NPSH Effect on Pressure Calculation

[jebatty]

I am trying to refine some performance calculations by using pressure gauges on both sides of a circulator to more accurately determine flow. It appears that I may have NPSH on the circulator inlet, at least the 30 psi gauge does not shift the needle off "0." For example, at Speed 2: outlet pressure 8 psi, inlet pressure 0. While at Speed 1: outlet pressure 6.25, inlet pressure 1.0.

May I use the 8 psi pressure to determine flow, or does an adjustment have to be made for the (undetermined) negative pump suction head on the inlet side? I will need a vacuum gauge on the inlet to determine actual NPSH, which I do not have.

Thanks.

 

[jimbom]

Jim, Please confirm the fluid is water, temperature is the same at both speeds, the conditions of the circuit are unchanged for both speeds, and you have a centrifugal pump. If the NPSH for water at some temperature is given in your pump specs that would be helpful. Is there a flow meter in the circuit? Do you have 10 pipe diameters of straight pipe before each gauge?

With regard to the gauges. If you switch them, do you get the same readings. I wonder if the accuracy, precision, or range of these gauges might make it difficult to get a good handle on the numbers.

I am willing to try a solution that would answer your question.

Jim

 

[jebatty]

Thanks for help. This is a boiler, shut-down now and water at room temperature. I installed the gauges on the flanges for the circulator, which has 1/4" plugs, and the circulator is installed at the bottom of the boiler. Static water pressure as read by both gauges, circulator "off," is 2 psi.

My understanding is that installing the gauges on the flanges is a proper location for the gauges. No flow meter in the circuit, and if there was, I think I would have my gpm's determined. Procedure: 1) shut-off the isolation valves on both sides of the circulator; 2) remove the 1/4" plugs and screw in both gauges; 3) open the isolation valves; 4) power the circulator; 5) read the gauges. Then, repeat with the gauges switched. Same readings.

 

[ewdudley]

I am trying to refine some performance calculations by using pressure gauges on both sides of a circulator to more accurately determine flow. It appears that I may have NPSH on the circulator inlet, at least the 30 psi gauge does not shift the needle off "0." For example, at Speed 2: outlet pressure 8 psi, inlet pressure 0. While at Speed 1: outlet pressure 6.25, inlet pressure 1.0.

May I use the 8 psi pressure to determine flow, or does an adjustment have to be made for the (undetermined) negative pump suction head on the inlet side? I will need a vacuum gauge on the inlet to determine actual NPSH, which I do not have.

All that matters is the pressure difference, as long a there's no cavitation. So if the low side is pulling below atmospheric pressure, all you need is a gauge that can read absolute pressure, or at least a gauge that can measure both above and below atmospheric with enough range. So yeah, you need a 'vacuum' gauge, but it needs to read above atmospheric as well.

If the goal is to let the operator tell at a glance whether the unit is operating normally, or if the HX is becoming fouled, then you need a differential gauge or two gauges. For occasional system performance monitoring you could consider a single precise gauge with valving that would let you measure one side of the pump and then the other. This takes the variability of the accuracy of the two gauges out of the equation.

--ewd

 

[jebatty]

I am using two gauges, which by then using the pressure difference and density of water I can calculate the pounds of head, which then shows the gpm flow by using the pump chart. That's my goal. I did a calculated pump head by measuring length of pipe, fittings, etc. Use of the gauges is to get a more accurate determination of pump head, and thus flow, to more accurately determine btu performance of the boiler.

All that matters is the pressure difference, as long a there’s no cavitation.

This is the point of my question. If there is NPSH, and possibly cavitation, what is the effect on flow? In other words, if outlet pressure is 6.5 psi, and inlet pressure (measured with a vacuum gauge, for example) is -2 psi, is the pressure difference for purposes of calculating flow 8 - - 2 = 10 psi? ... or is it just 8 psi? ... or is/might it be something else? So long as inlet pressure is positive, I think pressure difference will enable me to determine flow. But what if inlet pressure is negative?

 

[Clarkbug]

Thanks for help. This is a boiler, shut-down now and water at room temperature. I installed the gauges on the flanges for the circulator, which has 1/4" plugs, and the circulator is installed at the bottom of the boiler. Static water pressure as read by both gauges, circulator "off," is 2 psi.

It might be a bit too early for my brain, but 2 psi standing pressure seems low to me. I know it has to do with the configuration of the rest of your system, if you are all one level, etc. This low pressure could contribute to the negative number at your circulator suction.

 

[jebatty]

I could measure to get water height exactly, but 2 psi = 4.62 feet would be pretty close. This boiler is one level, isolated from higher levels by a plate hx, so water column from the circulator near the bottom of the boiler to the top of the boiler piping is what is involved.

 

[ewdudley]

I am using two gauges, which by then using the pressure difference and density of water I can calculate the pounds of head, which then shows the gpm flow by using the pump chart. That's my goal. I did a calculated pump head by measuring length of pipe, fittings, etc. Use of the gauges is to get a more accurate determination of pump head, and thus flow, to more accurately determine btu performance of the boiler.

All that matters is the pressure difference, as long a there’s no cavitation.

This is the point of my question. If there is NPSH, and possibly cavitation, what is the effect on flow? In other words, if outlet pressure is 6.5 psi, and inlet pressure (measured with a vacuum gauge, for example) is -2 psi, is the pressure difference for purposes of calculating flow 8 - - 2 = 10 psi? ... or is it just 8 psi? ... or is/might it be something else? So long as inlet pressure is positive, I think pressure difference will enable me to determine flow. But what if inlet pressure is negative?

It's only negative relative to atmospheric pressure. There's nothing special about standard atmospheric pressure except that it's a pretty handy reference in many situations. All the water cares about is the absolute pressure relative to the boiling point and we can get in trouble if the water is hot enough and there are points on the impeller where the pressure is low enough to flash a little steam. For any given pump there needs to be a particular pressure above the boiling point to insure that there will be no flashing inside the pump.

So if the water is say 180 degF, you look up the pressure at which 180 degF becomes the boiling point of water, and then convert your measured pressure to the same units, and then you look up what the minimum safety margin is for your particular pump at the speed it's running.

And yes, you're right, if outlet pressure is 8 psig and inlet is -2 psig then the head the pump is generating is simply 10 psi. If your gauges were psia then they would read 22.695949 outlet and 12.695949 inlet and you'd still have 10 psi of head. If the outlet reads 8 psig and the inlet reads 4.0720414 inch Hg vacuum, same difference. If it were a pressurized system you might see 22 psig and 12 psig and the pump would still be pushing the same amount of water, ignoring density.


--ewd

 

[jebatty]

I understand the flash to steam issue. I have ordered a vacuum gauge so that I can determine the amount of negative pressure on the inlet. At this point I have on Speed 1: outlet 6.5 psi and inlet 1.0 psi. At Speed 2 I have: outlet 8.0 psi, inlet 0.0 psi. At Speed 3 I have: outlet 9.5 psi, inlet 0.0 psi. Thus the need for the vacuum gauge.

 

[jimbom]

EWD gives an excellent explanation. Since this is an engineered system, I am sure the designer has provided for positive suction head with all variables included. So your pump should not experience cavitation when operating with water. Even a replacement pump of normal commercial design will experience no problems.

If you are checking to see if your pump curve is accurate, then you are going to have to pop that pump out of there and put it to the test.

My initial thought was pressure drop is proportional to a constant times velocity squared. Example is the Darcy-Weisbach equation. The constant accounts for the specific system characteristics. Evaluate the constant at one known point. Then, for any pressure drop, you could arrive at the velocity squared. The problem with my approach is it needs at least one representative point where there is pressure drop plus flow data.

 

[DaveBP]

But wouldn't a 2-port differential pressure gage remove all this relativity and double gage calibration complexity?

 

[jebatty]

A differential pressure gauge probably would, but these are expensive as compared to individual pressure gauges, or a single gauge with valves to read each port separately. I also bought a used differential pressure gauge, hooked it up as recommended, and got some large anomalous readings as compared to the calculated flow rate. Thus, i switched to 2 pressure gauges. I suspect that either I was mis-using the differential pressure gauge or it is defective.