Understanding pump curves

[jerfy]

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1434655147/tips/VFD_Pump_curve_xoyddk.pdf[/url]

Goodafternoon,

I am hoping to get a better understanding on pumps and what exactly the manufacturer curves tell me. I don't understand why the attached butane pump curve seems to imply that having a greater static fluid column results in a lower flow rate. Is this exclusively for pumping against gravity? My intuition tells me that this relation would be inverse if we were using this pump to flow liquid downhill rather than uphill, with the greater fluid pressure helping rather than hindering.

 

[bimr]

A pump does not create pressure, it only creates flow. Pressure is a measurement of the resistance to flow.

kinetic energy of a liquid coming out of an impeller is harnessed by creating a resistance to the flow. The first resistance is created by the pump casing which resists the liquid and slows it down. When the liquid slows down in the pump casing some of the kinetic energy is converted to pressure energy. It is the resistance to the pump flow that is read on a pressure gauge attached to the discharge line.

If the speed (revolutions per minute) of the impeller remains the same then the larger the impeller diameter the higher the generated head. Note that as you increase the diameter of the impeller the tip speed at the outer edge of the impeller increases commensurately. However, the total energy imparted to the liquid as the diameter increases goes up by the square of the diameter increase. This can be understood by the fact that the liquid's energy is a function of its velocity and the velocity accelerates as the liquid passes through the impeller. A wider diameter impeller accelerates the liquid to a final exit velocity greater than the proportional increase in the diameter.

 

[jerfy]

Thanks bmir,

perhaps I was not clear in my question. Our butane tanks are uphill and our pump is downhill pumping into an even further downhill line. If you view the attached pump curve, let's say we want 150 GPM from a 5 stage 40 HP pump. The curve says we can overcome 600 ft of butane fluid head, but if we want to see 300 GPM, we can only overcome 500 ft of head. Why does the greater gravitational force of 600 feet of fluid have a lower flow rate? Isn't our static head adding pressure and thus increasing flow rate to the system?

 

[bimr]

You will get more flow in your system when pumping downhill.

You need to plot a system curve and the point of operation is where the system curve meets the pump performance curve.

If you look at the diagram, the static head offsets the curves by the static head difference.

In the illustration above, some flow rate will occur by gravity head alone. But to obtain higher flows, a pump is required to overcome the pipe friction losses in excess of "H" - the head of the suction above the level of the discharge. In other words, the system curve is plotted exactly as for any other case involving a static head and friction head, except the static head is now negative. The system curve begins at a negative value and shows the limited flow rate obtained by gravity alone. More capacity requires extra work.

You need to develop the actual pump curve to graphically compare the pump head and the static head.

 

[DubMac]

Jerfy
Maybe I'm not understanding your post, but I think it's just a terminology problem. when you say:

"Why does the greater gravitational force of 600 feet of fluid have a lower flow rate? Isn't our static head adding pressure and thus increasing flow rate to the system?

The 500 or 600 feet of head values shown on the pump curve are for what the pump sees on the discharge side of the pump, they don't represent "gravitational force" or static head supplied by your uphill tanks to the suction side of the pump.

In terms of having a large gravitational head on the suction side, remember, the pump curve does not know that, it is only showing you the DIFFERENTIAL energy it will add to the fluid between it's suction and discharge flanges.

The pump curve is saying that if there is 600 ft of head on the discharge side, then it will be able to supply 150 GPM. If the head on the discharge line is lowered to 500 ft, the pump will then be able to supply 300 gpm. The less resistance to flow the pump sees at it's discharge flange, the more fluid it will be able to move.

Hope that makes sense.

 

[georgeverghese]

At low to some higher flow, due to the static height difference, this butane may be flashing out inside the pump and downhill line - the capacity control scheme should prevent this. Pump capacity / integrity may be affected at these flows.

 

[LittleInch]

Ierfy,

You're looking at this the wrong way around by thinking that the pump is determining the flow rate. Centrifugal pumps will increase flow until the resistance to flow matches the pump outlet head. Look at bimrs system curve.

To get 150 gpm, you need to reduce outlet head compared to 300.

The pump supplier has no idea about your system our how it is controlled.

If you want the pump speed to control flow you need a positive displacement pump.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[jerfy]

Thank you all very much,

Sorry if I am confusing, I am fairly new to the workforce and pump engineering is not what I studied.

I think I see where my misunderstanding lied, between understanding static and dynamic head. Now I am wondering how exactly I can put together a pump curve for our system? The pump is typically held set at 75 psi, but based on what you all explained I want to say this is the discharge head, not the total dynamic head.

I understand TDH=discharge + suction + friction. So I have a metric for the discharge. I must calculate the suction head based on fluid column height? The suction head calculation is confusing me. And how can I know how much pressure is lost from friction?

 

[bimr]

For a start, it should be helpful to review the McNally site:

http://www.mcnallyinstitute.com/home-html/Technical_paper_index.html

Calculating the total system head:

http://www.mcnallyinstitute.com/07-html/7-01.html

How pipe friction, liquid capacity and pump head are affected by the amount of liquid flowing through a pipe:

http://www.mcnallyinstitute.com/11-html/11-08.html

 

[jerfy]

Thanks, I have been reviewing this resource all morning to begin piecing together my equation.

Our butane tanks are pressurized to maintain liquid state. I'm wondering if this pressure is contributing to the suction head? I would imagine so, and that this is the "gauged head" but want to confirm my suspicion.

 

[DubMac]

Yessir! Have you tried getting a gauge measurement at the suction flange? Remember, the pump doesn't know about anything that happens outside it's flanges.

 

[Artisi]

And doesn't really care what is happening between the flanges.

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.)

 

[LittleInch]

jerfy,

It's best to use full names rather than abbreviations when you're doing something like this.

TDH means Total Developed Head at the outlet flange = inlet head + differential head of the pump. That's it - No friction

The differential head is the figure you get from the pump curve versus your volume flow and takes account of all losses inside the pump from inlet flange to outlet flange.

bimrs excellent graph shows you your issues. You have a fixed head from one tank to another and if you open the valve you will get a certain amount of flow assuming that both tanks are held at the same pressure. That flow is the intersection between the horizontal dashed line and the system curve.

To get more flow you need more head at the outlet which is supplied by your pump. So continue your system curve until yu get ot the flow you want. Your pump curve will be provide that EXTRA head on top of the static head you have.

Yes the inlet head/pressure will be influenced by the pressure your tank is at. In your case that pressure is effectively cancelled out by the same pressure in the receiving .tank, but it is better to add this to both sides of your equations in case they are a bit different.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.