I am working for a client who is specialised in design od sulphuric acid plants. As per their design standards, for strong acid pump suction lines, the first elbow should be anchored(Bolting to a structure taken from the foundation). It is basically done to ensure that the pump is absolutely frre from the piping loads. But is not mentioned that how the expansion of the pipelength between the anchor and pump nozzle is taken care. It will case tremendous loads on the anchor and pump flange because of lack of flexibility. Since it is an acid line, use of expn joint is prohobited. Can anyone suggest me any solution for this problem?

a really bad idea!!!

You could use non expandium for the straight spool which is an alloy of non obtanium!!!

This is usally the solution proposed by ignorant people when I say ignorant I am talking of people who are ignorant and overlook the simple physics... metal or indeed most materials expand or contract on the basis of temperature...

So I suggest handing a printout of this reply to your client ignorance can be overcome by education and thinking!

Sometimes the pump is not anchored,so we can fix the pump suction pipe close the nozzle.Check the client's standard,whether the pump is fixed or not.

Arun,

If the client requires an anchor directly at the pump suction, then one might implement the design without a pump loads analysis, justified by complying with the client requirement and insistence. The ordinary loads report would just frighten people from results of using default 10E12 stiffness for restraint instead of a more realistic stiffess.

The type of pump may have different restraints for the piping. Also the 'anchor' or other type of support may need to have realistic stiffness assigned for obtaining real pump and restraint loads.

Many large acid pumps are submerged vertical pumps, so that there is not suction piping of concern. The LEWIS Pump Company installation manual recommends their pump discharge piping be run with 3 elbows in two planes with an "anchor" at the third elbow. The 3 elbows provide a small amount of flexibilty. The "anchor" is intended to restrain piping movements and loads. This "pump anchor" is a large triangular plate that is very stiff in vertical and axial directions of the piping. In the lateral direction the "anchor" is more similar to a 'wobble' plate on the bearing end of horizontal end suction pumps.

LEWIS Pump has an external mounted Form'E' acid pump that does have piping connected to the suction of the vertical pump. Their recommended installation is to GUIDE the suction piping at the vertical inlet pipe just below the pump. Their concern is any lateral loads that would deflect the column and shaft. The guide would allow the column to thermally grow down - some pump columns are more than 100 inches long. The suction piping typically has spring supports for piping weight on the external mount vertical pump suction connection.

Horizontal end suction pumps would have other restraint designs. If the pump suction anchor is designed as a 'wobble' plate then it could be considered very stiff in vertical and lateral directions with some flexibililty in axial direction. A line stop nearby could provide the axial restraint.

non expandium see.... http://www.google.com/search?hl=en&q=non+expandium&btnG=Google+Search

Arun,

If the project specs really calls for an anchor at that specified elbow (suction line), then you could model the pump as sliding type, that is you may need to model the pump itself from the base to centerline and of course you need to put the reaction in the base as +Y with 0 gap and lateral gaps too say 1/8" - meaning, the pump can't go up but can slide horizontally to a maximum 1/8". You may need to specify that the pump's anchor bolt should be slotted or bigger than the bolt to allow horizontal movement. With this, you don't have a chance to anchor the discharge side without expansion joint or flex connection. I think this could be done but don't forget that you need to evaluate still your nozzle load on both suction and discharge.

Dear Arun,

Are you sure that there is no metallic expansion joint in the pump suction piping before anchor ?

You have to check whether expansion joint can be used for such service at all !

regards,

sam

Sometimes designers of such systems will install special spring supported pump and driver frames that will allow a CERTAIN AMOUNT of flex. The analyst will have to model the frame with fairly rigid elements (get the temperature right) and check to see what the deflection is at the springs (compare this to the OEM's limits). I have worked with these things at many refineries but it has been hard to get good spring data from the spring frame OEM.

The thing is that the pump will STILL see a fair amout of loading (depending on how flexible the springs are) and flex analysis is not really an exact science. Good luck.

Regards, John.