Catwalk Design 2

[Lesali]

I (sole proprietor) work in the cellular industry where the catwalk rails of elevated water tanks are often used to hang antennas. These antennas that once weighed as little and 14 lbs each, now weight 97 lbs with pipes and mounts. With 3 carriers x 4 antennas each, failure has occurred on one catwalk and the owner is requesting analysis on his other towers. I'm imagining needing to use a 3D FEA program (like RISA or SAP) to analyze these as the rails appear to be structural space frames supporting the plate deck. But those programs were not available in the era these tanks were built (1930-1950). If anyone has had experience or know of resource material on the original design of these catwalks I would appreciate the guidance.

 

[CELinOttawa]

One of the senior Ketchum's books touches on this... Structural Engineer's Handbook I believe. Effectively you're not going to be able to rely on the original design approach, however, as you are now changing the imposed loading and I doubt the condition is completely undeteriorated.

You need to apply modern methods to these historic steels. If you've not done it before, it can be a little awkward. You might want to read up first, and I'd suggest "Structural Analysis of Historic Buildings: Restoration, Preservation, and Adaptive Reuse Applications for Architects and Engineers"by Rabun.

 

[JStephen]

The primary purpose of that catwalk is as a stiffener for the support of the tank, not for personnel access. I would expect that the balcony supports the handrail, not the other way around. Typical handrails have diagonal members, but those are not universal, and some just use conventional vertical posts for handrails. On occasion, customers would want to avoid the balcony to prevent vandalism, and in those cases, there would be a stiffener inside the tank instead.

There are some rules for the design of the balcony and leg support system in AWWA D100 (maybe it's just me, but those rules never have entirely made sense, either!) However, on older tanks, that was a very approximate approach at best, and used very conservative stress levels, and relied to some extent on past experience or strain-gauge surveys for confirmation. I've heard of a tank being built and then sagging down around the legs, so presumably, others ran into problems with it, too. It is most definitely not a cookbook approach where you could take someone else's tank and calculate the same stresses the designers used when they built it.

If you go to analyze it using FEA, I'm not sure how you'd know where to start other than analyzing the entire tank, and that would present its own challenges, especially if you didn't have drawings of the original tank. Also, note that due to the approximate analysis used, you could very well find out a tank is "overstressed" before you do anything to it.

 

[Lesali]

Thanks for the responses - good food for thought. I already replaced a catwalk that was rusted and deteriorated on an "abandoned" water tower. But in that instance I was able to redesign new elements based on today's standards and stiffen the remaining cantilever supports. I had not considered the catwalk as a tension ring stiffing the tank base. I'm hoping the tower owner can come up with some drawings otherwise the carrier will have to order a tower mapping. But as I imagined, justifying an existing condition can be a big can of worms.

 

[SteelPE]

I remember attending a seminar given by the AISC where the lecturer mentioned that he once ran the analysis of an existing structure using modern methods (I think for an addition). He said that the structure failed his analysis badly under the existing loading conditions.... but the building was standing and in use for over 50 years????? They decided not to make any changes to the structure.

A few years later he came across a historic paper that described a way to perform the analysis on the type of structure he was looking it. He ran the existing structure through the methodology described in the paper and what was an overstressed structure quickly became one that worked. He then went on to say that as long as you provide a positive load path the structure will find a way to transmit the loads through your system to the ground.

Point being, if you are looking at an existing structure that was not designed using FEM and you suddenly decide to use FEM you may find more problems than you expected. Good Luck!

 

[paddingtongreen]

I agree with SteelPE. I have used this approach, when needed for years. It started when an acquaintance removed the steel column in the middle of his basement, without knowledge and without shoring - the house didn't collapse, in fact, it showed no distress. I concluded that the elements of the wall framing above, the top and sole plates being the "flanges" and the wallboard on both sides of the studs being the "web", acted together as a deep girder.
On a job that should have actually failed, I concluded that the wall board had to be carrying a surprising amount of the vertical load.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[CELinOttawa]

"He then went on to say that as long as you provide a positive load path the structure will find a way to transmit the loads through your system to the ground."

This is how plastic analysis works, and cornerstone of Capacity Engineering, though not at all how CE works...

"I concluded that the elements of the wall framing above, the top and sole plates being the "flanges" and the wallboard on both sides of the studs being the "web", acted together as a deep girder."

Agreed and entirely proper, except I hope you had them increase the fastening... The failures of such systems are spectacular and without the required warning (or "soft" failure) in advance.....

 

[CELinOttawa]

All of this reminds me of Wooten's third law...

http://www.aisc.org/WorkArea/showcontent.aspx?id=18544

 

[AELLC]

LOL, CELinOttawa, a topic dear to my heart, since I studied what we called Shakedown Analysis in University back in 1975. It would be more properly described as "cyclic gradual collapse of steel frames as the loads keep reversing and moving" - progressive collapse.

 

[CELinOttawa]

AELLC: Did you ever do a pushover/real world shakedown analysis of the classic single degree of freedom system? You know, the old weight at the top of a column? I was shocked the first time I saw it, and then immediately had the light-bulb moment as I connected plastic design and what was happening on the shake table.

As the column started to lean, we started taking bets on whether or not the column would manage to lay down on the table. It did.

 

[AELLC]

No, never saw that occur. All my studying was trying to complete a Master's Thesis - I used only research papers and textbooks as sources. Never actually saw something in the real world. I still have the papers, all hand-written. Mostly 2D frames, single bay portals.

I did actually apply my thesis (somewhat) at my first job, related to a "minor" problem in oil producing structures in the Gulf of Mexico. The wave action was evidently failing a steel support, no one could figure out why. They were only in 350 feet of sea back then.