motor sizing for centrifugal pump 2

[mirghaffari]

Dear friends,
I want to select the suitable motor for pump. The hydraulic power as per this formula (Ph = q ρ g h / (3.6 106)) is 1115 KW
The pump efficiency is 0.73, so the shaft power is equal to (1115/0.73) = 1527 KW
As per our pump data sheet, the motor shall be selected based on” End of curve” and at this point, the required power is equal to 1685KW
Pump coupling is mechanical type. As per API 610 table 11 we have to consider the Safety factor 110% for this pump. But I don’t know which one shall be multiply to 110%?
- (shaft power) 1527 * 1.1 = 1680 KW
or
- (End of curve power) 1685 * 1.1 = 1853 KW
Could you please advise me to select the suitable motor for this pump?

thanks

 

[BigInch]

Your spec would seem to indicate the End of Curve, so it's at least 1853 kW => 2500 HP

Independent events are seldomly independent.

 

[JJPellin]

It is not so important to size the coupling for end of curve operation. A coupling is much easier to replace than a motor. But, with that said, we still size our couplings with a factor of safety above the full nameplate rating of the motor. In the event that the pump was to seize up, the full torque capability of the motor up to overload would be experienced at the coupling. If the pump is wrecked, adding a wrecked coupling does not greatly change the cost and impact of the repair. But, I would prefer to not have a coupling failure.

For motor sizing, end of curve power for the installed impeller is a good idea. If you think you might have need to upgrade the pump to a larger impeller in the future, and if the cost to upsize the motor is not too great, you could consider sizing the motor for end of curve and maximum diameter impeller. But, for a pump as large as yours, this is probably impractical and unnecessary.

I may be misunderstanding your question. But, you seem to be asking if it is appropriate to size the motor based on the coupling. I would say no. I would size the motor based on the load (end of curve shaft power). I would then size the coupling based on the motor (motor power times 1.1 to add the factor of safety).


Johnny Pellin

 

[Artisi]

You really need to ask yourself, 1.) that is / where is "end of curve" and 2.) can the pump ever operate in this region. I would expect that this will be a motor build to suit your requirements so I would involve the motor manufacturer in any discussions that will finalise the motor selection / sizing making them fully aware of the operating requirements.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[mirghaffari]

thanks all,

dear JJpellin,
my consern is not aboat the sizing of couplig, my clear question is this:

the pump manufacture sujest the 1700 KW motor for our requirment (as explaine in my thread). i want to check that this power is enough or not?
based on our rated flow, this is good selection and cover our requiment (1700>1680) . but based on end of curve, i am not sure that i have to compare the motor power with (1.1* end of curve power) or not? at this regard this motor is not suitable (1700<1853)
thanks for your good advise.

 

[ScottyUK]

The electrical engineers will set the motor protection to prevent the motor being damaged by overloading, so the motor will trip if it operates in the end-of-curve region for any length of time. If this isn't acceptable then you need a motor which can deliver enough power without overloading.

 

[mirghaffari]

Dear artisi,

As I know, in some project specification, only for parallel operating pumps, the motors shall be sized based on end of curve. I think the reason is as follow:
Please consider two same pumps are working in parallel and they pass the 100 m3/hr flow in the line. In this case each pump handle (50m3/hr).
Now consider that suddenly one pump run out and failed. What happen for other pump? That pumps work with (Q=50m3/hr) or not?
If you draw the performance curve and also system curve, you found that in this case the flow of the pump increase. Now It may work with (70 m3/hr). so the designer select the motor for end of curve for parallel pumps operation.
This is my reason for this clause of some specification. If other friends have better idea I like to hear it.and if i am in mistake please advise me.
Thanks a lot.

 

[BigInch]

We always have a better idea, but if you're working to spec, it hardly matters what we say about that. Discuss any options to change that with your client.

Independent events are seldomly independent.

 

[DubMac]

Something else to consider:
If you are on one baseplate, have it sized to accept the largest motor the pump can handle. At least ask the pump supplier at what size motor you would have to move up to the next size baseplate.

 

[poli60]

I think, with all these doubts, you should go for the solution covering the 1853 kW.
Anyway, which is the difference in cost between the two solutions?

 

[rotw]

It may happen that specs or standards require to size the driver to the maximum absorbed shaft power on the power curve OR applying 10% margin on top of the rated power, whichever is the greatest. (referring about expected figures)

Oversizing equipment is not always a good idea. It increases your costs. Your absorbed power is quite high, suggesting we are talking about a relatively costly equipment.
Efficiency is low, suggest we are dealing with very high pressure injection pump instead of high flow duty. is that the case ?
Typically making the coupling even bigger might also be demanding for pump's rotor dynamic.

For example if you are in an upstream, production environment (example injection into a reservoir) you might expect changes in your operating conditions versus the production profile. The context is also something to weight.

The process is also something to look at. Do you have limitation devices in your flow lines, etc.?

When you size a motor to the end of curve with 0% margin, keep in mind the motor still has a built in internal margin, although it may overheat a bit but typically could withstand some extra horsepower. That is to say, effectively it is not 0% margin.

Do you start direct on line or with VFD ? start up is also a consideration for the final rating of the motor.

All in all, I would say first ask a clarification to your client or technical counterpart before taking any decision.

 

[ScottyUK]

On a modern motor any margin is paper thin: manufacturers are not even slightly generous with stator iron or winding copper because they are expensive. Any practical margin is likely to exist because the motor likely operates at a lower ambient temperature than the rating is defined at, which is typically 40°C for an IEC machine.

 

[Pumpsonly]

A pump /motor of this rating should not be sized based on guideline. A motor detailed study should be made based worst scenario condition on process requirement.Talk to the process people what could be the worst possible operating condition the pump will encountered and then decide how you could decide on the motor rating. The motor of this kw will be customised any way.

 

[rotw]

Pumpsonly,

quoted
A motor detailed study should be made based worst scenario condition
Unquoted

Yes but generally speaking a worst case scenario study may be more easy said than done; it may easily turnout to be a complete mess with irrealistic or highly theoretical worst case scenario that will be brought to the discussion by expert with sometimes no clue of what will really happen in the real field or ending up with lot of assumptions and than a sizing according to assumptions.

Basically I agree with your point, I am just saying that make a worst case analysis, depending on the complexity of the process, might not be always obvious.

 

[Artisi]

Posting the pump curve/s and the calculated system curve and a discussion on the operation might go a long way in getting understandable and useful answers. At the moment everyone is having to guess what is going on / likely to be going on with the system.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[fe2655847]

Mirghaffari,

So long as the motor nameplate rating is greater than your end of curve power, you are okay. But there are also specs that allows the end of curve to be greater than the motor rating as long as this end of curve power is covered within the percentage
of overloading the motor i.e, multiflying the motor rating to it's service factor your end of curve is below this value. Thanks.

 

[z00b]

Pump manufactures will generally include power draw as part of the Flow Vs Head curve, my recommendation is to go with that!!! If you want to slow your motor down you can apply the affinity laws.

Cesar

 

[SNORGY]

Be careful with "end of curve" sizing basis, especially with centrifugal pumps with a reasonably shaped characteristic. On occasion, you run into pumps where the maximum power required from the motor is actually not in the runout condition.