Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
BEP Purely Geometric?
- Thread starter jdogg05
- Start date
No. It is based on several things,
1.) the flowrate into the impeller closely matching the volume that the impeller's speed at the time and its "bite" can transfer to the discharge (geometry means nothing without considering speed).
2.) the friction of bearings and stuffings and seals
3.) the friction caused by the speed of the fluid moving through the pump.
4.) the recirculation from impeller discharge side, back to suction between impeller clearance space with casing
5.) the differential pressure between discharge and suction driving possible recirculation of fluid
Independent events are seldomly independent.
1.) the flowrate into the impeller closely matching the volume that the impeller's speed at the time and its "bite" can transfer to the discharge (geometry means nothing without considering speed).
2.) the friction of bearings and stuffings and seals
3.) the friction caused by the speed of the fluid moving through the pump.
4.) the recirculation from impeller discharge side, back to suction between impeller clearance space with casing
5.) the differential pressure between discharge and suction driving possible recirculation of fluid
Independent events are seldomly independent.
What BigInch said is completely right. I just would like to emphasize that the first topic he mentioned is the one which contributes the most to the best efficiency of a pump.
Depending on the outlet triangle velocity and the diffuser geometry, the BEP can be estimated, but only confirmed after testing.
Operating at the BEP, the amount of recirculation would reach its minimum value
Depending on the outlet triangle velocity and the diffuser geometry, the BEP can be estimated, but only confirmed after testing.
Operating at the BEP, the amount of recirculation would reach its minimum value
You will find today, most of the major pump manufacturers have impeller designs that go back in some cases maybe 50+ years that allows them to model any scale-ups or scale-downs with a high degree of performance accuracy. And yes, maybe the original designs were simple geometry based, it's called hydraulic design and has probably reached its high point nowadays with the assistance of computers. The comments made by others re BEP is valid although predictions can be made very accurately, it still requires physical testing to confirm the facts.
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.)
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.)
- Thread starter
- #5
OK, so when a pump manufacturer specifies a BEP, they are essentialy specifying a fluid then too... I thought pumps were specified independent of fluid type. Addressing BigInch's points, the pump manufacturer then has to make certain (important) assumptions about the fluid and the system. The only thing the manufacturer would have control over is:
"1.) the flowrate into the impeller closely matching the volume that the impeller's speed at the time and its "bite" can transfer to the discharge (geometry means nothing without considering speed).
2.) the friction of bearings and stuffings and seals
4.) the recirculation from impeller discharge side, back to suction between impeller clearance space with casing"
"1.) the flowrate into the impeller closely matching the volume that the impeller's speed at the time and its "bite" can transfer to the discharge (geometry means nothing without considering speed).
2.) the friction of bearings and stuffings and seals
4.) the recirculation from impeller discharge side, back to suction between impeller clearance space with casing"
What is so important about "geometry" what are you asking / trying to establish?
If you are discussing centrifugal pumps, be aware that all design is based on as is testing on pumping water at standard conditions, any other fluid requires the pump performance curve to be re-rated.
For more information I suggest you look on the internet, there will be a life-times reading available on anything you need to know, design, testing, re-rating, effects of temp., viscosity, atmospheric pressure, altitude, entrained air / solids, speed change, etc., etc.
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.)
If you are discussing centrifugal pumps, be aware that all design is based on as is testing on pumping water at standard conditions, any other fluid requires the pump performance curve to be re-rated.
For more information I suggest you look on the internet, there will be a life-times reading available on anything you need to know, design, testing, re-rating, effects of temp., viscosity, atmospheric pressure, altitude, entrained air / solids, speed change, etc., etc.
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.)
- Thread starter
- #7
What I am trying to establish is how exactly the BEP is established and why only one BEP is specified for a pump (and not multiple based on different operating conditions).
OK, thank you. The fact that centrifugal pumps are rated for, and tested with, water makes things make a lot more sense.
Just wondering though, if for instance, an EPC approached a pump manufacturer with a list of system specifications, would the pump manufacturer then go and re-rate (re-test) certain pumps to establish what would work best for that customers particular application? How do they incorporate the already esablished BEP into the specification when the customers specification does not match what they tested to in establishing that BEP?
OK, thank you. The fact that centrifugal pumps are rated for, and tested with, water makes things make a lot more sense.
Just wondering though, if for instance, an EPC approached a pump manufacturer with a list of system specifications, would the pump manufacturer then go and re-rate (re-test) certain pumps to establish what would work best for that customers particular application? How do they incorporate the already esablished BEP into the specification when the customers specification does not match what they tested to in establishing that BEP?
You're trying to establish why there is only one 'BEST' efficiency point for a pump? Probably because of the definition of the word 'best'.
BEP is the point at which a particular pump design is most efficient. It has nothing to do with what the customer wants, unless you are doing a custom pump design.
MikeHalloran
Mechanical
>>>What I am trying to establish is how exactly the BEP is established and why only one BEP is specified for a pump (and not multiple based on different operating conditions).<<<
AFAIK, there is no way other than to build a pump and test the hell out of it.
One part of the testing is to measure the pump's efficiency at a representative and exhaustive set of operating points.
From that data, a contour map can be generated and superimposed on the P-V curves that describe the pump's performance.
The BEP is within the highest measured efficiency contour.
Most pump manufacturers publish curve sets that include efficiency, so you can examine those contours for yourself.
There might be some possibly bizarre pump geometry that produces two or more loci of high efficiency, and there may be applications where that would be beneficial, but I haven't noticed such a pump yet. ... not that I spend any time examining pump curves when I don't have to.
Mike Halloran
Pembroke Pines, FL, USA
AFAIK, there is no way other than to build a pump and test the hell out of it.
One part of the testing is to measure the pump's efficiency at a representative and exhaustive set of operating points.
From that data, a contour map can be generated and superimposed on the P-V curves that describe the pump's performance.
The BEP is within the highest measured efficiency contour.
Most pump manufacturers publish curve sets that include efficiency, so you can examine those contours for yourself.
There might be some possibly bizarre pump geometry that produces two or more loci of high efficiency, and there may be applications where that would be beneficial, but I haven't noticed such a pump yet. ... not that I spend any time examining pump curves when I don't have to.
Mike Halloran
Pembroke Pines, FL, USA
- Thread starter
- #10
Ok, so the BEP is the theoretical efficiency limit and it would be difficult to spec a pump that would operate EXACTLY at that point.
What about pumps that are made for fluids with significantly different densities and viscosities than that of water (acids, emulsions, etc.). Those surely aren't benchmarked using water..?
What about pumps that are made for fluids with significantly different densities and viscosities than that of water (acids, emulsions, etc.). Those surely aren't benchmarked using water..?
MikeHalloran
Mechanical
No, BEP is not theoretical; it's measured, on a hopefully representative sample, and allowing for manufacturing tolerances, it should be possible to make any nominally identical pump reach BEP, in a system that allows it.
However, operation at BEP is not always at the top of the design goal list for real pumps in real systems.
Oh. Pumps for services other than water surely are benchmarked using water, because it's 'standard' and available at little cost in most locations. Corrections for other fluids are relatively simple to apply, and reproducible enough for most purposes.
To not answer your original question, it should be possible to generate a pump's curves solely from its geometry using CFD, but if anyone has actually done it, they're not talking about it, at least not loud enough for me to hear.
... and I don't expect it to happen real soon, because CFD is compute- intensive, and pump manufacture is basically a foundry operation, which is not compute- intensive.
Mike Halloran
Pembroke Pines, FL, USA
However, operation at BEP is not always at the top of the design goal list for real pumps in real systems.
Oh. Pumps for services other than water surely are benchmarked using water, because it's 'standard' and available at little cost in most locations. Corrections for other fluids are relatively simple to apply, and reproducible enough for most purposes.
To not answer your original question, it should be possible to generate a pump's curves solely from its geometry using CFD, but if anyone has actually done it, they're not talking about it, at least not loud enough for me to hear.
... and I don't expect it to happen real soon, because CFD is compute- intensive, and pump manufacture is basically a foundry operation, which is not compute- intensive.
Mike Halloran
Pembroke Pines, FL, USA
Basically the OP question would make sense for a centrifugal compressor.
If you change the gas inlet conditions then the you shift the map/curve.
During testing, Manufacturer build the performance database of the impellers using a standard gas. From that the curves are re-calculated for different/actual conditions using corrections (similitude laws and regression techniques). So for instance, when it happens that the real/actual gas and conditions used for prediction are close to the as-tested conditions, you will see immediately the manufacturer reduce their safety margin/tolerance on efficiency to gain some competitive edge.
Back to pump, BEP is nothing more, nothing less than a region of max efficiency in the iso-contour of the performance map. So the question of OP reads to me as follow do you shift the map/curve (up/down, left/right, rotate/steep) if you change the pumped liquid properties operating conditions by x,y % of course for a fixed impeller geometry/speed? I would say YES but needs someone to confirm.
I would add that the OP question is theoretical as in real world applications, before to notice some visible change in the curve/BEP (if any) you may have to change considerably your properties and operating conditions, that would already lead to issues upfront :
- Pump casing design parameters are limited (pressure, temperature)
- NPSH margin can be compromised
- Material selection of the pump is set for a specified medium
- All points mentioned by BigInch (bearing, sealing, recirculation, etc)
If you change the gas inlet conditions then the you shift the map/curve.
During testing, Manufacturer build the performance database of the impellers using a standard gas. From that the curves are re-calculated for different/actual conditions using corrections (similitude laws and regression techniques). So for instance, when it happens that the real/actual gas and conditions used for prediction are close to the as-tested conditions, you will see immediately the manufacturer reduce their safety margin/tolerance on efficiency to gain some competitive edge.
Back to pump, BEP is nothing more, nothing less than a region of max efficiency in the iso-contour of the performance map. So the question of OP reads to me as follow do you shift the map/curve (up/down, left/right, rotate/steep) if you change the pumped liquid properties operating conditions by x,y % of course for a fixed impeller geometry/speed? I would say YES but needs someone to confirm.
I would add that the OP question is theoretical as in real world applications, before to notice some visible change in the curve/BEP (if any) you may have to change considerably your properties and operating conditions, that would already lead to issues upfront :
- Pump casing design parameters are limited (pressure, temperature)
- NPSH margin can be compromised
- Material selection of the pump is set for a specified medium
- All points mentioned by BigInch (bearing, sealing, recirculation, etc)
Here is how this works: start with fluid conditions (including one design point, one head one flow), select pump, review the preliminary curve, buy pump, test pump with water, factor water test data to actual fluid conditions and see if it is within testing tolerances, ship pump, install pump.
I would disagree with the statement that "a pump" only has one BEP. There are a lot of physical variables (impeller diameter, impeller mix, underfile, surface condition of hydraulic passages, case and impeller) that can affect the BEP of "a pump" while it is being manufactured. Even after manufacturing is complete and the pump configuration is now a constant* there are changes that can shift the BEP (running speed, impeller lift, correction factors for viscosity.) Now, with all things remaining constant, yes there is only one BEP.
*Never safe to assume that all things remain constant, bearing and wear ring clearances will deteriorate and there will be reductions in efficiency because of this. If you want a pump to have more than one BEP, run it for 10 years and then test it again.
No, it would be difficult to spec a pump with BEP matching that point during the design phase. It would be impossible to spec a pump that operates exactly at that point. Your system curve will not match the original design because everybody was conservative, so the pump will be oversized and run further left on the curve than anticipated.
I would disagree with the statement that "a pump" only has one BEP. There are a lot of physical variables (impeller diameter, impeller mix, underfile, surface condition of hydraulic passages, case and impeller) that can affect the BEP of "a pump" while it is being manufactured. Even after manufacturing is complete and the pump configuration is now a constant* there are changes that can shift the BEP (running speed, impeller lift, correction factors for viscosity.) Now, with all things remaining constant, yes there is only one BEP.
*Never safe to assume that all things remain constant, bearing and wear ring clearances will deteriorate and there will be reductions in efficiency because of this. If you want a pump to have more than one BEP, run it for 10 years and then test it again.
No, it would be difficult to spec a pump with BEP matching that point during the design phase. It would be impossible to spec a pump that operates exactly at that point. Your system curve will not match the original design because everybody was conservative, so the pump will be oversized and run further left on the curve than anticipated.
Viscosity correction factors are used for head, flow, and efficiency.
The shutoff head (zero flow) is the same regardless of viscosity, but at increasing flow the curve "sags" more with viscosity (lower head due to head correction factor.)
BEP is shifted left due to the flow correction factor.
Power is increased at higher flows due to the efficiency correction factor.
For specific gravity changes:
Flow is unchanged
Head is unchanged (it is in units of distance (feet or meters) and not pressure so SG has no effect)
Power is adjusted by the ratio of specific gravity.
Pressure is adjusted by the ratio of specific gravity.
No change in curve shape
For speed changes or impeller diameter changes, see "affinity laws."
The shutoff head (zero flow) is the same regardless of viscosity, but at increasing flow the curve "sags" more with viscosity (lower head due to head correction factor.)
BEP is shifted left due to the flow correction factor.
Power is increased at higher flows due to the efficiency correction factor.
For specific gravity changes:
Flow is unchanged
Head is unchanged (it is in units of distance (feet or meters) and not pressure so SG has no effect)
Power is adjusted by the ratio of specific gravity.
Pressure is adjusted by the ratio of specific gravity.
No change in curve shape
For speed changes or impeller diameter changes, see "affinity laws."
EdStainless
Materials
Good summary Gibson,
Unless you move to a non-newtonian fluid or a flow that is multi-phased then you can scale pump performance.
= = = = = = = = = = = = = = = = = = = =
Plymouth Tube
Unless you move to a non-newtonian fluid or a flow that is multi-phased then you can scale pump performance.
= = = = = = = = = = = = = = = = = = = =
Plymouth Tube
- Status
Similar threads
- Locked
- Question
- Locked
- Question