Hello to all.
What do you thing about this pipe routing for pump suction.
Details:
suction size 4"
line size 8"
Tope=110°C
On suction nozzles are ecc red from 8 to 4"

One pump is working another standby. Temp up to first valve for standby is 50%Tope and after valve Tabm.

This is client request but I already try this and no chance to achieve allowable loads on suction nozzles.

Notes check by stress means minimum. I am just interested for opinion if anybody want to comment this routing especially between pumps. I made routing which works but client just say NO. My routing is on another image.

Pump is piston inline type.

Thanks for comments if somebody want to discus:)


client suggestion:
http://i56.tinypic.com/2z8yp07.jpg

my solution:
http://i52.tinypic.com/j9lb42.jpg

Hi Kemal,
With the pump suction NPSH has to be considered and that might be the reason why the client insists on simpler routing.
The issue with this routing is that the branches are too rigid and supports are too close to the nozzles. You can try the following:
1. Move the tee further away from the pumps and closer to the vessel to allow more flexibility in the branches
2. Provide only one support per branch and further away from the nozzle
3. Try to omit the directional anchor and see what happens
If none of the above works:
4. Ask pump vendor if they can supply heavier base plate and if that would help to increase allowables
5. Check if you can reduce the branch sizes right after the tee to 6” , move valves after the elbows (watch NPSH)and support under the excentric reducers.
6. Extend the line coming from the vessel to the further right from the tee. By this reroute you will not add elbows but the length will be longer so again watch your NPSH
Good luck!

P.S. Izuzetno sam prijatno iznenadena da se u SA rade ovakvi poslovi I da imamo cak I piping stress inzenjere! Drago bi mi bilo da porazgovaramo izvan foruma. Javi mi se na ranka.sofijanic@thyssenkrupp.com a mozes mi i poslati Caesar model i pump data pa cu jos pogledati

Thanks,

I tried many routing and results are better more or less but not enough to satisfy vendors requirements.

I guess there is not much elevation differance between liquid level in the vessel and the suction of the pumps. First of all they need to provide sufficient differance to obtain the required head not to cause cavitation in the pumps. It is not a good practice to leave the problem solution to stess analysis.

It is always expected that there will be small allowables on the pump nozzles.

It seems to me that the pumps need to be relocated in addition to elevation the vessel or lowering the pump levels to be able to provide less resistance to the suction piping as well as complying with the allowable nozzle load under different operating conditions.

Your client needs to provide you this kind of flexibilities. Otherwise solving the problem in stress analysis may be an endless game. Your client and you need to support each other to provide a proper solution.

Hope you will get this help from your client.

Regards,

Ibrahim Demir

yes elevation between vessel outlet nozzle and pumps suction is small. almost everything is in horizontal plane.

lowering pumps is not possible, also moving vessel up.

Thanks Mr.Ibrahim.

I start this topic not to ask people for solution. I just want to discus if somebody had similar calculation. If stress engineers share their problems and solutions it is very useful.

Please understand that all the replies come from the experience on the similar/same applications.

This type applications require very close relationship between process and stress engineer. Stress engineer finds that the allowables can not be achieved with the available route. Process engineer does not want to increase pipe length and introduce more fittings to the systems not to increase resistance. Thease are all normal for this applications. Sometimes the process engineer/vendor may accept that the pump can cavitate.

There are other things that stress engineer should consider for this applications. The suction line should be as short as possible to reduce the flow resistance and thermal expansion in the system. Keep the horizontal supports as minimum, or do not introduce unless it is essential. Watch the location of vertical supports not to cause additional bending moment at nozzles and pump flanges, put low friction vertical supports if necessary.

The horizontal support that you introduced are the source of the problems in case there is a temperature case. You added too many bends and increased the suction line length, this increased pipe and local friction losses in the pipe. Additionally, when you create new pipe route you need to watch the maintenace and removal access to the pumps.

Start analysing the original route without horizontal supports. Add them if they are necessary. In case this does not satisfy the allowables, and pipe code requirements, you consult the process engineer to plan new locations for the pumps in case you can not change the elevations. Do some stress analyses fpr your proposal before going to process engineer. Talk to the designers for other restrictions as well.

I guess this is sufficient for the moment.

Regards,

Ibrahim Demir