4 replies to this topic

[drdave]

Hello Friends,

My query is how to work out total differential head for the centrifugal pump shown below with data measurement available. Suction & discharge valves are full open.

Attached Files

•  Slide1.JPG   32.24KB   93 downloads

[djack77494]

drdave,
I'm not sure what it is you wish to accomplish, but I suggest that you seek to determine the differential pressure developed by the pump. The differential pressure will vary depending on the volumetric flowrate through the pump. You have set up this question in such a way as to make that relatively easy to determine. Start with the suction side of the pump, and simply add the static pressure (atm.) to the static head (1500 mm W.C.). Don't forget to adjust the static head for the fluid specific gravity. Express both in common units such as kg/cm2. This is the pressure at the pump's suction flange (centerline). For better accuracy, you should reduce this value by a calculated hydraulic pressure loss.

The discharge pressure is available from the gauge on the discharge, so the pump's differential pressure is the difference (Discharge - Suction), expressed in kg/cm2. This can be converted to total differential head by dividing by the appropriate conversion factor, corrected for the fluid's specific gravity. In Imperial units, we'd normally calculate the differential pressure in pounds per square inch (psi). Then divide by 2.31 and divide again by the fluid's specific gravity to get the pump's TDH in feet. In SI units, you'd have to apply the appropriate conversion factors, but the procedure is identical.

Changing parts of the system, for example by partially closing one of the valves, would impact the flowrate through the pump. Changing flowrate would change the TDH in accordance with the pump's curve.
HTH,
Doug

[djack77494]

drdave,
Many apologies. I neglected to consider the suction side pressure gauge in responding to your inquiry. Essentially, my response would not change, except that I would be curious as to how the pressure I expected (+1500 mm WC) decrease to -150 mm WC. What happened to the +1.35 m of fluid static pressure? I am assuming that the tank's elevation is relative to the centerline of the pump's suction flange (which is the elevation used by the pump manufacturers and in most calculations). I also assumed that the PG was at this same location. If either assumption in incorrect, you would need to better define the problem to get a quality response.

If both assumptions are correct, then other pressure losses would have to be significant. The only two losses I can think of are hydraulic pressure losses and fluid acceleration losses. The supplied information suggests that both should be negligable.

Please elaborate on the reason for the low suction pressure and I'll try to improve on my response.
Doug

[drdave]

Hello Doug,
I forgot to track this question so delayed in feedback. Sorry.
Yes, Problem pertains with basic engineering knowledge.But,actually,We are experiencing negative suction to some extent.Suction butterfly valve is full open.Minor abnormal sound is also coming from pump casing.We are thinking to increase water level in basin to the extent of negative suction value.Presently, 150mm level rise causes basin overflow.
This pump has to discharge through process to C.T. top deck at 15M height (i.e. static head).It is to be taken into consideration.
I think this data may be adequate to comment further.Here,We want to pinpoint system resistance location on actual performance curve.

Thanks.
Dattatreya

[djack77494]

drdave,
I confess to remaining perplexed by the negative pump suction pressure. Typically in a system such as the one you describe, the pump would be located close to the tank feeding it, and I would expect very small hydraulic losses in the short line between tank and pump. That being true, I would expect the pump's suction pressure to be approximately equal to the tank's static head. Thus the suction PG should read close to +1500mm WC, and not the -150mm WC stated. I find this discrepancy disturbing, and suggest that there is a significant unexpected hydraulic loss occuring in the suction line (or the gauge is wrong). Also possible could be the formation of a vortex, with air being sucked into the pump. Either air entrainment or cavitation (caused by low pressure/flashing liquid) could generate the noise you indicate; increasing water level would help both situations. [Thinking more about this, cavitation should not be occuring since the fluid (water) needs to experience significant negative pressure (much lower than -150mm WC) at ambient temperatures before it will vaporize.]

Looking at the discharge, I again note that the gauge reads differently from what I would expect. Assuming that the line length is not excessive, the PG should read the equivalent of 15m WC + hydraulic losses. There is one check valve and one butterfly valve shown in this line. The losses through these valves should be minor, so I'd have expected the discharge PG to be only slightly above 15m WC. If we call it even 20m, then the PG should read 2.00 kg/cm2, as opposed to the stated 3.75 kg/cm2.

By the way, and in answer to your original query, the pump TDH is calculated as so:
The discharge gauge reads 3.75kg/cm2(G). Convert this to 37.5 meters of water column.
The suction gauge reads -150mm or -0.15meters of water column.
The pump is generated the difference (TDH) or 37.5 - (-0.15) = 37.65 meters.
(This neglects what should be minimal hydraulic losses between the pump and the gauges.)

I suggest that you try to obtain or estimate the liquid volumetric flowrate. You have the pump's TDH, and I assume you would have the pump curve for such a large pump. Confirm expected pump performance by checking the flow and head against the curve. That should tell you a lot.

One more thing to check. Confirm that your liquid velocities are reasonable. At the pump suction, the velocity should normally be not much more than 3 m/sec.

Let us know how this works out.

Good Luck,
Doug