Practicalities of Load Testing Spreader Beams 1

[bbqq021]

We are due to complete an offshore lift of approximately 370 tonnes in about 2 months. The structure is a mid-water-arch which will support 2 risers for an FSO in the gulf of Thailand. Our installation contractor has specified the dynamic factors which considers the motion of the floating vessel and the added mass as the structure is lowered through the surface and then through the water column, long story short we have come up with a beam of 600 tonnes SWL. The design of the beam will follow B30.20 as well as the supplement ASME BTH-1 for design. The beam will only be used once.

According to BTH-1 the Design category of this beam is “B” (environmental loads are not easily predictable) and so the design factor is 3.00 compared to 1.67 from AISC or 2 for Cat A. This is all fine and understood.

My problem is that the company I work for mandates that all portable lifting equipment should be proof loaded according to BTH and OSHA (125%). This is not practical and in the past when we have had large lifting equipment we have used NDT and the design of the beam to demonstrate integrity and compliance. The problem is BTH has very weak language like “should be load tested to 125%”. I know NDT and design is a common practice when it comes to large lifts however I cannot cite a code or practice. Can anybody point me in the right direction?

My second question is what kind of NDT? I guess the answer to the first question may include an answer to this however at the very least I was thinking 100% RT where possible and 100% UT for other primary structure.

Thanks for your help in advance.

Will

 

[BAretired]

What is an FSO? What is SWL? What is BTH? I assume NDT is non destructive testing. Please confirm. What is RT? What is UT?

When you use acronyms, you must realize that we are not all in your particular line of work and we probably are not all familiar with all of the codes which apply. You should define acronyms at the beginning of your post, then use them freely from that point on.

I am not sure what a mid-water arch is but if you provide a sketch of it, I might be able to contribute some little tidbit of advice.

If you are using a load factor of 3.0 instead of 1.67 because some loads are not easily predictable, it is not clear why load tests should be required. If in doubt, use a load factor of 4.0.

BA

 

[bbqq021]

FSO - Floating Storage Offloader (not important)
SWL - Safe working Load
BTH - Below the Hook (ASME Standard) American Society of Mechanical Engineers
NDT - Non Destructive Testing
RT - Radiographic testing
UT- Ultrasonic Testing
MWA - Subsurface Structure weighing 370 tonnes.

 

[ToadJones]

So are you trying to avoid load testing the spreader beam?

I'm not sure you can avoid it especially if the owner wants you to do it.

I do work for a company that does a lot of this kind of work.

They load test spreader beams quite often.

Sometimes they have to make spreader beams to load test the spreader beams!!! They often do this using enormous water weights.

They are often contracted just to perform the load testing of the spreader.
I have seen some enormous spreaders that they tested used only once and then scrapped. It made me sick to see these big beautiful pieces of engineering scrapped.

I'll check with the guys at their office tomorrow if I remember and see if they have been able to circumvent the testing process in the past.
I am almost certain they have loaded tested to 600 Tons.

 

[rb1957]

why is testing to a load level impractical ? if it was designed for this load ? or is this a schedule constraint ?

or is the spec saying design to a factor of 3 and then test to a factor of 4 ?

in which case, wouldn't you design for a factor of 4 ?

at the end of the day, what does your customer want you to do ?

 

[bbqq021]

Thanks ToadJones

Yes, this is only a single lift and then the beam will be scrapped. We are based in Thailand and so i am looking to see if there is a test house here that can do the work just incase we need to.

The ironic thing is there are 2 endless slings on the end of the spreader bar connected to 2 trunnions on the structure we are lifting. Due to the size of the lift it is impossible to test the trunnions since they are part of the structure, in these kinds of lift there always comes a point where the testing has to stop. (You don't test the structural integrity of the container you are lifting).

Water bags? 750 tonnes? (600 x 1.25)

I have worked with 4000t lifts (not the design). The spreader bars for these were 2000t and they were not load tested.

The other problem i have with proof loads is that they only test up to 125% of the SWL. NDT and engineering verifies the integrity of the structure up to the yield stress of the steel (300% in this case)

 

[bbqq021]

Sorry forgot to say, i am the customer.

 

[hokie66]

bbqq021,
I understand and agree with your philosophy that it is illogical to load test this weldment. I wouldn't really call this "portable lifting equipment". Design it, check the design, and do the NDT, which will necessarily be some mix of X-ray and ultrasonic. Selling this approach to your management sounds like it may be difficult, but that's why they pay you petro guys the big bucks.

 

[ToadJones]

I am fairly certain that they have load tested 600 ton gantry cranes to with water bags.
The bags are gargantuan.

Quick google search

http://www.waterweights.com/en/oil-gas.html

 

[dhengr]

It seems to me that the biggest sales job you have here is selling your own management on the lack of need for the testing that they are insisting on. How much experience have you had for them on these kinds of structures? How many successes vs. problem jobs or designs (I hesitate to say failures), with and without proof testing? Why would you design, conduct fabrication NDT and then load test the spreader beam to a higher level than has been used on the load itself; the trunnions, their surrounding structure and welding, etc? Obviously, one of the considerations in this lifting job is at what point it could/might fail, and who’s responsibility that would be, and who’s insurance company would be on the hook for the loss. And, that decision lies with your management, their faith in your design and your lift management abilities, and their level of discomfort with any uncertainties. They really can’t damn you for doing what they tell you to do. And, all you can do for them is lay all your engineering facts on the table to help them make that decision.

A 370 tonne load and a load factor (L.F.) of 3 doesn’t add up to 600 tonnes safe working load, does it? Which is it, 1.67 or 3? And, at this L.F. is the design to Fy or Fu and/or buckling failure? A good and thorough structural design is essential on this type of equipment, as is good clean weld design and detailing. I’ve never been convinced that all welds need elaborate NDT, and the type of test you would do is a function of the weld and joint details, and the weld stress level, for meaningful results. The quality of your design details and directness of load path from the upper hook to the trunnions on the load probably has more to do with a good spreader beam design and a successful lifting operation than load testing the spreader beam designed with a L.F. of 3.

Some of the L.F’s., testing criteria, etc. in ASTM BTH-1 are based on a piece of equipment which will be used many times, under many different sets of conditions, different shackles, different sling lengths, etc. and different loads. Your’s is a one time use and seems fairly well defined if you can trust your contractor’s loads.

 

[Teguci]

So scenario #1 - Beam is fabricated and load tested in the shop and fails. Cost will now include refabrication, and retesting and might infirnge on the lifting schedule.

Scenario #2 - Lifting beam is delivered and rigged. During the lift it fails and now a new coral reef can be introduced to the area. Schedule is impacted, your company is on its way to bankruptcy, you have nothing to show for it, and some very powerful people are asking you some very uncomfortable questions.

Scenario #3 - Lift beam works no problems.

As a frugal person, I love scenario #3, its just the distinct non-zero probability of #2 occuring that has, in the past, kept me from sleeping some nights.

Just to confuse the issue more, we do erect buildings and bridges without load testing them first. The bottom line for me is that this is a question an engineer can't answer. With a load factor of 3 I could see myself foregoing the load testing but, in case of a failure, I'd make sure those powerful people I mentioned above were in the room with me when those uncomfortable questions were being asked (CYA - should be a universal acronym BARetired).

 

[BAretired]

Yes, Teguci I am familiar with that one.

BA

 

[Teguci]

Just ribbing ya - I didn't know what FSO was either (smile)

 

[pwildfire]

B30.20 does not Require load testing. It does Recommend load testing to Exactly 125% rated load.

My company produces lifting devices for various industries, although not on the scale of yours. We load test many of them, but not all. Spreader beams are typically simple enough in design that a thorough engineering analysis can be easily performed, and in most industries it is acceptable to have this analysis evaluated and stamped by a registered professional engineer.

NDE typically is only used to modify the design factors to achieve a more accurate analysis, for example weld efficiency is assumed to be higher based on the level of examination. In other words, NDE is always a good idea, but only in conjunction with an appropriately designed structure.

Basically I think this comes down to the comfort level of the companies and their legal personnel more than any code required engineering practice.

 

[bbqq021]

Just to put a little more perspective on the question. Designing and building a new spreader beam will be about 25 to 30 grand. Load testing to 750 tonnes (considering this company has only gone to 600 tonnes)- I am guessing 200K. (not to mention the fact that i don't think there is a testing house of this scale in Thailand).

http://www.waterweights.com/en/oil-gas.html

Of course the consequential losses of not meeting first oil dwarf any cost of any test however that argument does not help since that would also warrant making it out of titanium and flying it round the moon and back.

I have spoken to a few spreader beam supplier companies and the answer i got from them is similar to what i am getting here. There does not seem to be standard or code, it is just industry practise. Just for the record and for people in the future, this is what i am doing.

Static weight of the MWA (inc rigging) is 370 tonnes.
The floating crane is not heave compensated so i use a DAF of 1.3
The structure has large mudmats so i will use a snagging factor of 1.2 as it passes through the surface and through the water coloumn. (TOTAL factor = 1.56).

370 x 1.56 = 577 tonnes (BEAM RATED to 600 TONNES)

There were actually 3 different load cases which i needn’t go into. It turns out the half in the water and half out was the most onerous.

100% in air
50% in air
100% in water

BTH (Below the Hook) splits the lift into CAT A and CAT B, Broadly speaking CAT A is where the environmental loads are easy to predict (an onshore lift perhaps). CAT B is where they are not. It also goes into ratios between dynamic and static portions of the load.

I think i could have justified a CAT A lift since we are only lifting it once however to add some conservatism and to deflect the criticism i chose CAT B. Design factors for yielding and buckling are 3 for CAT B, 2.0 for CAT A.

I am using regular GR 50 steel and 100% UT on the longitudinal weld and the padeyes. All welds will be AWS D1.1 full pen. The beam is 914mm diameter and probably 2" steel. RT is no good on these kinds of thicknesses so UT was chosen.

Because there is no standard i am also writing an internal technical note explaining what i am doing and why. I will then get all of the relevant people (within the company) to buy into it.

Thanks for your help.

Will

 

[hokie66]

If you are forced to load test it, rather than using actual weights hung from the beam, I would look at doing it with hydraulic jacks. Would require a big foundation to jack against, but that should be simpler than doing an elevated test.

 

[Ussuri]

My experience is smiliar to pwildfire. You dont have to do a load test. For really heavy lifts it is demonstrated by design and by NDE. Smaller beams are load tested by then it is easier to do so. For instance, how do you load test if you are lifting a 1500 tonne topside module? However, I cannot point you at a definitive reference to support this.

On a side note, a DAF of 1.3 is applicaple for an offshore lift in air. All bets are usually off when you pass through the splashzone and deploy subsea. The wave loading on the structure may be significant, depending on arrangement.

You said you structure has large mudmats, have you mad an assessment of the hydrodynamic loading on your structure? Mudmats tend to present a large projected area so pick up high drag and inertia loads. The DNV rules for planning and execution of marine operations give you a method for doing so.

For small structures a DAF of 2.0 is commonly used for the subsea case. For larger structures where this it is more critical we have the lift and deployment analysed in software such as SIMO or Orcaflex. This allows us to reduce the DAF.