4 replies to this topic

[Mech.Engineer]

Hi Everyone....!. I would like to calculate NPSHa for a pump wich is gonna work with LPG 70% propane/ 30% Butane. Propane is taken from a sphere and in this case i would like to know all about LPG properties...how to keep it in liquid state for avoiding cavitation in pump suction (Vertical can and Horizontal pumps)..... and principally if someone here have a spreadsheet for calculating friction losses in pump`s suction pipe using LPG...I know it`s different than using water...!. IF SOMEONE CAN HELP ME, PLEASE LET ME KNOW...!.

[go-fish]

Frictional losses method should be same as water, you need to calculate the density and viscosity of the LPG at the pumping conditions by using 70-30 mix ratio of propane and butane. For this, you can either use HYSYS or any other simulation program.

Also, you need the vapor pressure of LPG. You can again use HYSYS. However, since LPG is stored in spheres, it is in equilibrium, so the vapor pressure and the vessel operating pressure are considered to be same and cancel out while calculating NPSHa. I am not sure on this part, so you should wait for other experienced members have to say.

Once, you have all the inputs, you can use the NPSHa equation.

[kkala]

I agree with go-fish: Pressure on liquid surface in sphere is same as vapor pressure at pump suction nozzle, so NPSHa calculation has to be based on LPG static head only*.
Be generous with NPSHa margin, once I applied 2 ft over manufacturer's NPSHr (liq NH3 pump) but the first two impellers of the vertical pump were often sent for repair to maintenance shop.
Calculated pressure at any point of suction line (for sphere LLLL) should not go below LPG vapor pressure, to avoid (even temporary) vaporization. This is normally valid for nearly vertical suction lines; I had made a pressure - suction length diagram to check it in the case of mentioned pump.
"Propane is taken from a sphere" is not well understood; "propane" may mean LPG (composed of 70% propane in this case).

*if inert gases (e.g. N2) are dissolved, gas pressure in sphere can be a little higher than vapor pressure at suction. Temperature a bit higher at upper volume of sphere could have similar effect too. These had been observed in liq NH3 storage, but had not been considered in the NPSHa calculation (correctly judged as occasional and non conservative).

[Mech.Engineer]

Hi Again KKLA...!. Thank you for aswering...!. I should say that it was an excellent answer for me...

I would like to ask you something:
I have an sphere from where i`m gonna take propane 100%. Initially this sphere was designed for working with the mixture of 70% propane and 30% Butane but now is gonna work with propane (100%). With the first condition of sphere (70/30) We had a problem with pumps NPSHa...this was very negative and we had to change horizontal for vertical can pumps. Now our client wants the sphere working with the second condition (100%). How do you think this change is gonna affect calculations now..?. What do i need to take in count..?. Do you have a PROPANE PROPERTIES CHART?.

I appreciate your comment o.k....!. Thank you. (Take alook on attache png).

Attached Files

•  lpg sphere&pumps.png   8.4KB   122 downloads

[kkala]

I have an sphere from where i`m gonna take propane 100%. Initially this sphere was designed for working with the mixture of 70% propane and 30% Butane but now is gonna work with propane (100%). With the first condition of sphere (70/30) we had a problem with pumps NPSHa...this was very negative and we had to change horizontal for vertical can pumps. Now our client wants the sphere working with the second condition (100%). How do you think this change is gonna affect calculations now..?. What do i need to take in count..?. Do you have a PROPANE PROPERTIES CHART?.

Similarly to what written before for LPG (70/30), propane operating pressure at liquid surface in the sphere will be same as vapor pressure at suction nozzle. So NPSHa shall be liquid static head (at LLLL)-frictional pressure drop in all the suction line (datum is suction centerline).
Attached LPG.xls gives some data, as well as references for propane properties. Given that the vertical pumps have solved the problem of NPSH for LPG (70/30), no problem is normally anticipated for propane (changes seem minor). Nevertheless hydraulic calculations should be made to confirm this and specify any changes (e.g. in pump flow rate), as explained in LPG.xls.
Design pressure of sphere (receiving now propane) can be an issue. Pumps shown in attached PID are horizontal, understood to be vertical now.

Attached Files

•  LPG.xls   19.5KB   310 downloads