Starting current for squirrel cage prime mover on a cent. pump 13

[jlhaas]

This is probably a bit basic but I was asked a question the other day regarding the procedure for starting up a centifugal pump. The procedure usually involves shutting the discharge valve, turning on the pump, then openiong the discharge valve. I was told the reason for this procedure was to limit starting current on the motor (small motor - no starting resistors). I haven't been able find a good description for why the discharge valve being shut helps.

 

[rmw]

Marke,

What you said about 'no flow, no work' is certainly true, and a pump working against a shut off valve is, in fact only circulating fluid in the pump housing, but that is not cavitation.

Cavitation, and there are lots of threads in other places in this forum, is a condition of flow, not the lack thereof.

Rodmcm is correct in his assumption that discharge head is produced against the shut off valve. It is called 'shut off pressure' in pump parlance. And, if left in that condition too long, the electrical work done on the pump at no flow will just heat up the fluid in the pump, and in the case of water, will after a finite time, boil or flash it, vapor locking the pump. While often confused with cavitation, it is not that.

A pump always draws some horsepower, if no more than just to overcome bearing and seal friction. It draws the least amount of horsepower possible at 'shut off', and the most horsepower at wide open flow.

Apart from water hammer , which is an problem in and of itself, getting to the point at which the hammer occurs, or filling an empty pipe to the point of hammer, is a wide open flow situation, and the pump is requiring maximum HP from the motor, all while the motor is trying to start and come to speed. It is the worst of all worlds, hence, the procedure of shutting the discharge valve to start the pump.

rmw

 

[GOTWW]

How about PF correction, that is the best hope for a weak line, starting pf is low, so maybe a switched capacitor bank. Not really desirable.

Best yet, tell the utility that thaey are failing to me ansi c c84.7 voltage drop requirements, it is their problem now.

 

[jbartos]

Comment on electricpete (Electrical) Mar 18, 2004 marked ///\\
The pump will affect how long that starting current lasts.
///True.\\ The more torque the pump draws during startup, the longer it takes the unit to get up to speed and longer the starting current lasts.
///It depends. If the motor HP is proportionately increased and the correct NEMA Design Letter is applied, the larger motor-pump set may start faster than some smaller motor-pump set with a relatively small HP motor for the pump.\\\

 

[Marke]

Hello rmw

Yes I agree, the fluid spins within the housing, but I am led to believe that there is also cavitation withing the pump and this can cause damage to the impeller over a period of time. - I am not a pump expert, so this information may be wrong, but the description given by many is that it is this cavitation loss that is reduced when a pump with restricted flow is slowed down. Is this a reasonable explanation, or are the losses elsewhere?

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[PUMPDESIGNER]

The defining factor for the pump under starting conditions in short is as follows:

Pumps are variable torque loads.
The power curve for a pump defines how the pump loads the motor at any given flow rate. This power curve is supplied by the pump manufacturer.

However, lacking manufacturer's power curve, Pump Specific Speed (Ns in the US) is an indicator of the relationship of the power curve to flow rate. In general the following applies based on Pump Specific Speed:
Low Ns pumps (< 2500)draw least power at low flow rates, while high specific speed pumps (> 9000) draw maximum power at low flow rates. When Ns is between those two values the power curve is less dependant upon flow rate.

In review.
Since High Ns pumps (called axial flow pumps) draw maximum power at shut off, those pumps should not be started against a restricted output or the motor will be heavily loaded,

while,

Low Ns pumps (Radial Flow and lower value Francis Vane impellers/pumps) draw lowest power at shut off or low flow rates, so those pumps are often started against maximum resistance. In reality though, for Low Ns pumps less than 50 hp, starting against closed valve or open valve usually does not matter because the starting time for these pumps is so short that the effects of load on starting characteristic is very low, perhaps extending the duration of the high inrush by a small amount I would guess. For High Ns pumps I would think big trouble could occur if started against a closed valve or high resistance, inrush current might only fall to locked rotor value at most.

PUMPDESIGNER

 

[Rodmcm]

Thanks Pumpsdesigner, but could you educate me a bit more, how do you derive the power of a donkey motor starter from the pump curve. Do pump curves show zero flow, against a shut off head? I don't believe I have ever seen this.

 

[MikeHalloran]

Back to the original question.

Another reason why starting a centrifugal pump against a closed discharge valve is good practice: It's entirely possible that the pump is already running, backwards.

All it takes is a check valve that's failed open, and a downstream pressure source like an elevated reservoir or a parallel pump; it happens.

-Mike-

 

[PUMPDESIGNER]

Follow this link to see Power Characteristics for different pumps. Near the bottom of that web page there are 3 other links that take you deeper depending on how interested you are.

http://www.irrigationcraft.com/power_curve.htm

I am not an electrical expert, and I am not sure how starting characteristics relate to the power curves of pumps. Starting is a fast event when going direct across the line, and pumps are variable torque loads, so the effects of power curves on starting are unknown to me.

Much earlier in this thread I asked electricpete (who is an expert in my opinion) about the affects of pump type on starting current and he agreed that start duration time may be affected by pump type.

PUMPDESIGNER

 

[Rodmcm]

Thanks for that, I hope I'm not wasting time here but the queston is important. If you have a closed discharge valve, and the electrical start options are not viable ,then how do you calculate the power of a small donkey motor to raise the pump/main motor speed to full speed. You are running the pump from 0 to full speed over, say, for fun 2 minutes, as the speed increases the pump must start to deliver water to the closed valve, hence the pump sees 'shut off pressure'. I can see nothing on a pump curve that allows me to determine torque or the work expended over the two minutes, that's my problem. Anyone with any ideas?

 

[PUMPDESIGNER]

I see why we are missing each other. You do not understand the power characteristic of a pump and the operating window or range of a pump, or possibly what those things imply.

The power characteristic is the power required to maintain the pump at a constant speed against varying degrees of resistance.
If the pump is held at a constant speed, and the flow rate is then taken from zero flow to maximum flow while power input is measured at points through the flow range, then the power characteristic is produced as a line on a graph with head and flow rate as the comparison.

Then, to obtain the power at varying speeds the Pump Affinity Laws are used. Of course you could just test the pump that way, start the pump at 1 rpm then ramp it up through the flow range, measuring power draw at specific points to produce the curve. Most pumps are not tested that way however, the manufacturer assumes you will derive the power characteristic in the form that you require by the Affinity Laws. The Affinity Laws are not always dead accurate however because they assume efficiency remains unchanged. But the laws are used extensively and are pretty reliable. Only testing can reveal the dead accurate power curve.

I am not a whiz at typing formulas as text, but here are the various forms of the affinity laws, where:
Q = Volumetric Flow Rate
N = Rotative Speed (RPM)
P = Power or BHP
D = Diameter of the impeller
H = Head

Q1/Q2 = N1/N2 (Solving for flow rate vs. rpm). Where the flow rate and rpm are known under condition 1, then under condition 2 either flow rate or rpm is known and the other can be solved for.

H1/H2 = N1squared/N2Squared (Solving for head vs. rpm). Notice that head varies directly as the square of the speed.

BHP1/BHP2 = N1cubed/N2cubed (Solving for power vs rpm) Where power is known at rpm1, power required for rpm2 is then solved for.
Notice that power varies directly as the cube of the speed.

I have not taken the time to correct something erroneous I have seen on this thread. It is incorrect to state that a centrifugal pump should be started against a closed valve. The reason is that there are different types of centrifugal pumps. Specific Speed is an indicator of the power characteristic and the operating range of a pump, and thus the reason why a pump should be started against either a closed valve or an open valve. However the pump manufacturer's published information is more accurate and must be used. Some centrifugal pumps would either not start or would damage themselves quickly if started against a closed valve, some centrifugal pumps don't really care much if valve is open or closed, and some pumps prefer to start against a closed valve. Specific Speed and the pump manufacturer's published data are the sources that tell us how wide the recommended operating range of the pump is, and how the power characteristic relates to flow rate.

The operating range is both a Subjective and Objective determination by the pump manufacturer. The range takes into account the pump's responses to operationg off BEP including power, vibration, Recirculation Cavitation, and materials of construction including shaft stiffness or R3D4. Bottom line here is that some pumps cannot tolerate shut off for even short periods, while others work well at shut off.

I am throwing out what I think you can use without trying to tailor it to your abilities because I do not know what you may or may not understand.

I have the time now to pay attention to this thread, so ask on if you want.

PUMPDESIGNER

 

[Rodmcm]

Thanks I will think on that for a while