Why impeller diameter should not to exceed 95% of the max impeller ??

[Issarin]

Hello,

In API 610 9th edition article 5.1.6 states that :

"Pumps shall be capable of at least a 5% head increase at rated conditions by replacement of the
impeller(s) with one(s) of larger diameter ..."

And I've seen some project specification (for seawater pump service) which is modified from article 5.1.6 by adding :

"The diameter of the impeller(s) provided (rated impeller and the one that gives 5% head increase), shall not be greater than 95% of the
maximum impeller diameter that may be installed in the pump."

Now, my question is why they have to specify the diameter shall not be greater than 95% of max impeller.
What is this margin for ?

As my understanding the smaller impeller means more clearance which is less efficiency so I don't understand the specification above.

Ps. The motor for this pump is sized for the 5% head increase impeller only not cover the max impeller.

Thank you for your help.

 

[BigInch]

I think this provision is for the case that somebody gets the required system head wrong. It is often stated that accuracy in hydraulic analysis, after all typical ranges of variables are considered, roughness, temperatures, viscosities, fitting losses, etc., is about 10%. That would supposedly would be +/- 5%.

After round-off and selection of the next higher common size, its not unusual to find a motor sized at 10-15% over what is actually needed.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

It's called the fudge factor - just in case the specifying engineer / designer gets the total head wrong, although in my experience unless a genuine mistake has been made with the head calculation, head is usually over specified resulting from all the margins added by the various parties along the way to final delivery of the pump unit.

 

[Issarin]

BigInch,

Thanks for your quick reply.

I understand that 5% head increase is for case that someone miscalculates the head so we have enough space to install the larger diameter impeller to get the required head.

But in this case, if that provision (not exceed 95% of max impeller) is also considered as a margin for someone miscalculates the head then they have to get the new motor for the impeller that exceed 95% of max impeller and the size is quiet big (1100 kW - Motor rating) so this would cost a lot.

Therefore, I'm not sure this is the real reason or not or maybe there are some reasons that I can't think of.

Do you think is there another reason except the one we're talking ?

 

[Artisi]

Well a smart engineer selecting a pump thats need a 95% impeller to achieve the duty would probably select a motor capable of running the 100% impeller - just in case.

 

[BigInch]

Maybe the same can be said for the pump design. Its hydraulics too. Remember, in situ pump flow tests don't always get the curve they advertised on the back of the box either.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

BigInch - I would say that in general, any pump supplied from a reputable manufacturer will meet the published curve on a properly controlled test. However, in saying that and speaking from experience it is "normal" for pumps to not achieve the "curve" on site, this is usually not a pump fault but rather a different system requirement than was anticipated or poor on-site testing protocol.

A number of times I have had pumps removed from site for retest - sometimes tested by other pump companies to remove any bias in testing just to prove that the pumps perform as offered.

What must be remembered here also is that all testing is subject to a testing code tolerance and some end users have difficulty in accepting a +- 5% or what ever the code allows. There is a good argument as a supplier or buyer to offer / request certified witness testing prior to delivery as this removes all doubt as to the pumps performance, it could also explain the 95% diameter rule being discussed here.

 

[JJPellin]

I think the other replies have hit upon all of the major points to this. But I can't resist adding a few thoughts. In my plant, we tend to purchase our pumps with excess head capability which we bleed down across control valves in order to be able to control the flow over a range. This wastes a lot of energy, but gives us the control that we need. However, in some applications and other industries, it is not unheard of to purchase a pump to meet the expected system curve with no control valve and expect the pump to run at the point where the pump curve crosses the system curve. We have a couple of high pressure pumps that run this way. Unless you specify a guaranteed minimum performance point when you buy the pump, then the OEM is only obligated to hit the rated performance within +/- 2% (I think). As noted above, the system head requirement is probably overstated because of fudge factors, but this is not certain. Even with the pump built to accommodate an additional 5% head increase, you could have a problem because of the shape of the curve. We have a high pressure water wash pump the runs up to the system curve with no control valve. The performance of this pump has degraded by 4% in head over the past few months. But, because the pump curve is pretty flat, this has caused a drop of more than 20% in flow. Even just 4% off the curve, we are at the minimum allowable water injection rate and at risk of fouling and corrosion problems.

If a pump was purchased for 100% of expected head, and it came up short, it would be very difficult to correct. The foundation, base plate, piping, motor driver, conduit, wire, starter and switchgear were all designed for that pump. If you had to change to the next size larger pump, it could be a monumental project. It is better to allow a little bit of wiggle room for slight underestimations (not necessarily gross errors) such that the original pump and motor could meet the higher head requirement.

P.S. BigInch. Do you ever sleep?


Johnny Pellin

 

[BigInch]

I didn't like it.



**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[dabluffrat]

I don't think I saw it mentioned, but trimming the 5% away also helps increase your cutwater gap and reduce vane pass vibration as well.

Did you know that 76.4% of all statistics are made up...

 

[Artisi]

That's if you have vane pass vibration.

 

[wimple]

Guys

Just a question...

If for any reason, the pump during test or commissioning fail to achieve the required head, so actual head is lower then expected. Let say we move to 100% impeller diameter to recover. Then why the absorbed power at the recovered head should change ? Is it due to some efficiency difference ? what about the fact that 100% impeller diameter will allow reduction of leakages even quite marginal.

Thanks & Rgds

wimple

 

[BigInch]

When you change impeller, head will change accordingly. You would normally calculate that by taking the head curve you have with the first impeller's radius and ratio that to the second, without changing the efficiency. Efficiency is likely to be affected positively, however small. If you do get an efficiency increase, consider it a bonus.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[Artisi]

If the head tested out lower than expected at a reduced diameter impeller firstly you need to establish why - but you would normally assume that the input power would also be reduced unless there is an impeller hydraulic problem.

Fitting another impeller at an increased diameter will result in a different hydraulic performance.

If H and Q are meet then the only way power can change is for the impeller / case having a different hydraulic performance.

Remember, performance testing has a code tolerance on Q,H, Eff. which must be acceptable by both manufacturer and buyer.

In many cases a manufacturer having to prove performance on a critical performance test may well fit a slighly oversized impeller for a pre-test prior to any official testing taking place.

If a pump fails on test there are a few things that can be done to try and enhance performance - but this is another story.

 

[wimple]

/Quoted : But in this case, if that provision (not exceed 95% of max impeller) is also considered as a margin for someone miscalculates the head then they have to get the new motor for the impeller that exceed 95% of max impeller and the size is quiet big (1100 kW - Motor rating) so this would cost a lot.
/Unquoted.


So at the opposite of what is pointed out above, this is definitely not to rise issue on motor power margin. Usual 10% motor oversizing, should be fine.

 

[BigInch]

Another thought.

For attaching a VFD and increasing the impeller size to yield a curve equal to the non-VFD impeller curve.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[wimple]

Biginch

Why the VFD 100% speed and non-VFD constant speed should be different?
Are you meaning for example some possible slip in the rpm due to VFD design to be corrected by impeller diameter to match the head with 0 negative tolerance or to keep the gearbox aligned if any ?

 

[BigInch]

Ya to the slip part.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[sridhar1312]

I have situation Primary chilled water pump operating with 1.8 Kg/cm2 discharge and enters the chiller and after chiller the pressure is 0.8 Kg/cm 2 then it goes to secondary chilled water pump where suction is vaccum(-1 M) and thus the pressure drop from chiller outlet to secondary suction is 1.8 Kg/cm2 or 18 M.

Kindly let me know Whether the pressure drop of 18 MWC is due to?

 

[Artisi]

sridhar1312
Start a new post to discuss your problem.