Lamella roof beam repairs

[archeng59]

I am working with a client to repair several damaged wood beams that are part of a lamella roof. The gymnasium was constructed in the early 1950's using 2x lumber for the lamella framing. Several members are cracked and split. A few are completely broken. The client does not want to demolish the building. I have prepared several details for repairs using wood screws, such as Headlok screws and Simpson SDS wood screws, and steel plates. But, never having designed one of these much less made repairs to one, I'm curious if anyone else has made repairs to a lamella roof system? The attached photo shows a typical area of the roof framing. I don't have any specific questions. Just putting this on here and asking for comments about your experiences with a lamella system.

 

[dhengr]

Archeng59:
I’ve never done the design of a Lammela arch roof, but I’ve always got lots of opinions, ideas, and some imagination and experience with wood design. Each repair might be an individual design and condition as a function of the failure type, location, etc., and as likely as not due to some condition with the individual wood member, a weak spot, a knot, a small split which grew, in just the wrong place. And, my first repair tack would be the same way we might try to repair any wooden beam/column which had suffered some damage. You probably also want to give some thought to a shoring, vert. jacking, spreader members/lt. beams (relieve member compression), tie members/lt. beams (relieve member tension), gluing, gang drilling of steel pls., in place, etc. etc. Many times the means and the methods are as important to a project like this, in terms of its success and quality, as fine tuning design numbers. Can you make any comment (photos too) on the various failures, any commonality within them, etc? Are the distressed areas out within the members or are some at member joints? The roof has pretty much proved itself in terms of withstanding the test-of-time, but how old are these distressed areas and was there any one-sided loading (wind, snow, etc.) or any concentrated loadings (mech. units, roofing material piles, etc.), any movement or distress at wall bearing points? Let’s see your details and discuss them, along with photos of the failure they are intended to fix. You could probably use some computer software on the whole thing to get a good picture of the loads and stresses in members at various locations if you wished. Below you will see some simplified, long hand methods to take a first shot at these. These where undoubtedly the methods used originally, same erra.

For your reading pleasure:
1. “Structure in Architecture,” by Salvadori and Heller, pub. by Prentice-Hall, 1963.
2. “Structural Design in Architecture,” by Mario Salvadori and Matthys Levy, pub. by Prentice-Hall, 1967.
3. “Timber Design and Construction Handbook,” by Timber Engineering Co., pub. by McGraw-Hill, 1956.
4. “Timber Construction Manual,” by AITC, pub. by John Wiley & Sons, 1966

 

[archeng59]

There are no roof top units. As far as how long the damaged members have been there, I have no idea. The school system is a small rural school that has a lot of administrator turnover. There is no one currently on staff who has knowledge of when any of the damage was first noticed.

Bearing points along the concrete walls appear to be ok. Most of the damage is splitting along the member length. The split locations vary a lot. Some splits are near midspan and mid-depth. Some are near one or both ends. The broken members, two of which are visible near the center of the photo I provided, are typically broken at or near the midspan where adjacent members are connected to the beam.

One reason for my request for comments is that the school contacted a retired structural engineer who told them he wanted no part of the project to repair the framing. He told them to tear down the gym. Made me wonder if he knows something I don't know.

My final concept was not to replace any of the members, although that was my first idea until I started thinking about how removing and replacing a member would actually occur. Instead, I detailed long screws installed from the bottoms of the 2x members through the split areas. For the broken beams, I detailed adding U-shaped plates that are screwed to each side of the broken area. The beams are jacked up to close the member before adding the plates. This is similar to another repair that was made sometime in the past that appears to have performed well since.

 

[OldBldgGuy]

I have always thought that these were amazing roofs, structurally and aesthetically. I haven't repaired any, but I've repaired other light glulam arch, built-up lumber, and similar wood structures. I think you have to investigate each failed member and see if it was a flaw in the member that was the reason for the failure, rather than some extraordinary load, and repair the member to restore the capacity that was originally there (or better). You have to jack the broken members before repairing them, you have to have the repaired member sharing the load. If you just patch them now, they don't do anything until overstressed again, and they don't relieve any of the stresses that they have now transferred to other members.
My opinion, but I don't like visible steel hardware and reinforcements, I want to restore the aesthetics as well as the structural capacity.

 

[oldestguy]

I'd first try to figure if the split is something related to shrinkage, possibly failed originally strained zone, present in the original log before sawing. Assuming no "built-in stress finally showing itself", consider compressing the split sides and glue the closed split together. You don't want another split to emerge beside the glued first one. When it comes to visible reinforcing, etc. how many of the owners, users, etc. ever look critically at the ceiling anyhow? With plates and flat head screws in flat cone shaped holes, there isn't much to see.

 

[Triangled]

i'm jealous,
definitely agree owner should not disassemble...
a long time admirer of these. there is one local i like to take the new grads to.
same books as dhengr

regarding the 2 broken members, the vertical break suggests a weakness at a knot to me, and then the adjacent seems more like a bending stress failure, maybe after sharing load from its adjacent member. The wide stripe of different color sheathing.... is this a repair of some kind? maybe the repair contractor stacked too many materials at the location of the two breaks. "small rural school that has a lot of administrator turnover" maybe nobody paid attention, contractor heard a sound, pulled off the materials, didn't "see" anything and proceeded.

regarding other horizontal splits unseen in this photo, they might not be a big deal, depends a lot on where they are. I remember talking to an AFPA guy years ago about Fv = 95 and increasing with reduced split restrictions (anyone remember that?) - he basically said there was enough reserve capacity that, from the Fv standpoint, the board could be split end to end, and 95 still be ok. can't say i ever did the math on that, but i have noticed, sure enough, allowable horizontal shears have gotten a lot more while almost every other lumber allowable has gone down.

 

[oldestguy]

That long dark streak in the roof sheathing may be a replacement of that area for some reason and likely the sheathing takes a part in sharing the compression loads. Suppose this is a replacement of sheathing, then any compression loads in the sheathing there get dumped on the various beams. Overloading those beams and maybe putting a dip wrinkle in an otherwise smooth roof surface might be looked at. If so, jacking up the roof and installing new sheathing tight to adjacent sheets replacing the dark strip would be required, in addition to any beam repairs. Without replacing so that original compressive stresses are in the sheathing, my guess is this will fail some day even if the beams are reinforced.

 

[thebard3]

That looks to me like something that wouldn't meet building codes today, but I'm no structural guy.

It's interesting that the members do not meet end-to-end. It seems like that would limit their load carrying capability. I guess maybe the fact that they mare joined at an angle might negate any benefit of an end-to-end arrangement.

Brad Waybright

It's all okay as long as it's okay.

 

[archeng59]

The long strip of different color sheathing is where the original skylight was removed at some time in the past. There is no one currently working at the school that remembers seeing a skylight, so it must have been removed more than 20 years ago as the current custodian has been there since 1998. With regards to the splits, there was no rhyme or reason to why they occurred or where. The broken members were damaged due to a load issue, in my opinion. But, the custodian said they have been broken since he started working at the school in '98 so don't know how long the damage was there. He told me, "Everything you see is how I found it when I started working here. Nothing has changed."

 

[Agent666]

I'd think because some bits are broken/damaged that they currently aren't carrying any or much load as they have redistributed their load.

So if you're cutting out the old and replacing, then you're not going to get back to the original state of the new members actually taking any of the dead load. Just something to be aware of I guess. But if they have been like that for 20 years then it obviously shows the degree of redundancy that must be present in these types of systems. I guess you can analyse the structure with the members being taken out and see if it works or not.

 

[Sawbux]

Don't see any tie-rods across the gym. Are there big concrete buttresses keeping the arch from spreading? If these have been modified or removed, it would allow the arch to spread, causing bending failure in the segments.

I would repair with screws and epoxy. There is a lot of redundancy, so likely not too critical unless a whole series of these are breaking.

Is this in snow country? Maybe some unbalanced load caused flex failures.

 

[LaminatedTimber]

Wooden lamella roofs were known for not handling unbalanced roof loadings and there were many, many failures and collapses. Some lamella roofs are doing fine. I know of at least five in Wisconsin that have been performing fine for 80 or so years. A school gym was built in Fargo, ND in the 1960s with solid wooden lamella and it collapse within months of completion. The guy who sat at my desk before me bid it as a glued laminated timber radial rib dome and when told it would go to lamella he "warned" the general contractor about problems. Hate to say "I told you so" but that's what happened. Luckily there were no injuries in the collapse.

I cannot help with your design and repair method.

Lamella is a real groovy looking type of material. The historical and iconic Brown Derby in LA has a lamella roof.

Andreas

 

[McSEpllc]

The lamella roof was developed and patented by Friedrich Zollinger after WWI as a way to address the housing shortage (it uses about 46% of the lumber compared to the previous wood construction techniques used in Germany.

https://en.wikipedia.org/wiki/Friedrich_Zollinger

http://www.holzbau-statik.de/ibr/do...uweise _Bautechnik 1992-Heft 4_S. 190-197.pdf

Here is one paragraph of a German article:
"Static calculations alone are not useful. Because the rigidity of the structure under load depends largely on the properties of the bolted nodes, which are beyond mathematical modeling, such as for example manufacturing tolerances or pre-tensioning of bolts. For this reason, endurance tests make sense today."

It might be good to review the connections up close to check whether they are loose, tight, or pulling through or something. It might reveal what the nature issue is.

The overall dome structure should be axially loaded, but the two failed members in the photograph are bent down visibly when holding a straightedge against them - Gauging from the rafter thickness I guess about up to 4"?

The offset and through-bolted wood members likely experience prying action at the ends for any localized deflection (maybe from unbalanced snow loading).

Personally I do not feel comfortable to propose a repair to an owner unless I feel confident I know what the failure mechanism is. That is probably why the previous engineer passed on this project. Once I know the failure mechanism, I can figure the load that the members has to be able to resist (and couldn't.)

 

[chicopee]

The picture shows only one cracked section near the original skylight placement. That section may have been damage during removal of skylight or when the new replacement planking was installed. Examine the cracked sections carefully to see if the damaged were man made during the roofing work. My first inclination, eventhough I am not a structural engineer, would be to realign the cracked joints and double strap (plywood sheets on both sides of realigned joints) secured with thru-bolts.