Lifting Trunnions 3

[MikeG7]

Not sure if this is the best location in the forum to post this question:
The WRC-107 analysis done for a lifting trunnion assumes that the trunnion is connected to the shell and calculates the stress in the shell and in the pad if a pad is used.

I have a situation where I want to weld the trunnion only to the pad. The stresses in the trunnion would be the same but the effect on the vessel would be different as only the pad is transferring the load to the vessel.

Would it be unreasonable to use WRC-107 for this configuration?

 

[jte]

Not sure if this is the best location in the forum to post this question

Yes, I'd say this is the most on-topic forum for this question.

Would it be unreasonable to use WRC-107 for this configuration?

Yes, I'd say it is.

If I wind up reviewing this vessel, I will ask for an FEA. Said FEA will be nonlinear in the sense that contact between the repad and shell will be explicitly modeled.

My gut feel is that this isn't going to turn out well - but as a friend famously says, "Show me the numbers!" If the loading is light enough and/or if the repad is stout enough, perhaps this could work.

 

[TomBarsh]

I might do this, using the reinforcing pad as the attachment to the shell. But then bear in mind that you would need to investigate the stresses in the reinforcing pad itself as it is transferring the entire load from the trunnion to the shell (actually, vice versa). It is no longer a "reinforcing pad" (or plate); instead it is a structural component in the direct line of load transfer. Its analysis would be a bit "weird", I suppose you might find a case in the Roark-Young book...circular plate supported around the perimeter, subjected to an applied bending moment.

Personally, I would do this only for the very lightest/smallest vessels.

Other than that, I would follow JTE's sage advice.

(Now that I think about it, there was a similar thread at eng-tips within the past year or so.)

 

[TomBarsh]

more discussion:

http://www.eng-tips.com/viewthread.cfm?qid=358182

 

[MikeG7]

Thanks Tom & jte
You both got me a bit worried about this. The vessel in question is quite large it is a storage tank about 20ft in diameter and 30ft high, weighing about 14 metric tonne.
Compress report warns that the WRC-107 analysis assumes the trunnion to be welded to the shell and that the pad is just considered to give strength to the shell and not transfer the load.

If the pad was sufficiently large and thick and we ignore the tank wall completely (meaning leave the shell thickness out of the analysis) and only consider the trunnion pipe on the pad, then the numbers would show what's happening in the pad. This would be OK right?

 

[MikeG7]

Thanks for digging that one up for me Tom

 

[SnTMan]

I think the only valid WRC-107 approach would be to treat the pad as the attachment. Assuming that dimensionally it is within the ranges(s) covered by the various tables.

Regards,

Mike

 

[MikeG7]

I'm not familiar enough with the WRC method yet but all the advice seems to steer me away from what I am wanting to do.

 

[JStephen]

I would suggest:
Check using trunnion as the attachment, repad as the "shell", with repad thickness only.
And check using the repad as the attachment, tank shell as the "shell", with shell stress only.

 

[dhengr]

MikeG7:
This tank sounds more like a storage silo than a pressure vessel. At 31kips and 20' in dia., by 30' high, by what thickness (about .375" max. for that wt.)? Is there a roof and a bottom in this tank, what are they? In what position is it built or assembled, and/or tranported? What position are you lifting it from and to? Are there any stiffeners on this tank shell? Maybe you could involve them in a lifting means, easier than adding repads. On too thin a tank of this dia. two trunnions as a lifting means would likely cause buckling of the entire shell, as well a very high localized stresses in the shell at the trunnions, and the possibility of localized buckling. A big enough and stiff enough repad as part of each trunnion would reduce the localized loads , stresses and buckling, but they would probably not prevent the overall buckling issues. Is there any internal stiffening system in the tank for transport which might be used for lifting?

 

[MikeG7]

Hi Dhengr

I should have been clear it is a storage tank with a stiffened cone roof and the shell is 6mm (0.236") stainless steel. The vessel is very flimsy, the roof is quite stiff.
The tank is shop built and lifted from the lying down to the vertical. there would be some internal cross bracing temporarily included to reduce buckling and removed after installing.
I will take your advice and use a large pad.
For those who posted advice earlier I have taken the advice and decided against using the trunnion only connected to the pad and to connect it to the shell directly with a pad.

 

[MJCronin]

I believe that a silo design of this giant diameter and thin wall should have longitudinal and circumferential stiffeners.

Were they included in this design ?

How are the proposed trunions located in relation to the silo stiffeners ?

How about some drawings ?

Has your company fabricated and off-loaded any of these huge silos in the past ?

How are you transporting a 20 foot diameter silo anyway ???!!?

It seems to have too large of a diameter to be transported by truck or rail ?

It has been my experience that silos up to about 13 foot diameter can be removed from a truck by using two "top of the shell" lifting plated located at 180 degrees apart.

Is this the first silo design that you have ever done ??

MJCronin
Sr. Process Engineer

 

[MikeG7]

MJCronin
The decision to transport the tank to the site fully welded is the fabricator / client preference. They have has transported similar size tanks before. It's quite a process I can assure you, telephone lines must be moved and special permits obtained first due to the height, but it is done. This is Africa! things get done here that shouldn't/wouldn't in other parts of the world. The picture attached is one example.

I have designed a number of vessels of smaller and larger proportions. By tank standards this is not massive, but it is big for a shop build. Bigger ones are usually sent in sections and field welded on site.

I usually design lifting plates (usually 4 plates) mounting to the top of the cylinder and extending onto the roof cone. But in this instance we are restricted in that no plates can be welded in this area.

There are some permanent circumferential stiffeners welded to the tank outside to keep it stiff (included primarily for resisting wind buckling and also aids in transport by keeping tank form).

Picture attached shows similar tanks of slightly smaller proportions being trucked out.

 

[MikeG7]

picture attached

 

[dhengr]

MikeG7:
The problem with trunnions for lifting this thin shelled tank is that they induce significant moments into the tank shell, with high local stresses and potential buckling, even with a large/stiff repad. As MJCronin suggests, and as you seem to be aware, lifting plates (lugs, in the plane of the shell) are much preferred for this application if there is some way you can make them work. They would have pin holes to accept a lifting clevis, or some such. Keep the cantilever length of your trunnions to a min., and if possible try to incorporate then with your external stiffeners, roof stiffeners, or the internal transport stiffeners, so there is some back-up structure to help distribute the concentrated loads. Where are the external stiffeners and any roof cone stiffeners? How far does the roof cone overhang the tank shell? This will determine your trunnion canti. length. Say the overhang is 4" all around, for discussion. You might need an 8" trunnion canti. and the round pipe trunnion really concentrates stresses at the top and bottom of the pipe shape, on the shell, and/or the repad. Why not take a (actually 2 of these 180̊ apart) light 10" WF or S structural shape and extend it vertically btwn. the top two external stiffener rings, where the reactions can be taken out and distributed into the structure. Weld as required and reinforce the top of its web for a clevis pin hole and lifting? This could be fabed like a WT with a web to the shell and a flange at a greater radius (10") from the shell. Another lifting plate alternative might be a 12" wide pl. (some width) standing vertical and welded to the outside of the tank shell, in its lower vert. portion. This pl. would be bent out and up at 60̊ to get outside the roof overhang, then be bent to be vertical again. In the top vert. region, this plate would be shaped and have doubler doughnuts for a pin hole and lifting clevis. These pls. are bent in a funny Zee shape, and when loaded there will be pl. bending and particularly bending at the two kink points (bend lines); the lifting pls. will try to straighten out. Some people call this shape a cranked beam or column. But, I think you could design these pls. for this condition, maybe with some stiffener pls., and it lessens the load concentrations in the tank shell, changes their orientations. Letting this canti. bending happen in the lifting pls. lessens the moment concentrations induced on the tank shell.

 

[chicopee]

Why not fabricate a cage that would contain, tightly, the entire length of the tank and then use the cage as the structure attached presumably to two cranes, to be lifted into place, then remove the cage. It is folly to build such tank with no forethought on how to erect them safely.

 

[Spoonful]

My 2 cents,

I don't see why can't use WRC107, except vessel/attachement are not within WRC107 limitations.

With trunnions only welded to the pad. Forget the pad for now, and use WRC107 to analysis for the vessel, and use the Pad outside dimension(assume pad is welded around the outside), instead of the trunnion's. because this is where all the load is transferred/applied to the vessel. If it pass, go back to check for the pad, separately, using Roark's formula for ring shape, and I would expect if the pad as flat can handle the load, it would most likely also ok when rolled to vessel shape(assume load is in longitudinal direction)

 

[MikeG7]

dhengr
Apologies for a belated reply to your post...
In this application, (picture attached) due to the circumferential stiffeners that are in the way at the top of the vessel, and "lifting plates" in the plane of the shell are not practical. As far as the idea of bending these plates so that they clear the obstruction, it would be practical to do so but I am afraid that the trunnion option has been already been incorporated but in a way that takes into account suggestion of keeping cantilever short as possible and per your suggestion to incorporate the trunnion with some backup structure to distribute the concentrated loads.

I looked into the option of incorporating a structural steel shape between the horizontal stiffeners but there is not sufficient space between them to get a significantly robust section in vertically but the suggestion is very good. Thank you for your valuable input.

 

[MikeG7]

Hi Chicopee
As it turned out this vessel was supposed to be built in sections and final welded on site. The size is very large to transport it but that was a late decision in the project so the lifting design is an afterthought. I think the "cage" would add significant cost?

 

[chicopee]

Well MikeG7, it appear that your company will have to bite the bullet and spend additional money to do what I have proposed in light of the fact that you seem to have rejected other options presented by the above responders. It is unfortunate, however, one must realize that not all projects will be profitable.