Rotational Restraint

Posted by: krish

Rotational Restraint - 01/24/06 06:52 AM

Sir,

I cannt find more about rotational restraint in the caesar technical reference, so please send the complete information and help me.

thanking you
Posted by: Richard Ay

Re: Rotational Restraint - 01/24/06 08:59 AM

What information are you looking for?
Posted by: krish

Re: Rotational Restraint - 01/24/06 09:47 PM

About rotational restraint ie about RX, RY, RZ
Posted by: krish

Re: Rotational Restraint - 01/25/06 06:58 AM

When i tried to reduce the moment at the nozzle load i used RX, RY, RZ it reduces the load & now i want to know, is it possible to restrict only the moments without forces & i also need where i get more information about ROTATIONAL RESTRAINT.
Posted by: John C. Luf

Re: Rotational Restraint - 01/25/06 07:09 AM

IMHO This is a consulting question. How to do something is up to the designer not the software manufacturer.

Coade is not being paid by your client to do analysis and design of work in your contract.

As to the practicality of the question it depends on how clever you are.
Posted by: foglamp

Re: Rotational Restraint - 01/25/06 07:19 AM

Find out what linear movement causes the rotation and try limiting that linear movement using conventional restraints.

Hope this helps.
Posted by: NozzleTwister

Re: Rotational Restraint - 01/25/06 12:55 PM

I may be getting off track a little bit but this thread reminds about a stress audit some of my colleagues and I were involved with some years ago.

We reviewed some stress calcs done in an overseas office. We found in several calcs, strategically placed rotational restraints just before equipment nozzles. We couldn't find these restraints on the drawings or on the stress comments, only in the computer analysis. We quickly dubbed them as MOMENT FILTERS, they worked wonders for nozzle load compliance.

Now, back to REAL WORLD moment restraints. They can be designed, but you're limited to a location where you can actually install them and restrain the rotation for real. It's controversial on how much rotation they will actually restrain but if designed properly, they'll help. I can't think of a design right off that won't restrict any forces. I think you'll have to restrain at least one linear direction.

Moment restraints should be the last resort. Limiting the linear movement causing the rotation by using conventional restraints is preferred as foglamp pointed out.
Posted by: Edward Klein

Re: Rotational Restraint - 01/25/06 02:19 PM

Moment Filters - I love it, Kevin.

Man, I could use some of those on a turbine problem I'm working on. Wonder if PTP has any on the shelf.

laugh laugh laugh
Posted by: Andrew Weighell

Re: Rotational Restraint - 01/25/06 02:28 PM

Golly gosh, and I thought the only method these guys (by/checker/approvers) had to reduce loads was "forgetting" to put an anchor at the nozzle and show a large amounts of surprise if found out. As Sam said in recent posts, you get what you pay for.

Otherwise, I agree with Foglamp and Kevin.
Posted by: John C. Luf

Re: Rotational Restraint - 01/25/06 03:02 PM

More rework for somebody else... keep your calendars open!
Posted by: SUPERPIPER

Re: Rotational Restraint - 01/26/06 03:48 AM

Krish.

By observation
Rotational restraints usually require very little
actual movement to reduce these moments, the result of which, they are almost impossible to design, manufacture or install to any great effect.

I'd stay away from them and re-route the pipework to suit.
Or ask someone with alot of stress experiance for
Advice.
Posted by: krish

Re: Rotational Restraint - 01/27/06 06:24 AM

I have lots of confusion, its correct practice to give rotational moment in front of any nozzles.
Posted by: John C. Luf

Re: Rotational Restraint - 01/27/06 06:41 AM

I don't agree with this statement "I have lots of confusion, its correct practice to give rotational moment in front of any nozzles."

Maybe if you practice snake charming and slight of hand magic but designing and building a true rotational restraint is expensive and difficult.

I suspect your model may be more slight of hand then reality.
Posted by: A.M.

Re: Rotational Restraint - 01/30/06 02:38 AM

Hi Arun,

Providing a Rotational restraint to reduce the moment is not a correct practice. So it is better to look for routing change or someotherway to qulify the nozzle laod.
Posted by: sam

Re: Rotational Restraint - 01/30/06 04:55 AM

Luf Sir's comment is apt.

How can one with 'lots of confusion' tell his 'practice to give rotational moment in front of any nozzles' as correct?

'When i tried to reduce the moment at the nozzle load i used RX, RY, RZ it reduces the load & now i want to know, is it possible to restrict only the moments without forces & i also need where i get more information about ROTATIONAL RESTRAINT.' - Arun Kumar's this entry shows he needs some formal training!

regards,
sam
Posted by: SUPERPIPER

Re: Rotational Restraint - 01/30/06 08:21 AM

i suspect language could be the problem here.
Posted by: John C. Luf

Re: Rotational Restraint - 01/30/06 08:50 AM

Maybe so but my German according to my relatives is even worse than my U.S. based English!


A rose or a rotational restraint are the same in any language....
Posted by: John Breen

Re: Rotational Restraint - 01/30/06 09:15 AM

Hi Arun,

If you specify either (and) X, Y or Z restraints the appropriate (as specified) translational (linear) movement of the pipe will be stopped (it will be made zero in the model) and the applied force(s) (reaction) that is (are) necessary to stop the movement will be calculated by C2 and the force will appear in the "restraint" report. Similarly, if you specify either (and) RX, RY or RZ restraints the appropriate (as specified) rotational movement of the pipe will be stopped (it will be made zero in the model) and the applied moment (reaction) that is necessary to stop the rotation will be calculated by C2 and the force will appear in the "restraint" report.

What your colleagues here are telling you is that it is very difficult to design a "real world" restraint that will absolutely restrain (stop) a moment (RX, RY or RZ). More commonly, we will locate a PAIR of translational restraints (X, Y or Z), such that the moment that we wish to diminish (control) will be broken-up into a force couple of opposite signs. It is much easier to design restraints against forces than it is to design restraints against moments.

Arun, we piping engineers often see cases where an uninformed "analyst" will "solve" a problem of excess loadings on a piece of equipment by placing a rotational restraint (RX, RY or RZ) in his/her analysis model. The resulting calculated numbers will look quite "nice", but the truth is it is usually the case that it is impossible to actually fabricate the restraint. Remember, we typically fabricate restraints from structural steel shapes and plates that are in fact elastic (i.e., they will deflect under loading). They will NOT result in the moments being completely restrained (stopped). Sometimes problems are "solved" on paper (in the computer) but the resulting design does not protect the equipment that the piping is attached to.

Look at the list of references provided on the COADE web site and try to obtain one or more of the books that address structural design and analysis of piping systems. Also, ask your colleagues for help and/or review of your restraint designs.

Regards, John.
Posted by: Dave Diehl

Re: Rotational Restraint - 01/31/06 08:28 AM

Several years ago a Mr. Bob Robleto (KBR) presented a paper at the ASME PVP conference on the design, construction and installation of a moment support. I asked Bob about it again yesterday and here's what he said:


The paper is titled “Moment Restraint Support For 84 Inch Conference Compressor Suction Line”, Pressure Vessel Piping Conference, Orlando Florida, 1997

We built one for an 80 inch line in an ethylene plant and it worked fine. The tolereances were important and we had a man go down and inspect the installation. The design comes from A. Kieth Escoe in his book Mechanical Design of Process Systems. On pages 57 and 58 of Volume 1, pictures and an application of an MRS, (Moment Restraint Support) are shown. It constrains the moments but not axial motion.
Posted by: SUPERPIPER

Re: Rotational Restraint - 01/31/06 09:43 AM

Horses for courses
Scale that monster down to 8" and it bet i don't work !!!!!
Posted by: Andrew Weighell

Re: Rotational Restraint - 01/31/06 01:24 PM

I am not familiar with Mech Des of Process Systems. I am guessing that the MRS is some form of long guide. If the effective points of action for lateral loads are a large distance apart, the lateral force and hence stiction / friction will be low and vice versa. I'd be more interested in the vice versa.
Even with very small clearances, there will be a finite rotation and give before load is taken up.

My money is on Superpiper's horse at least until I've seen the MRS on the paddock.
Posted by: anindya stress

Re: Rotational Restraint - 01/31/06 08:51 PM

I believe that if we put an axial stop or a guide with zero gap, it will be a rotational restraint also. But that is purely hypothetical.Support with zero gap is not possible ( unless somebody welds it which then becomes a six DOF restraint).

So in my opinion rotational restraints are not possible in real world. Also the amount of rotation that a piping node is usually subjected to, say, in the range 0.1 degree ( just an example). Preventing that : wow !

Regards
Posted by: sam

Re: Rotational Restraint - 02/01/06 09:47 PM

I can recollect one main steam piping analysis for a captive power plant in Jamshedpur,India where just near the steam turbine the analyst used a gap 3-D(XYZ) restraint. When I objected about the constructability & sensitivity of the gaps in NEMA SM-24 conformance result of steam turbine, I was silenced in the name of value addition of chopping up a pair of P22 elbows from an alternative flexible layout without such 3-D(XYZ) restraint.

So, it is easy to know the answer, but selling that is not so easy in a lowest bid EPC market.

regards,
sam
Posted by: first2buy

Re: Rotational Restraint - 02/02/06 11:38 AM

Going back to Escoe's book, he describes a 10 inch MRS, I supersized it to 84 inch. A simple example of an MRS mechanism is a base support with all bolts slotted. Free to move axially but rotation in Y is restrained. More complex mechanisms can be designed to allow freedom in all directions and moments restrained in all directions.
Posted by: Andrew Weighell

Re: Rotational Restraint - 02/02/06 03:29 PM

What were the shear loads on the bolts? How much rotation were you trying to restrain? i.e. What was the rotation before fitting the support? How does this compare with the clearance rotation of the actual support?
Posted by: first2buy

Re: Rotational Restraint - 02/03/06 04:08 PM

Andrew

The bolted mechanism is just an example. the 84 inch consisted of a clamp around the vertical pipe which was attached to box beams. The boxed beams allowed for tight clearances. The actual clearances, friction and flexibility of the structure were input into the caesar model.

I will attempt to describe it this way. Draw 2 rectangle on on top of the other. This is the plan view. The pipe runs vertical in the middle. A clamp with two box beams is welded to the clamp. The pipe is allowed to move vertically through the clamp. At the ends of these box beam is a box beam allowing a box beam to go through top to bottom in your plan view. The beam is allowed to slide.

The top and bottom are fixed. This allowed translation horizontally, top to bottom but restricted rotation. The other horizontal motion was also restricted. It was a welded construction. I do not recall the loads but it was effective enough to reduce the loads at the compressor.This was built 9 years ago and have heard no complaints about the operation.

Regards
Posted by: John C. Luf

Re: Rotational Restraint - 02/03/06 06:07 PM

I did not intend to say that rotational restraints are "impossible" to design and construct.

But in my experience they are not a normal support placed in front of equipment to make moments drop to low values.

I have at times past designed and detailed these but only as an absolutely last alternative.

My opinion stands they are not "normal practice perse.
Posted by: krish

Re: Rotational Restraint - 02/06/06 09:45 PM

Thank you all for your replies
Posted by: krish

Re: Rotational Restraint - 02/07/06 06:25 AM

If there is any other way to reduce moments in the turbine nozzle without providing rotational restrient.
Posted by: John C. Luf

Re: Rotational Restraint - 02/07/06 06:51 AM

Change the geometry of the piping layout to add flexibility to the system or add linear restraints to counteract the linear component of your moment.
Posted by: SUPERPIPER

Re: Rotational Restraint - 02/07/06 08:13 AM

Move the turbine........ laugh
Posted by: Richard Yee

Re: Rotational Restraint - 02/07/06 11:16 AM

Krish,

There are two design principles that would help to control moment loads on a turbine nozzle, if there is sufficient space for flexible piping and the freedom to revise the piping geometry.
First, the piping supports are to be located in plan location so that the centroid, or center of gravity of the vertical support loads, is near the nozzle location. If say there are 3 supports in a triangle pattern, and the nozzle is located within the triangle, then the summation of the moment loads from the three supports will be much less, than if the 3 supports were located such that the turbine nozzle was located outside of the triangle of 3 supports. Visualize it as a table with 3 legs, the nozzle being the table and the supports being the legs. It would also help to locate supports near weights of valves, etc.
Second, the piping restraints for horizontal thermal growth could be located at neutral planes of thermal growth to match the turbine nozzle thermal movements. The neutral planes normally would be at the turbine shaft and at the anchored end of turbine casing (discharge end). If the piping is routed to cross both of the neutral planes, and a line stop restraint located at each of the neutral planes, then there will be restraint against thermal forces in those two directions. The remaining vertical thermal growth would be accomodated with spring hangers on piping with adequate flexibility. The best spring hanger locations would also apply the center of gravity principle to minimize moment loads resulting from the spring hangers.
If the piping geometry has already been fixed, or if there are limited options of revising the piping geometry, then the thermal deflections must be studied to decide where to add restraints with gaps to control the deflections, and hence the loads and moments on the turbine nozzle. The forces will probably need to be reduced to much lower than the force allowables in order to achieve the moment allowables.
Posted by: John C. Luf

Re: Rotational Restraint - 02/10/06 09:33 AM

Mr. Yee an excellent post....