on the wall I'm reworking, but the other three walls of the building have an interior plywood wall covering and I cannot see the LFRS elements. There are no drawings available for this building. And of course no one wants me to get into to analyzing this whole building, which would cause the business to shut down in order to move a lot of equipment and remove all the interior wall coverings.

So my plan is to proceed with the design with the 10% rule for lateral loads - I'm not designing any additions, so new lateral loads will be induced in the system - while also designing the portal frame to be quite stiff and reduce the amount of load the other elements of the LFRS will take as redistributions.

Just wanted to get some extra thoughts on this approach. I know the existing elements of the LFRS which are perpendicular to the new portal frame will take load if the portal frame is too soft, so I am primarily trying to eliminate this from happening as much as possible.

 

[Aesur]

Based on the measurements of the building, I anticipate you could figure out the lateral loading along that line fairly easily.

 

[phamENG]

"Excuse me, please design this structural modification for me. Oh, you need to be blind folded and do it with your right hand tied behind your back (or left, if you're left handed)."

I'd pass. Not worth the hassle. The only exception would be if I thought I could get a moment frame in there stiff enough to approximate the lateral deflection allowed by the notoriously sloppy and loose cables.

 

[JedClampett]

Options in order of likelihood of success:
[ol 1]
[li]Ask the owner to contact the PEMB to have them design this. I know, you don't know who it is, and if you did they wouldn't do it. But next time the owner says they want the cheapest building, this might plant a seed.[/li]
[li]Have them pull off all the plywood, which no one wants you to do. Then measure a lot of things and analyze based on a lot of remaining guess work. Make sure it's in your agreement, if you say you see something else wrong, it will have to be fixed. Make sure your fee is about 3 times what you usually charge, and you'll need it.[/li]
[li]Take the phamENG path and turn it down. You'll sleep better.[/li]
[/ol]
We had the exact same thing happen to us. We needed to cut out a brace in a PEMB.
[ol 1]
[li]We called the original manufacturer and asked if they would help us out, for a fee. They wouldn't. This despite the note on their drawings saying the structure couldn't be modified without their permission. I know, they can make more profit doing almost anything else, including browsing Eng-Tips.[/li]
[li]We put in the drawings that the contractor needed to hire an engineer to design a fix. They said no one would do it.[/li]
[li]We checked, and they were right. Absolutely no one wants to touch these.[/li]
[li]We told our management that if we touch the building, we would own it. To keep the project moving forward, we ended up doing it in house, which we really didn't want to do.[/li]
[/ol]

 

[WesternJeb]

PEMBs usually use diaphragm bracing to get the forces where they want them. How do you plan to get the diaphragm forces into your portal frame without a full analysis like this or full renovation of the roofing system?

The odds of getting a portal frame to be anywhere near as stiff as the braced bay 50' or so away from it seems pretty difficult.. Or once you show them the framing sizes, they will probably laugh at it and rethink their idea. Either way, I am more concerned about how you plan to get the diaphragm strong enough to transmit the forces you need.

 

[SWComposites]

Is there already a similar size door in the building? If so, can you copy the framing around that door over to the new door location?

 

[phamENG]

SWComposites - Metal Building Systems rarely work like that. They put diagonal cables in bays with no openings. These braces are typically mirrored and appear in the same bay on the opposite side, and also in the bay in the roof - they don't use a diaphragm like "normal" steel buildings do. So it's just another way they are incredibly efficient...and incredibly difficult and costly to modify from their base form.

 

[KootK]

I know the existing elements of the LFRS which are perpendicular to the new portal frame will take load if the portal frame is too soft, so I am primarily trying to eliminate this from happening as much as possible.

I feel that this would be a pretty straight forward exercise. I don't feel that the portal frame needs to be of comparable stiffness to anything else.

I take your point regarding the perpendicular LRFS however:

1) That effect should only occur if the roof diaphragm is rigid. And that seems unlikely given that PEMB diaphragms are usually discretely braced as Jeb mentioned.

2) It would take a lot for those walls to see more load from torsion that they would see from the load case that loads them in direct shear.

At first blush, you approach strikes me as being pretty reasonable.

I'm assuming that the braced frame that you plan to remove is the only LFRS on that wall.

 

[HTURKAK]

Without considering the legal responsibility , technically it is possible to remove X bracing for new doors. In past , i have got a similar problem but for industrial building . I have decided to install portal truss frame bracing . I just provided similar lateral stiffness rather than checking the overall building.

The following picture from METAL BUILDING SYSTEMS ( By Alexander NEWMAN )

Use it up, wear it out;
Make it do, or do without.

NEW ENGLAND MAXIM

 

[Eng16080]

Adding to the comments by JedClampett, if you decide to do the project, I would additionally ask the client for a large portion of the fee up-front. I've had several projects similar to this where the client/owner was requiring a modification to an existing structure and ultimately did not like the solution I came up with. At that point, they hired another engineer in the hopes of arriving at a cheaper solution. In several of those cases, we were not paid for our work.

I can even think of a few projects where the client was just shopping engineers to find the cheapest solution with a stamp on it, usually coinciding with the most negligent engineer of the bunch (not that I'm bitter about that, or anything).

 

[hemiv]

This ship's already sailed, no chance of turning down the project now. I don't sell them, I just design them. I'm actually not too worried about it. As far as the legality goes, I have no idea - no one knows who made this building. A local contractor built it in the 90s. It's a very rural area with no code enforcement, so nobody has any drawings on anything.

@westernjeb - The roof bracing looks fine and is already doing it's job. I'm not asking it to change where or how it's already delivering load.

@kootk - Good point rgd the flexible diaphragm. Yes, you are right about the perp LFRS not taking much new load so there's probably no reason to increase the stiffness of my portal frame to mitigate that. I guess I'm just antsy about not increasing the unity by more than 10% under this particular load case. i.e. going from 0% to not more than 10%. And I think it will help provide a similar level of performance. Yes, it is the only LFRS on that wall. I'd be replacing one X brace with one portal frame.

 

[bones206]

We told our management that if we touch the building, we would own it. To keep the project moving forward, we ended up doing it in house, which we really didn't want to do.

I unfortunately get roped into a lot of these PEMB retrofit projects. I've learned to lean into the fact that I'm going to "own" a bunch of these frankensteins over the course of my career. As long as you understand the building system and have have confidence in your design you'll sleep fine at night. Someone has to be willing to take these on and git'r done.

Echoing KootK's comment, I wouldn't worry about the portal frame stiffness relative to the orthogonal bracing elements. I would come up with a target drift based on typical PEMB practice to proportion the portal (~H/60 for 0.42W is typical for metal-panel-clad frames). As a back check, portal frame stiffness should somewhat match the existing X-bracing stiffness. You can also consider base plate rotational springs to really dial in the drift calc.

 

[phamENG]

Someone has to be willing to take these on and git'r done.

Or we could all just boycott them. Then maybe owners would start to understand there's no such thing as a free lunch. You want to skimp and buy one of these things? That's great. They are excellent if you want a barn or a warehouse that will never change. But if you want anything else or there's a chance you want to modify it in the future? Well...your savings is going to cost you dearly.

 

[bones206]

I used to boycott them but they are a steady stream of work, and since I'm not my own boss I can only say no so many times. You just have to be very explicit in the proposal about cost and what needs to be done to make it work. I don't typically care if they end up paying or not -- not my department.

 

[hemiv]

@bones - H/60? Wow, I didn't realize they went so low. Do they typically use a maximum limit? H/60 would take me to 3" of allowable deflection. I also don't think I have a good way to approximate the existing X bracing stiff. I bet the deflection required to get the sag out of the rods would be more than the rods themselves would allow. I modeled it a rod with no pre-load, but I think that's giving a pretty high estimate of the stiffness.

 

[phamENG]

Yep - H/60 is standard. How do you think they get their prices so low? Like I said - great for a warehouse or barn. Just about anything else...they're not as cost competitive if properly specified and designed. But they slip H/60 onto every proposal they can and see if it sticks. Since most people only hire an "EOR" after they already ordered the building....(there are reputable companies that don't do this, but they seem to be the exception)

Do you have rods or cables? Either way, it's typically as easy as PL/AE and figure out the distortion of the frame from that to find your sway. The 'compression' buckles early enough to be ignored, so you only have to deal with the tension side.

 

[hemiv]

@pham - rods. Yes, I can come up with that and have already modeled the frames. But that's really only applicable to situation where the rods are tightened. The rods are typically left loose, so I assume that has a big effect on the amount of deflection that would occur at service levels.

 

[phamENG]

They're not supposed be loose. When those go up, they do tighten them up. But...these things relax. They don't get maintained. So the building gets sloppier. The original intent was likely to have them pretensioned at least to the point that they would engage immediately upon application of a lateral load.

 

[bones206]

I typically see H/60 in their specs for wind, but I suspect they keep deflections well under that limit when using diagonal bracing, like H/200 or better.

To get the stiffness, I usually model the X-braced plane frame and apply a unit lateral load at the eave to get an overall drift deflection.

 

[bones206]

This is a photo of one of my future frankensteins I scoped out over the summer. The cable bracing is slack because I suspect it yielded under overstress. It looks like bracing was removed on the other end of the building for a new overhead door, and so this brace had to do more work than it was designed for. You can see in the far top right the end-wall roof bracing was all buckled as well. Yet... no visible damage to the roof or wall paneling.