Overloaded pipe sleepers 1

[npthao121]

Dear All,
At our site, we have installed H-244x175x7x11 mm (grade ASTM A36) beams as pipe sleepers. These sleepers will support some gas pipelines (30¡±, 20¡±¡¦) and piping contractor finished pipe installation already. After that we found the H beams seem to be slightly overloaded. By visual investigation we can measure deflection of 30-35 mm per each beam (the span is 6 m). However the beams is still stable.

I rechecked the beams by AISC-ASD89. The demand/capacity ratio is about 1.11; then I backed to take a look at the calculation sheet and found H-294x200x8x12 mm beam was designed for the sleeper originally. There is a difference between calculation sheet & design drawing. It seems to be that this situation due to the designer¡¯s mistake, they did not check CAD draft work, I think.

The owner does not know about the problem until now. We want to correct the problem by reinforcing the H beam, but it is very difficult to reinforce or replace the beam. We cannot touch or move the pipe & other cable tray¡¦installed on the sleeper already. Any way the gap between the sleeper & the ground just about 300 mm. Therefore, we cannot do anything unless we remove all things on the sleeper.
Unfortunately we are the EPC contractor so that I can not blame to designer.
Anyone can suggest the right way to overcome the problem? Can I do nothing & just keep silent. In all honesty I am not much worry about the beam strength, it can be OK with the ratio of 1.11.
Thank in advance.

 

[prex]

If you really can do nothing to overcome the problem, then it's up to you to decide: even if the structural safety was not much at risk, the problem could be discovered by your client in the future and you could then face a more difficult problem.
Otherwise you should inspect in detail the present situation and finally decide whether some type of modification is possible.


prex

http://www.xcalcs.com

Online tools for structural design

 

[corus]

Design codes give recommended design limits for a structure but there are instances where these can be overlooked. If a design has been known to operate successfully in the past for the same loads and conditions then there is no need to assess the operating loads against design recommendations. In your case you have exceeded the design limits by 11% for a static load case and the structure has not failed. Bearing in mind that the design codes do have a safety margin in for unknown factors your design appears to be operating successfully. If the design is subject to variable loads however where fatigue would be a consideration for example, or where seismic loads might need to be considered, or wind loads etc. then I'd worry.

 

[Ron]

npthao121....don't ignore the issue. It could come back to haunt you and others.

First, make the designer aware of the issue. He has the most culpability in the issue. Tell him that either he must inform the owner of his mistake or that you must do so as an ethical matter.

Gas pipelines and other code-covered design have code requirements for many reasons, one of the most important of which is protection of the public. While I see "Corus's" point, that isn't the way it works in the public realm. You must have and show valid reasons for deviating from code minimums, not just that someone made a mistake and it probably won't fail based on history. History is only one predictor of structural performance....if used as the only predictor, we would have a lot more structural failures than we have. Most structures rarely see their design loads, but they are to be designed and built so that if that event occurs, they will not fail as a result.

As for the solution, you can gain capacity in the existing beam by adding a channel section along the bottom flange. The channel should be sized compatibly with the beam, installed "leg up" toward the beam flange to create a "box" section at the bottom flange and welded with a continuous seal weld (adequately designed fillet weld and/or partial penetration weld). If that cannot be achieved, a channel section of width greater than the beam flange width could be installed with legs down, and welded to the bottom flange of the beam with a horizontal fillet weld from the top side. This would be an easy fix assuming you can achieve adequate compensating capacity.

Good luck.

 

[npthao121]

Thank everybody for your comments,
Ron, I have the same mind with you regarding the problem. Actually, i considered the solution to reinforce the H beam by welding channel to the bottom flange of the beam. but the problem now is we do not have enough working space for our welder. as i already mentioned above, the gap between bottom flange & ground level is just about 300 mm in average. I think if we decide to reinforce the H-beam we have to take out everything and it will be a disaster for us in the aspect of schedule & cost. I am considering to soil digging by manual to increase the distance, but still difficult for welder to work in such condition.

 

[dbuzz]

Ask the designer to explain the discrepency beteen the calculations and the drawings. You may find that the change in size was an approved change, late in the piece, but that the calculations were not upgraded to include the change. Although this is poor practice it does happen from time to time.

How have you calculated your load factor of 1.11? Have you assumed point loads or distributed loads (which may be valid if the pipes have a saddle seat)?

The measured SPAN/200 to SPAN/170 deflection due to dead load looks to be a bit of a problem. What additional load and deflection are expected from the pipe contents?

If the beam proves to be non-compliant with design code requirements you really are obliged to inform your Client.

 

[npthao121]

Dbuzz, thank for you joining,
I talk with the designer already and he said it is a mistake, the right section is H294x200x8x12 instead of H244x175x7x11. He did not realized when reviewing the AFC drawing.
1.11 is not a load factor; it is the demand/capacity ratio as AISC-ASD89. Load value & partern provided by piping designer (concentrated load at several positions on the top flange of the beam) there are some guide & shoe supports to connect pipe to beam.
What i am worrying now is the hydrotest for piping. some additional load (by water) will be applied. But we can make some temporary supports during the hydrotest.

 

[dbuzz]

I take it the demand/capacity ratio of 1.11 is for pipe self-weight plus pipe contents (oil). What is the demand/capacity ratio for pipe self-weight plus pipe contents (water)?

Has the designer suggested any remedial modifcations, stiffening or reduction of span (if that is even possible)?

One option may be to install flange "doubler" plates. Here is a quick calc based on acheiving equivalent stiffness:
Ixx(294x200x8x12) ~ 113 x 10^6 mm4
Ixx(244x175x7x11) ~ 61.2 x 10^6 mm4
Add 165mm x 12mm "doubler" plates to top and bottom flanges
Ixx(built-up section) ~ 125 x 10^6 mm4

You'll also need to check the built-up section for strength and design the longitudnal welds to suit.

 

[DSB123]

npthao121,
You seem to be concentrating on the supports for the piping and not considering the effects on the pipework. For example if the pipework is connected to any equipment and the supports have deflected by the amount you mention then the loadings onto the equipment could be high and of concern. Suggest you have a word with the piping engineer regarding the possibility of excessive loads on terminal points.
As regards the strengthening of the H beam suggest that a channel or fabricated channel from plate could be welded to the bottom flange of the H beam from the top using a CFW provided the channel depth is made greater than the flange width of the H beam. This would not be restricted by the ground clearance then.

Hope this helps.

 

[Ron]

npthao121...300mm is adequate clearance to repair as noted. Welder can use wire-feed process (Flux core (FCAW) process would be preferred for outdoor use), or use "clipped" electrodes (clip about 100-150mm off each electrode and weld in short sections...takes more time but will work)

 

[dik]

Is it possible to do anything at the supports to shorten the span a tad?

Also is it possible to provide a bottom support for the beam and use high strength cables to 'lift' it?

 

[dbuzz]

DSB123 makes a very valid point about the consequences of the excessive deflection on the piping.

Regarding the repairs, I would be disinclined to weld a channel to the underside of the bottom flange, toes up to create a box section. The built-up section will have an irregular distribution of bending stress and very convoluted shear flow. I would also be disinclined to weld a channel to the underside of the bottom flange, toes up such that the toes run past the edge of the flange. This will create a "box" for a build-up of debris and dust which will retain moisture when wetted and may contain contaminants - conditions that can lead to accelerated corrosion of the steelwork.

Instead I would recommend you use either (i) flange "doubler" plates descibed in the post above, or (ii) a beam or column tee with the web coincident with the 244x175x7x11 beam web.

 

[npthao121]

Thank everybody for your valuable ideas,
I discussed with the designer and he also proposed to reinforce the beam by welding aditional channel. Fortunately, channel material is available in our site. Also i can mobilize the high quality welder from piping team. But welder may have to take the FCAW qualified test as CA requirement.
Anyway thank you one more time, I've got a point and ready for the meeting in the next Monday.

 

[dbuzz]

If you must use a channel to reinforce the beam may I suggest you place the channel on the top flange, toes down, with the channel sized such that the toes run past the edge of the beam flange (similar to a crane runway girder).
____
|-----|
|
|
|
-----

You'll probably find that this will give you satisfactory stiffness, section moment capacity and member moment capacity (ie. lateral-torsional buckling capacity).

However this may be time-consuming to install if the pipes are already in place. How may beams do you need to strengthen?

 

[Ron]

npthao121...let us know what is decided for the final repair. It is good to close the loop in these discussions when we can.

 

[Focht3]

Wouldn't it be faster - and safer - to add a mid-span support beneath as [blue]dik[/blue] has suggested? Just a simple support - nothing more. Dig and pour the footing, then use jacks to temporarily raise the beam to the proper level. Support the beam with "appropriate" concrete blocks that are shimmed with thin steel plate. Remove the jack. Install rebar for the pedestal and pour - leaving the concrete blocks and shims in place. Voila! Your problem is solved. And a key point: at no point during this repair method have you done anything to make the existing situation worse.

If the soils aren't expansive, this would be a very cheap - and effective - fix. And it would significantly increase the system's capacity, which is a benefit to the owner. Welding around all those pipes will be slow and difficult considering preheat, stress release (alluded to by [blue]dbuzz[/blue]) and difficult working conditions. And the owner will probably require 100% NDT (probably UT) to verify weld quality. (Considering what has transpired, it won't be your - or the designer's - call.)

And what will you do if you have a lot of bad welds to gouge out? You'll have no choice but to fix them. Ouch!

I'd explore this a helluva lot more before I committed to a welded solution. Talk to the project geotechnical engineer. Confess the mistake, tell him you're looking for a viable solution before you notify the owner. He will likely have simpathy for your predicament, and may offer up solutions of his own. It's worth a try.



Please see FAQ731-376 by [blue]VPL[/blue] for tips on how to make the best use of Eng-Tips Fora.

 

[CWIC]

Just a few comments to those already posted:
1.) I feel the 300 mm is an adequate distance to allow access of the repair by a welder at the bottom flange of the beam. Production and repair welds have been performed with much less.
2.) If a weld repair is performed from the bottom flange, the channel can be positioned and held in place by a jack. The jack can also be used to remove the deflection from the sleeper while the weld repair is in progress.
3.) Flange-up channel sleeper racks are not uncommon.
4.) The pipe welders ASME qualification (certification) may encompass some of the essential variables for welding being performed under the D1.1 structural code. A little research and review may be in order.
Where is the welding being performed specifically in CA? BTW: Welding performed in any petrochem or processing plant in the City of Los Angeles requires the welder to sit a written exam in addition to the performance testing. If the welding is not performed where a local jurisdiction prevails, then I would review the welders' current ASME qualifications before testing.
5.) I also agree with DSB's comment regarding verification of loads imposed on equipment or other piping systems. If these loads are such that repairs (including stresses encountered during a weld or any other type of repair) are negligible, then go for it.

Comment:
We modified existing pipe racks when additional ASME B31.4 piping systems were added to existing pipe systems on a job in the Middle East by adding inverted T stiffeners. It should be noted these were long runs, no equipment was subjected to loading during modification. Pipe flanges were the only elements that were of concern. An erection plan was required from the contractor to ensure method of loading, welding sequence and temporary supports were approved prior to modification. A hydrotest was required for the existing piping and the new piping systems. All passed without any problems.

IMHO:
I think this repair (in-place bottom flange stiffener) can be performed with satisfactory results.