Low rpm motors for pumps

[mohtogh]

Hello,
My boss says: "always select lowest rpm (900 ?)for electromotors of centrifugal pumps(with a specific Q and H) ,because they are long last .The lower rpm , the longer life of motor."
Is it right?
Thanks

 

[quark]

Yeah. Right but I think you are rich

 

[ccfowler]

Actually, motor life is not usually an issue of speed! It is the pump life that is usually longer for lower speed designs.

 

[electricpete]

quark makes a good point that slower speed motors are always more expensive than comparable higher speed motors of same horsepower.

Some people avoid 2-pole motors (~3600rpm in the US, ~ 3000rpm overseas) because those machines can be a maintenance headache. Small problems with misalignment, unbalance etc are compounded. I know the vast majority of "problem" machines at our plant are 2-pole. Also we have some 2-pole that run fine.

Avoiding 2-poles is gaining some popularity and I believe it would be a good practice. For some reason the same logic does not follow to lower speeds. I have never heard anyone suggest that 1200rpm or 900rpm machine has any advantage over 360rpm machine.

 

[electricpete]

last sentence should be: "I have never heard anyone suggest that 1200rpm or 900rpm machine has any advantage over 1800rpm machine."

 

[d23]

All:

Ok guy’s someone has to play devils advocate here, so I’ll take a shot at it.

As Pete pointed out the slower the motor runs the large the size gets. Is a larger heaver rotor better on bearings than a small one? The larger the motor gets the higher the cost both initial or repair when needed.

You will have a constant head and flow requirement due to the system. I didn’t look at any real numbers, but assume for this system a unit running at 3600 RPM you will require an 8 in impeller, at 1800 RPM you will require a 9 inch impeller and at 1200 RPM you require a 10 inch impeller. The larger the impeller is, doesn’t balance become a bigger issue? The larger the impeller is will the thrust load increase due to impeller surface area?

My opinion:
“Slower is better” is an arbitrary statement that may need to be redefined with a few facts. I like Petes second statement that leads me to believe there may be both upper and lower extremes to rotating speed.

 

[mohtogh]

Thanks all,
one pump manufacturer told me: "for submersibles pump ,only one rpm (1800?) exist in the market " .
What do you think about this?

 

[d23]

mohtogh:

We have a full product line at 3600. The submersibles used in the petroleum industry as oil production pumps are all 3600 RPM that I know of. We also use VSD's and will operate between 30 (1800 RPM) and 90 (5400 RPM) hertz.

More conventional:
Franklin, KSB, Toshiba etc. all manufactures 3600 RPM motors for the water industry. You may want to ask your manufacture what they are for.

For your application an 1800 RPM unit may be the best choice, but it's not the only one.

Good Luck

 

[Kawartha]

I agree with d23, one should not make an rpm rule without all relevent facts. There are pumps which operate at >20,000 rpm (of course clean liquids only), boiler feed pumps operate at 3600 rpm and higher (these pumps can be as large as 25,000 hp).

On the other hand, slurry pumps should always be low speed pumps. These include rubber lined impellers and casings and hard metal pumps. In this case the slower the better, but only because of the application.

d23, I think that mohtogh's comment is based on submersible sewage pumps. But even in that market, he's wrong. He is correct that most submersible sewage pumps are designed and operate at 1750 rpm, but there are also some small units at 3500, and most of the very large designs operate at 1200, 900 and slower rpm.

mohtogh,

Many general rules of thumb are helpful, "the slower the better" is not one of them.

 

[electricpete]

I guess there are other factors related to pump specific speed and pump efficiency. For instance in very low flow high dp applications (like power plant circ water) you would typically find an axial flow pump operated at low speed.

Maybe someone else can comment on that.

Avoiding 2-pole machines will be a good thumbrule to follow unless you have a compelling reason, IMHO. If you have doubts, I suggest you talk to someone who has monitored/maintained a variety of machines to compare experience with 2-pole vs slower machines.

 

[mohtogh]

Thank you again for your excellent answers.

 

[d23]

Pete, All:

I think in most cases the equipment rotating speed will be or at least should be determined by application. I believe that you (Pete) operate chilled water pumps that I would guess for the most part are two pole motors. In my nitch centrifugal market we are attempting to create high head in a limited physical area. Sometimes I deal with a casing ID as small as 4 inches and need a dp of 5000 PSI. Higher rotational speed is required (pump affinity laws) due to cost alone. From personal headaches I can assure you that minor problems like viscosity or solids cause me a lot of problems at 3600 RPM.

One question that I do have is do you expect a "motor" to last (MTBF) longer at lower speeds rather than higher speeds? One rating for bearings is the total number of revolutions. Two pole four poles etc. all have different bearings. Should I expect a longer life from a slower motor?

 

[ccfowler]

The bearing life vs. motor speed issue is very much in the "wonderful world of tradeoffs." Bearing life is affected by speed and load, but bearing selection taking these factors into account is what determines the bearing design life. The combination of installation, maintenance, and operation then determines how much of the design life is actually realized. Extended operation with misalignment, vibration, inadequate or improper lubrication can produce a very short service life even for bearing chosen for exceedingly long design life. Conversely, a well installed, operated, and maintained machine can realize a very long service life even with seemingly marginal bearing selections.

Lubrication is a very important element in realizing maximum bearing life. Keeping moisture and other contaminants out of the bearings and lubricants can have a dramatic effect on extending life. The use of desicant breather filters, carefully measured greasing, oil rather than grease, oil mist lubrication systems, etc. can be of great value in realizing the full design life from bearings. Seemingly expensive installation, maintenance, and lubrication "extras" can provide dramatic paybacks when avoided down-time costs are taken into consideration.

 

[sprintcar]

This is fine with Vertical pumps, but you may want to reconsider with horizontals. As mentioned above, matched speed motors are bigger and more expensive. PLUS - with the exception of VFD(VSD) capable units (even more expensive) you're stuck with One Speed.
This means when management wants production rates increased, you need either a new pump or motor, rather than changing a belt drive or gear reducer ratio.
In slurry pumps, any shock loading (cavitation, rocks, grinding balls) is transmitted directly to the motor armature and bearings, along with any dynamic axial end play in the pump shaft.
Alignment is more critical as well, since misalignment loads are transmitted to both pump and motor bearings.
The net floor area is about the same for both installations, but the less expensive motors will leave room in the budget for a spare.
Keep the wheels on the ground
Bob
showshine@aol.com

 

[electricpete]

I will offer some more comments on the subject of 2-pole vs other motors. I restrict my comments to fixed (matched) speed applications.

#1 - As I said, a disproportionate number of the problem applications at our plant (in terms of vibration and bearings failures) are 2-pole machines.
#2 - From conversations with machine condition-monitoring professionals at reliability-magazine.com I have heard they have similar experience and I have heard that some plants require 1800rpm or below for this reason.

Yes of course it is a combination of design/maintenance/operations factors. But the cards are stacked against you in a 2-pole machine for a number of reasons.
#1 - Let's say I meet a specific ISO balance standard where the allowable imbalance is proportional to a multiple of W/N where W is rotor weight and N is machine speed. It sounds like I am keeping the forces on 2-pole machine comparable to 4-pole machine, right? Wrong. A 2-pole machine balanced to ISO G1.0 has 1/2 the unbalance (inch-ounce) but twice the unbalance force as compared to 4-pole. The limit is inversely proportional to speed, but centrifugal force means the resulting force is proportional to speed squared. So in selecting bearings... we not only need to account for higher number of revolutions but higher force.

#2 - Alignment.... easy to say that properly aligned machines work great. But there are machines that will give you problems. Thermal growth is one of many challenges. And the equipement/techniques used have a limited resolution. Faster machines demand tighter alignment tolerance..... but typical maintenance mechanic has only one set of alignment tools applied roughly equally to all machines regardless of speed.

OK, off my soapbox. Now you know my opinion. You are welcome to your own.

 

[electricpete]

I forgot one other factor unique to 2-pole motors.... extreme sensitivity to vibration from unbalanced magnetic pull (due to assymetric air gap) and from soft foot problems.

Regarding unbalanced magnetic pull, it contains an ac and a dc component which can be on the order of 16 times as high for 2-pole motor as for 4-pole (ratio between 4-pole and 8 pole is much lower). Some will tell you that electrical-related vibration is not as severe as other types.... partly true... I certainly would not get as excited about seeing 0.1ips of 2*LF as 0.1ips of BPOR.... but it is still indication of a force seen on the bearings.

 

[Texaco]

Motor and pump life are also very much related to lubrication. A study at Texas A&M and literature from SKF both say that bearings lubricated with oil mist will achieve significant longer lives. SKF says that the L10 live can be improved by a factor of 6.

 

[Kei]

Hello guys.
I just joined this site.

When we decide the pump speed and size, the cavitation should be considered.
Large suction diameter and slow impeller speed are good for preventing cavitation. The large diameter impeller causes high speed on the tip, so we need to reduce rotation speed.

 

[vanstoja]

See also thread407-76076 on 4 pole vs 2 pole pump speeds which had 40 posts starting 10/22/03, ie, about the same time this thread was resurrected after a 7-month hiatus.
This one addresses motor life rather than pump life but they merge on the subject of bearings effects on machine life whether pump or motor. When bearings enter into the discussion of pump or motor life one should make a distinction between rolling element and fluid film bearings since the latter kind cause most of the rotordymaics problems associated with driveshaft speed. For motor electrical failures, which are more in the nature of "life-ending" than bearing failures, the critical concern is mostly motor winding insulation temperatures with respect to the insulation Class ratings as epitomized by the plus 10 degree C rule of thumb for halving of insulation life.