Good Day friend,

I have modelled wrc nozzle in caesar ii by converting the node to vessel nozzle and also defined displace only in -Y direction. This is because vessel is going to exapnd from fixed saddle end to sliding saddle eng as the tem is about 278 oF.

After that i modelled vessel as described in caesar ii manuals. The only thing i suspect is that, i modelled vessel fixed saddle end as an anchor and the sliding saddle end as free point i.e. neither enetered any displacement not any anchor.

>>> Did i modelled correct? <<<

after peoforming analysis the foces at my fixed saddle point comes out like this:

==========================================
580 Rigid ANC
1 OPE

FX FY FZ MX
0. -110605. 14430. -271000.

MY MZ
0. 0.
===========================================

can anybody tell me what is meant by these huge amount of focres/moments on fixed saddle.

how can i assure that my fixed saddle can bear such huge amount of forces.

Thanks

Perhaps I'm misunderstanding, but from your description, it sounds like you are carrying your entire vessel on the fixed saddle - you say you've not modeled displacements nor and anchor at the sliding end. Is there a support there at all?

Hi edward,

oh sorry .... my mistake... well now i put a +Z restraint (since z is my vertical axis) at sliding saddle and run again analysis..

But the rseult are as follows:

Sliding Saddle:
==============

vessel going up i mean the node has displacement of '0.0238 in' and so no forces and no moments at sliding saddle. The support is inactive. No use.

Fixed Saddle:
=============

all the forces and moments are same as:

FX = 0
FY = -110605
FZ = 14430
MX = -271000
MY = 0
MZ = 0

what to do now friend.

Thanks

I get nervous when an engineer asks "What to do now?".

Use the output to determine if your input is correct. You may have to model even more of the elements, loads and boundary conditions to get a better analysis.

I usually "ignore" loads on equipment anchors because not all loads are included. Equipment is usually included in the model to develop a better picture of what's happening to the piping, not what's going on (completely) with the equipment.

Is your piping moving properly? That's what's more inportant here.

YOur system looks planar with only FY, FZ & MX loads. It shouldn't be too difficult to get to the bottom of this. Did you call the vessel "rigid"?

AUB, I'm not quite clear on what you've done here. How have you decided what you displacement is?
If you've modelled your vessel including temperature and then added a displacement is it possible that your overconstraining something? Or that you have put the displacement at the wrong node in your model? (ie displaced the vessel node not the nozzle node)...

Just a few thoughts, I can't quite picture what you've done here... so maybe a slightly more detailed description will help.

-Shannon

Dear All,

i have got answer as said by Dave

----------------------------------------
I usually "ignore" loads on equipment anchors because not all loads are included. Equipment is usually included in the model to develop a better picture of what's happening to the piping, not what's going on (completely) with the equipment.

Is your piping moving properly? That's what's more inportant here.
------------------------------------------

so my piping loads at suction nozzles of pumps and vessel nozzles are all OK....

the last question regarding confirming modelling of pump nozzles is:

1) should i put anchors at pump nozzles for having foces and moemnet so as to evaluate api 610 compatability.

or

2) should i enter displcements and no anchors. if so how much nozzle displacements should i enter. Any thumb rule or any formula to calculate this.

Thanks all.

AUB,

Regarding whether to model pump nozzles as anchors or displacements: I think the best advice I can give is that you should try both methods and compare the results. Comparisons of alternative modeling approaches is the best way to gain the experience that will allow you to make appropriate judgements on your own.

That said, I usually model pump nozzles as anchors since the amount of displacement is usually so small it makes little significant difference to the results. However, if the results indicate that the pump is close to maximum allowables, I might remodel the nozzles as displacements to see what happens. (A far better approach in this situation though, is to see if a redesign of the piping can improve the situation.)

I should add that calculating displacements at pump nozzles is not entirely straightforward because, unlike most vessels, it's not always clear at which point the pump expands from.

Gringo,

Thanks for commenting. Yes i also use normally by placing anchors at pump nozzles.

But i can not check the sutuation by entering pump nozzle displacements.

Thea reason not have any idea how to calculate pump nozzle diplacements.

Any help will be highly appreciatable.

Thanks

Simple: Just apply the same principles you use for vessel nozzle displacements.

(1) Determine the origin point of expansion (for example, on a front suction-top discharge pump, the centerline of the discharge nozzle is probably the appropriate point).

(2) Determine the distance from the origin point to the nozzle face.

(3) Determine the coefficient of expansion for the pump casing material at the appropriate temperature.

(4) Multiply the distance by the coefficient and the delta-T.

Dear AUB,

Please don't misunderstand me. It is risky for your reputation to show ignorence in public. As an engineer from our subcontinent I expect you to prepare a bit more from references, mentors, CAESAR II training classes, if possible. Even by surfing the discussion forum & CAESAR II manuals & Coade Mech Engg News we have got our so many problems solved before raising a query. Your customers depend on you for safety; so, be confident.

regards,
sam

Hi gringo,

thanks for guiding me...

and sam you not know what is situatiosn here so no need to comment like this....

and yes i accept my mistake for not searching the coade forum for this topic... may be it is discussed earlier.. i am sorry for that.

Thanks

All right, this is just scary. You say that the sliding saddle is lifting about 1/32" and therefore inactive. Further, you show an upward load of 14 kips on the fixed saddle, so niether of your saddle supports is carrying any load. Now, in reality, this is just impossible unless your pipe is really large relative to your vessel. Your pipe should not be able to bench press your vessel.

I'm guessing that the vessel was modeled with weightless rigids so that the load on the saddles is not considering the effect of the vessel weight. Dave Diehl is right that we don't design equipment anchors from a Caesar model as all the loads aren't considered. Still, there needs to be enough there so that the supports behave as you would expect them to.

Now, that your piping system is unloading both of your saddle supports suggests to me that the pipe routing is very very very tight. I'm curious to know what kind of loads you are getting on your pump and vessel nozzles. I expect they are very high.

Hi Bajwa,

Just happened to read this thread a while ago and have not read all responses thoroughly and so just in case nobody has reminded you yet on sliding restraint... and as you said you have changed it to +Z . It is actually a +Z (vertical) with zero gap - meaning your vessel at that point can't go up because of zero clearance, and since it is normally a slotted hole right there, you need to put a guide there too with gap (say Y or X as applicable on your situation) Field installer need to bolt the vessel slightly tight enough to let it slide due to thermal expansion. A lock nut is I think a good idea to lock the main nut from loosening.

Jus a reminder

Hi freinds,

Thanks for your comments.

Lamigo, thanks for remoinder

Edward you are abdolutely right. My pump nozzle allowable forces are going above than api 610 limitations (say 60 %). i am just searching api 610 if it allows me how much i go beyond the allowable loads. But i could not find anything like this. Do you have any idea.

Also my suction pipe consist of 3 elbows. if i do piping with 4 elbows that reduces my allowable loads at pump a lot. But problem more pressure losses. So the client restricted me to to 3 elbows and not allowing go beyond this.

what is idea if i go for 4 elbows. is it recommened. any tumb rule how much elbows one should have in its suction piping.

Also note: two pumps doing suction using two different pipe from same vessel. The stress analysis is being done at an opertaing temperature of 278.3 oF. If go for design tenperature of 400 oF, loads are huge; can not control them with 3 elbows at least.

Thanks

Best Regards

Bajwa,

Some quick thoughts/comments/questions on your pump situation:

1. API allowables cannot be exceeded.
2. Are you including both suction and discharge nozzle loads in your API calc? I don’t have my API code with me, but there is a factor (Table 4 allowable, I think) which sometimes helps if one nozzle load is high and the other low.
3. There is no rule-of-thumb to the number of elbows in a suction system. If pressure drop is the concern, can you increase the available head by raising the vessel? Or, can you increase the diameter of the pipe? (Increasing the diameter will of course increase the loads, but it may allow you to use more elbows.)
4. With two pumps connected to a common suction/discharge line it is important to examine three operating situations, i.e.
a. ‘A’ pump hot, ‘B’ cold.
b. ‘A’ pump cold, ‘B’ hot.
c. Both pumps hot.
d. Look at the three above cases at both operating and design temperatures.
5. Make certain your design temperature is realistic. Process engineers often do a poor job of this. If the pump will never see the 400 degrees you stated, it makes no sense to analyze it.
6. Have you tried springs under the suction nozzle elbow? If springs help, it’s important to size them carefully considering the operating cases noted above.
7. Has the pump been purchased? If not, can the vendor provide a model with greater than API allowables? (Usually unlikely, but worth a try.) Or, can the pump nozzle size be increased by using a larger casing with a small impeller?

Bajwa,

You can use Appendix-F of API-610 by combining both suction & discharge system together.

This will help you in qualifying the pump nozzle loads.

Regards

Hi Bajwa,
The pump suppliers often allow 2 x API610 loads just by simply providing heavy duty base plate.

Carefully check you model. As Edward said it does not make sense that your piping lifts up the vessel.

Modeling vessel:
- model fixed point as an anchor, no displacements
- model sliding point as Z (it is held down), X and Y with gaps as appropriate, as well as RX, RY, RZ because rotation is also restricted. Displacements will be calculated by Caesar.
- model vessel with rigid, no-weight elements up to the nozzle / vessel intersection at appropriate temperatures.
- model nozzle with ANC and CNODE, or as WRC297 nozzle. Displacements will be calculated by Caesar.

Modeling Pump:
- Place anchor at the baseplate / pump centerline intersection.
- Model rigid elements up to the nozzle face at appropriate temperatures
- Connect with ANC and CNODE to piping. Displacements will be calculated by Caesar.

Good luck!

Dear Friends,

Thanks all for your comments. I now clear my model if it have any misatkes and then finally clear allowable nozzle loads...

Thanks again.

Best Regards.