Hello,

I have 3 pumps with the following loads cases. Am I correct with the expansion stress cases L17 to L37 per B31.3 ? Please help.

L1 - HYD
L2 - OPE - W+P1+T1 - pump A running, pumps B & C idle
L3 - OPE - W+P1+T2 - pump B running, pumps A & C idle
L4 - OPE - W+P1+T3 - pump C running, pumps A & B idle
L5 - OPE - W+P1+T4 - pumps A & B running, pump C idle
L6 - OPE - W+P1+T5 - pumps A & C running, pump B idle
L7 - OPE - W+P1+T6 - pumps B & C running, pump A idle
L8 - OPE - W+P1+T7 - pumps A & B & C running
L9 - SUS - W+P1
L10- EXP - L2-L9
L11- EXP - L3-L9
L12- EXP - L4-L9
L13- EXP - L5-L9
L14- EXP - L6-L9
L15- EXP - L7-L9
L16- EXP - L8-L9
L17- EXP - L2-L3
L18- EXP - L2-L4
L19- EXP - L2-L5
L20- EXP - L2-L6
L21- EXP - L2-L7
L22- EXP - L2-L8
L23- EXP - L3-L4
L24- EXP - L3-L5
L25- EXP - L3-L6
L26- EXP - L3-L7
L27- EXP - L3-L8
L28- EXP - L4-L5
L29- EXP - L4-L6
L30- EXP - L4-L7
L31- EXP - L4-L8
L32- EXP - L5-L6
L33- EXP - L5-L7
L34- EXP - L5-L8
L35- EXP - L6-L7
L36- EXP - L6-L8
L37- EXP - L7-L8

Thanks,

I realy dont know what your trying to do.
My opinion to your work will be...
Try evaluating your pump case in operating(As indicated in the code or contract), instalation and ambient. Think of things that are posible to happen... Usually you can reduce your case in visual judgement,like if higher temperature help to mess your pump or low temperature will do.

Study this one. This will realy help you...

Regards! Mabuhay!

B31.3 requires that you only evaluate the maximum computed stress range. You need only look at one of those load cases L10 through L37. But the high stress range may change from node to node so you'll have to pick and choose. You can set up another load case - L38:L10,L11,L12,...L37 and set combination to max to throw all the max. stress ranges for each node into one report.
But... B31.3 para. 302.3.5(d) says that when stress range varies, you evaluate the max. stress range using an allowable limit with an adjusted N. The number of cycles (N) used to set f is a function of all the other stress ranges. It's a cummulative damage thing. Quite the hassle.
If several of your load ranges produce high stresses, and if you have high cycles for several you might want to use your approach.
I think, in general, your approach here is uncommon.

L10 to L16 I think it would be enough and all the other I think it is questionable... Maybe you have something in thoughts you can share? Because it maybe easier for our senior to understand if you have a justification. For ex. L17 is not reasonable because it is just like two situation occur but stress/force..etc. are being subtracted.(I realy dont know how to say it...but all I know is,"It is wrong".)

Please send me a PM with your opinion cause I am realy excited to know your justifications..

Salamat...

Regards!

Thanks Mr. Dave and Mr. Bom.
I agree that load cases L17 to L37 are not common. But, I just want to know if it is correct or not according to the code B31.3.
L17 to L37 are for checking expansion stress ranges only.

I think with a proper supporting and flexible route you will not have any problems with "stresses" actually your main problem is allowable loads on the pumps nozzles by evaluating L2~L9.

Thanks Jozm,

But my concern is the correctness of the load cases to check the stress ranges between two different operating cases. It is true that pump nozzle loads usually govern the design, but the stress ranges must be checked according to the code.

Hi Julius,

You could only have an expansion stress during

ex.

L1=W+P1+T1(OPE)
L2=W+P1(SUS)
L3=T1(EXP)

L1-L2(exp)non-linear
L3(exp)linear

That's why perhaps in sticky topic you will find people asking "why there is no allowable stress evaluation for operating case". As I understand in the code, the whole operating system is to be evaluated in two type of stress/failure, that is sustain and expansion. Those algebraic forms explain why and when you will have an expansion and sustain stress.

Regards!

Be careful here. As soon as you have multiple temperatures you have the potential to cycle between them, therefore you have additional expansion cases to check. Check this earlier post on this same subject: http://www.coade.com/ubbthreads/ubbthrea...=true#Post10440

Sir Richard,

For that case it is applicable. But I think for this case, the stress range is not in both operating condition but between its operating and ambient/instalation as I understand the post. That there is no stress crossing the zero line.
Also im not comfortable with OPE-OPE=EXP(why EXP not OPE? I believe that its a software approach), Not just I didnt find it in the tech. ref. but also in the case of pressure change.
Maybe you can reffer me to which it is stated?

Thank you.

Regards!

Mr. Richard,
Thanks, although my question is not answered directly.

Mr. Bom,
I do not have any problem with the stress range from operating temp to ambient.
My problem is the stress range between two different thermal cases, which I feel is not commonly evaluated, but should be evaluated to comply with the code. I just want to verify If the load cases are correct.

Thanks,

Julius,

The stress range I mention was for the link Sir Richard has given. My point is, if your not crossing the zero stress(that means only expansion no contraction), then those L17 to L37 has no meaning.

Regards!

Hi Julius

In my openion, when you have three pumps(pl refer the relevant P&ID),among three pumps, either of two will be working and one will be standby.From that you can arrive the combination of load cases and the load case would be

L1 - HYD
L2 - OPE - W+P1+T1 - pump A/B running, C idle
L3 - OPE - W+P1+T2 - pump B/C running, A idle
L4 - OPE - W+P1+T3 - pump A/C running, B idle
L5 - OPE - W+P1+T4 - pumps A & B & C running
L6 - SUS - W+P1
L7 - EXP -L2-L6 checking stress range when Pump C is idle
L8 - EXP -L3-L6 checking stress range when Pump A is idle
L9 - EXP -L4-L6 checking stress range when Pump B is idle
L10- EXP -L5-L6 checking stress range when All Pumps are working.

Most of cases, when there are three pumps, "normally two will be operating one will be standby"

This is the typical combination of load cases.If some one have any comments on this, most welcome

regards

Sha...

Bom,

The (EXP) Expansion case is a "stress range". You compute a range by subtracting. The Piping Codes require that you check/verify the "extreme" displacement stress range.

In a situation where you have a single operating temperature, then the range is simply "OPE - SUS". However, there are other more complex situations, for instance:

Two Temperatures on either side of Ambient: In this instance, there are three ranges that must be checked (a) T1-ambient, (b) T2-ambient, and (c) T1-T2. In all probability, case (c) will be the "extreme" range that governs here.

Multiple Operating Pump Scenarios: In this instance, you have part of the system hot, part cold, and depending on the pump on/off status, these parts can change. As an exmple, take pumps A, B, C, where pump A is off. Now whem pump A is turned on, and pump B is turned off, you're probably not going to shut down the whole system. So your "range" isn't (T1-ambient), but rather (the Ta condition minus the Tb condition). If this was setup such that T1 is A off, T2 is B off, and T3 is C off, then you need to evaluate the following Expansion cases:

T1 - ambient
T2 - ambient
T3 - ambient
T1 - T2 (turning A on and B off)
T1 - T3 (turning A on and C off)
T2 - T3 (turning B on and C off)

All six cases need to be checked.

Richard,

Thanks, I agree with what you have stated.
On my example, I have considered seven operating cases, while in your example you have only three, but the idea is the same.

Sha,

I think you have to add more cases to comply with the code. Don't neglect the other stress ranges, you need to check it.

Thanks,

Ive just analyzed 3 pumps yesterday and I evaluate stress by its full stress range

L1 - HYD
L2 - OPE - W+P1+T1 - Maximum Design Temperature
L3 - OPE - W+P1+T2 - pump A running, pumps B & C idle (Operating Temp)
L4 - OPE - W+P1+T3 - pump B running, pumps A & C idle (Operating Temp)
L5 - OPE - W+P1+T4 - pump C running, pumps A & B idle (Operating Temp)
L6 - OPE - W+P1+T5 - pumps A & B running, pump C idle (Operating Temp)
L7 - OPE - W+P1+T6 - pumps A & C running, pump B idle (Operating Temp)
L8 - OPE - W+P1+T7 - pumps B & C running, pump A idle (Operating Temp)
L9 - OPE - W+P1+T8 - pumps A & B & C running (Operating Temp)
L10 - OPE - W+P1+T9 - Minimum Design Temp
L11 - SUS - W+P1
L12 - EXP - L2-L10

Please please correct me if I was wrong

Sir Richard,

I understand what you mean.. I was only pointing to limited case.
Engineer's have there own approach.. My limited understanding in this software is that, I have to solve for both -stress(contraction) +stress(expansion) then add both range. And I know this approach is quite long than your approach.

Thank you for your clarification, For without it I maybe mislead by that topic.

Regards!

Take a look at Example 3 in B31.3 Appendix S. This example decsribes the stress range calculation for a system switching the hot leg / cold leg operation. The bottom line - the range between two operating states may exceed the range from installed to operating. The Code wants the greatest range. Recommended cases from CAESAR II do not include ranges between operating states - you have to type those in.

Just as a side note, at least one other competing stress program, Autopipe, does this maximum stress range check automatically when you have multiple temperature cases and when you select the "max. range case" option. Maybe something to consider for next issue of Caesar, makes life a lot easier.

Hi Sir Dave,

I read the appendix S as you've said. I still have this doubt. I hope Sir Dave you could futher explain this subject matter and perhaps reffer to any book which can make clarification.

Thank you.

Regards!