respected all forum memeber

this is the question of pump suction line-eccentric reducer.
after caesar 4.4 version caesar has facility to model reducer
here i want to know if we model a eccentric reducer in pump suction line(top discharge-end suction),we found that at pump nozzle there will be a minor displacement(0.0002")vertical upward direction.as far as vendor allowable is concern we are not able to rich that allowable loads given by vendor(specially fy),due to that vertical eccentricity.
so my question regarding to that shall i model concetric reducer instead of eccentric in that case?

thanks in advance
bhavesh.shukla

Dear Shukla

By seeing that the vertical displacement is mentioned as 0.0002"(0.005mm).it seems to be very small.However, modelled as ecc reducer is correct only.If you want to reduce the displacement,you can try to

1.Shift the first support slightly or
2.modify the piping routing (ie upstream to the first support of suction nozzle.)to make the first support to take the load properly.Hope that it makes clear to you.If any comments, welcome.

ohh thanks

but sha still i am not satisfied,this kind of minor displacement is possible in practical case?,cant we ignore it?
thanks for reply again

In my view, if you are going to consider such small displacements, you should be designing a ladies watch, not a rough, tough piece of refinery equipment.

How about including the spring stiffness of the world around your pump and support, steel components etc ? That would lose your 'displacement'.

Bhavesh,

Having this small displacement is possible and acceptable provided you should check the forces and moments of calculated value with the allowable(API 610). If it is within that allowable, ok.Otherwise you have to go for one of the suggestion listed in my last reply.You can even go for a variable spring hanger,but think about the necessity and keep this as last chance(if it fails with API 610 allowables)

regards

Sha..

As far as I have understood there is a difference in geometrical interpretation between an 'element defined with length and slope', and 'offset'.

Have you considered the thermal growth of the pump casing? That is probably a more significant deflection than the contribution of eccentric reducer.

Regards, John

Hi John,

Normally , modelling of pump in caesar is such a way that starting from pump anchor point to the nozzle point(for suction and discharge nozzle).

I hope that this is nothing but thermal growth of pump caasing with respect to the anchor point. Am I correct?

regards

Sha..

It is not normal(unless rigid)...

It is safe to stop modeling at nozzle face.. Your vendor allowable are base in casing distorsion, I guess you dont wanna put some flexibility on your pump nozzle/casing. otherwise you dont need your vendor data.

And I think John's point is that the casing thermal growth are more to consider than a ecc. reducer.

Dear Mr Bom

I agree that vendor allowable are based on casing distortion.I also follow the nozzle point is the anchor point(with cnode)with respect to Pump anchor point.Modelling of pump is rigid only.

Hence,there is a thermal growth from the pump anchor point to Nozzle point(anchor with cnode)where the vendor allowable value is checked with calculated value.IF any comments, please welcome

thanks for the reply

regards

Sha.

Bhavesh/Sha,

For properly understanding analysis of pump lines, Pl. read the WRC Bulletin by Vincent Carucci "Guidelines to design and Installation of Pump Piping systems".

Regards

ya anindya

can u please forward soft copy of this wrc bulletin that u have mentioned already?
i know u are more knowledgable than me but if we model cocentric reducer instead of eccentric one what should be wrong matter?
is it related to geametrical diffrence?,and due to that eccentricity we get vertical displacement at that anchor point?
i know we can reduce that effect by shifting support .
please help me
because i am already confused betn this two.
thanks
bhavesh

Bhavesh

the diaplacement you mentioned not becoz of eccentric reducer's eccentricity. this is due to nearest support ( rest) in suction lines. if you want to reduce the same , you can adjust the support in suction lines in line with reaching the API-610 allowbales.

please model ecc reducer in sution line always. yes i agree that it's very difficult to maintain pump allowable. but pipe routing may help u in this regards.

if nay doubt , let me know

thanks

Hi Bhavesh,

Please always put Eccentric reducer in suction pump, coz if you do put a Concentric reducer it will cause a pocket, cavitation that will lead to more Vibration problem. Try to Modification the Type and location of the support; the first support after the pump and others around it, it works for me when had a same case like you, and Spring Hanger always be the last choice.I think you can read this WRC 449 Bulletin that Sir Anindya mentioned above
Thank you.



Best Regards

Tengku Syahdilan

Hello,

If I may add to Dylan's comment, FLAT SIDE of the eccentric reducer should be facing upward.


Best regards,