Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SSS148 on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Pump Curve question 4

Status
Not open for further replies.

asifraza0

Chemical
May 29, 2006
62
Ok, experts need your opinion on this..
I'm designing a cooling glycol circulating pump which is used to cool the inter-coolers and after-cooler of a fairly big Nitrogen compressor. This is a closed loop system where the pump discharges in an air cooled Heat exchanger, then to the compressor and back to the pump suction/glcyol expansion tank.

The design point is 465 GPM at 90 ft of head: Pump curve attached
The shut-off head is 102 quite close to the design point

The questions:

1) Whether the flat pump curves poses any issues related to operational flexibility. If during operation there is a need for a higher head the pump would then not be able to deliver any head more than 100 ft
2) The location is north Canada where ambient temperatures are low and this results in lower glycol temperatures. The operation folks would then throttle the compressor coolers valves on return line to control the temperature. Does this have any adverse affect on the pump operation with this flat curve?

Your comments/opinions are greatly appreciated
Asifraza0

Asif Raza
 
 Pump Curve
Replies continue below

Recommended for you

Neither of your attachments loads.

Try the click here to upload your file instead of the paperclip.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
attaching pump curve again
asifraza0

Asif Raza
 
Where ????????????????

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.)
 
Artisi: I'm not a computer moron but still not being able to upload the images.
I've tried the top green square (upload images in png or jpg format and also tried the bottom link to upload images but neither works. Pl show me how to, Thx

Asif Raza
 
The bottom link normally works by opening a second window ( do you get the pop-up?) Sometimes security on some computers / systems prevent it opening a second window.

Then browse to where your file is stored, click open, then click "Upload file" , WAIT, then click the link below the uploaded file name which says "Click here to insert your files link....."

Done here with a small picture.

If it doesn't work then I suspect your security settings won't allow it. Try a different computer.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
 http://files.engineering.com/getfile.aspx?folder=dc28a895-11dc-4db3-92cb-4f31782143eb&file=vuvuzela.png
Some questions:

- Did you buy the pump ? Competent pump vendors will supply custom pump curves for a variety of viscosity and SG conditions .

- Flat pump performance are undesirable for a variety of reasons .... but usually indicate that a "cheapie" pump was selected by cost savings management. Operational problems, of course, will be addressed by the field engineer.

- Is there a single pump in this service or is there an installed spare ? ( also called A/B configuration)

- Is the pump(s) on a VFD ?

- Why would there be an increase in pump head for a closed system ? Why would there be a need to throttle the pump discharge ?

- Winter/Summer system temperature control should be achieved by a bypass throttle on the coolers. Do you have such a bypass ??

MJCronin
Sr. Process Engineer
 
OK,

Even without the pump curves, the issues are there to be answered.

Your issue is how to marry your pump to its system curve.

A flat curve is not such an issue providing that the system curve is relatively steep at the duty point and you've only got one pump working ( i.e. not in parallel)

You are correct in that if the system needs more head then your pump doesn't have it, but that's true of most pumps.

so long as your flow is more than 50-60% of duty.point, your pump should operate Ok at lower flow. The flow might hunt a bit more if the flow is controlled by a fixed pressure drop, but without data it's difficult to judge this.

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

The pump hasn't been bought yet.
Yes there is a bypass around the air cooler
The pump is fixed speed and a single pump operation
The pump curve is steep beyond the duty point
I will try to upload the curve from a different compuyter tonight
Thx
Asifraza0

Asif Raza
 
pump curve enclosed..

20160127_145301_1_tolozm.jpg


Asif Raza
 
Flat curve: applied on closed piping systems with modulating valves. A flat-curve pump provides good pressure regulation where the flow rate fluctuates rapidly.

With a flat curve, the head rise is limited at reduced flows. A very high head rise as a result of closing control valves may cause problems. Excessive flow rates through the coils and greater pressure differences across the control valves are some of the problems that can be avoided with a flat pump curve.

Steep curve: usually for open piping systems (cooling towers), with high pressure, constant flow, multiple pumps systems. The use of steep-curve pumps is sometimes desirable because they permit some adjustment of the discharge pressure without changing impellers.


 
Unless you are very sure of your total head calculations, I don't think this is a good pump selection, a small change in total head could result in dramatic shift in flow. However without the total system head information (system curve) it is difficult to give you good meaningful information.
However the impeller selected in a reduced diameter allowing for performance change if necessary and if unsure if your final duty, maybe you should consider a baseplate configuration that will allow a motor size change.

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.)
 
Without seeing the system diagram or system curve it's difficult to work this out as artisi says.

You really need some sort of flow control on this system or you could easily go off the end of the curve with only a small change in system curve.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor