4 pole versus 2 pole 4

[HomeMadeSin]

Back when I was handling contract orders for IDP, a lot of the specifications called out for 1800 rpm (4 pole) maximum speed. Inevitably, sales and purchasing always worked around that spec. for the sake of savings, but I was wondering if that has changed at all. Although the pump size would be larger for slower speeds (typically), the bearings, seals and such would last much longer bringing down the operating and maintenance costs.

Any thoughts?

 

[stanier]

On the contrary the whole of life cost may be reduced because of use of a two pole speed pump. The componenets aree generally smaller and thus cost less to replace. The key is where you are located. ie cost of labour.

Two pole speed pumps are invariably more efficient than four pole so power savings accrue.

No one procrastinated their way to the top

 

[electricpete]

I have heard a variety of opinions on this subject. Each person's view is shaped by their own experience.

My job is monitoring motors and assisting in motor maintance and this experience has given me a strong opinion.

2-pole machines are probably 33% of our machines but 80% of our rotating equipment problems.

2-pole machines may save money in initial cost but will be much more painful to maintain.

 

[stanier]

It also depends on the quality of the 2 pole machine in the first place. A good 2 pole machine will out last an el cheapo 4 pole time and again.

No one procrastinated their way to the top

 

[HomeMadeSin]

stanier: sounds like your biased to 2 pole. Also, a good 4 pole machine will out last a cheapo 2 pole anyday.

You'll have to help me understand your comment about 2 pole being invariably more efficient than 4 pole. It really boils down to the pump design - and since the initial price affects the desire to design primarily for 2 pole, your point maybe empirically true. However, if pumps were designed for 4 pole speeds to begin with, that would take that out of the equation.

Having done a brief stint with preventative maintenance, I agree with electricpete. I've done enough vibration surveys to know that higher speeds require more work.

 

[stanier]

Actually I am not biased towars 2 pole at all. However I have dealt with many public authorities over the years who blindly insist on 4 pole speeds. When a 2 pole pump gives the best engineering solution it should be used. there is just a lot of ignorance on the part of specifiers as to why a pump should be 4 or 2 pole.

In respect of maintenance is a multistage four pole pump less of a head ache compared to a single stage 2 pole?

I have used pumps running at 15000 rpm on specialist gas applications. Vibration is not an issue there.



No one procrastinated their way to the top

 

[HomeMadeSin]

stanier: fair comments. Each application should be looked at individually. The opposite to your multistage 4 pole vs single 2 pole example is when customers pick a single 2 pole over two 4 pole pumps - blindly going for 1 pump than 2 w/o regard to operating efficiency at the required flow and pressure.

As far as high speed pumps, they need to be balanced to a gnat's .....well skip that part. I don't know for sure, but I would guess the rotating assembly would have to be balanced to ISO 1940 G1.0 at least. If not, I wouldn't stand next to it.

The point I'm trying to get at is with most companies feeling the economic pinch, the feasibility (at least in the bean counter's eye) to have highly specialized maintenance people is nil. You couldn't let a high speed pump (like the one you mention) slip in terms of maintenance. Speed does account for something. The same mechanical seal rotating at 1750 will fail sooner than if run at 3450 (all else equal!). Same with bearings. Most manufacturers design a pump for a given speed (3450) and also offer it at 1750. The seal cavity doesn't change, the bearings don't change, etc - because it isn't economically feasible to have a pattern for each speed.

 

[electricpete]

Some thoughts on speed as relates to reliability.

As was mentioned, higher speed machines have tighter unbalance tolerances. A familiar standard is 4W/N. Iso standards are also proportional to some multiple of W/N. So we restrict unbalance by 1/Speed = 1/N and everything is fine, right? Wrong.

Unbalance force goes with N^2 =Speed^2. If we reduce unbalance by (1/N) we still have unbalance force increasing in proportion to N.

We're not done yet. Rolling bearing life-time in revolutions is proportional to (Rating/Load)^3. If load goes with N then Rolling bearing life-time in revolutions goes with 1/(N)^3.

We're not done yet. That was rolling bearing life-time in revolutions. Rolling bearing lifetime in YEARS is propr to (1/N) times rolling bearing lifetime in revolutions.

Rolling bearing lifetime in years goes with (1/N)^4.

If we don't change bearings, and we double the speed and reduce the unbalance per standards, our lifetime goes to (1/2)^4 = 1/16.

OK, not fair... different bearings would be selected for high-speed machine. That is correct. Still there are economic pressures and I don't think you will find that high-speed machines are generally overdesigned enough to compensate for those higher stresses.

One more item: Your mechanics have the same tools to align the 4-pole machine as they do the 2-pole machine. In both cases they likely will expend effort to get it about as close as they can and then stop. A good program will specify tighter tolerances for the high-speed machines but it is not likely achievble using same tools.

Maybe I have overstated or mis-stated some of these points. People who live with the machines for the long-term see the difference. People who specify the machines never do.

High-speed machines above 3600rpm are a different ball game.

 

[electricpete]

Once again I should mention I was just expressing an opinion. No disrespect intended for anyone who carries different opinions.

If you want perspective of people who monitor/maintain equipment on 2-pole vs 4-pole, please post a question at

http://209.15.74.79/ubb2000/cgi/for...ration+Forum&number=2&DaysPrune=20&LastLogin=

It is a very good forum populated by vibration analysts and rotating equipment engineers.

 

[electricpete]

I have been thinking about it and there is one minor detail I overlooked in my discussion that makes a huge difference in the quantitative evaluation of speed upon rolling bearing life.

I have assumed that unbalance is the only load acting upon the bearing. In fact, if we use ISO G1.0 tolerance on 3600rpm machine, the unbalance force is less than 4% of the weight force.

Either you guys are too uninterested (in my long discussion), or way too polite (not pointing out my error).

Anyway, regardless of my misguided attempts at quantitative calcualtion of the effects of speed, my experience remains the same as reported above.

 

[vanstoja]

Power magazine's THIS MONTH'S PROBLEM in Sept. 1991 (pgs. 83-84) was "Should we specify lower speeds for centrifugal pumps?" with emphasis on 1800 RPM vs 3600 RPM. Among the 5 answers one said, "we randomly sampled 50 pumps each running at 1750 and 3600 RPM. Each speed group had pumps of identical manufacturers, models and sizes...result of our findings was that a 1750 RPM pump will run 2.54 times as long as a 3600 RPM pump." Another contributor noted that "once the 1800-2000 RPM range is exceeded, the wear rate for both impellers and housings increases rapidly because of cavitation and particle impingement." Another stated that "Vibration is more critical on a 3600 RPM pump than on an 1800 RPM pump." The 4th guy called it a draw and the 5th was all for 3600 RPM based on size and weight reductions when properly selected.
My opinion is that 1800 RPM pumps are apt to be less troublesome if they do not have to use (1) fluid film radial/thrust bearings and/or water-lubricated mechanical face seals. Fluid film minimum thicknesses in bearings and mechanical seals become much more critical to overall reliability at lower running speeds. If the pumps are operated in pressurized systems or with ample NPSHR margins over NPSHA, then cavitation and erosion problems are practically nonexistent and 1800 or 3600 RPM pumps should be equally reliable. If noise or vibration transmission are critical to the application, then there is definite advantage for lower speeds since pumped fluid broadband excitation is a cubic function of running speed (determined by pumping power) and 18 dB reductions in flow-related noise/vibration accompany each halving of pump running speed.
In short, the paramount consideration for pump speed selection should always be the reliability and functional needs of the particular application.

 

[RotaryG]

Many a times one does not have a choice of number of poles. Single stage pump speed is dictated more by the operating conditions than anything else. For large multi-stage pumps, lower speed can also mean more number of stages and problems with rotordynamics.

Each application requires its own consideration. No generalities can be given.


Regards,

Guru

"Evolution rather than Revolution"

 

[HomeMadeSin]

electricpete: I was out of commission this past with a broken collarbone (motorcycling) and didn't bother to check the board. I appreciate the repsponse and will try to get some thoughts into.

Sounds like there is a counter-point to every point, but i would be willing to bet that there is SOMETHING to it, which triggered at least an article (re: vanstoja).

However, it doesn't sound as if there has affected much in the way of pump selection these days.

 

[piffer]

HomeMadeSin,
I just read through all of the responses and had to go back to your original question to remember what you asked. It is my experience that the bottom line to reducing expenses leads back to efficiency. In the wastewater industry (pumping), energy usage can be many times more than initial cost or repair cost(s). The combined hydraulic and motor efficiencies are significant factors. On the subject of 2-Pole vs. 4-Pole, the higher speed machines tend to have more frequent service intervals (as several respnders have mentioned).
Regards,
Piffer

 

[stanier]

So the answer is to do a life cycle costing. refer

http://www.pumpzone.com

for a model.

If you work in the third world where labour is 1USD/day and all imported components have high duties and taxes the lower capex cost two pole version is a given.

In downtown New York with sky high contract labour costs and local spares then the four pole would be the answer.

You may also like to consider that your maintenance people are there anyway so geting them to do something may not actually cost you anymore.

These are the extremes I know but need to be looked at.

 

[HomeMadeSin]

I just talked with our primary motor vendor and he agreed strongly that 4 pole equipment (or more appropriately, slower equipment) outlasts 2 pole equipment. While that is a broadcast statement, I think it is one taken from years of experience (read warranty, after-market, etc).

Now as stanier points out, if you have maintenance people anyway, so what right? Well after doing a short stint at a paper mill, I've seen what happens to a maintenance crews over a period of cutbacks and lean times. Joe Maintenance not only turns shafts and performs maintenance on equipment, but has to jump on the paper machine when the thread goes down. Companies don't have as many Joe's as they once did.

Also, RotaryGuru, your point is also what I'm looking to convey. If the conditions of service were say 100 gpm at 100 feet, a quick look at Goulds (just picking the closest catalog)3196 pump could either be a 1-1.5x6 at 3450 rpm with 5 hp motor or 2-3x10 at 1750 with a 5 hp motor. Just an arbitrary point. Initial cost factors would obviously favor the 6" pump as the smaller (1.375" versus 1.875" seal, coupling and baseplate. However, what is the L10 of the smaller bearing at 3450 rpm, which is lubed (or greased) in a smaller bearing housing. How are the axial loads and radial loads? How does the shaft deflection affect seal life?

It would be interesting to see if any lifecycle studies have been done to compare the differences. cost versus cost, operating and purchase. And I definately agree that using a pump as close to BEP as possible will outweigh all other considerations. Two pumps in series often beat 1 high speed alone in hitting the sweet spot.

 

[stanier]

In answer to Homemadesin's original question. Dont let Procurement make the decision! The decision should be made by Engineering.

its too complex an issue for the clerks of the world to decide. They have a conflict of interestin any event. They happy buying anything as long as its under budget and they get a good lunch out of it.

 

[HomeMadeSin]

stanier: I'm with you on that one. However, in more cases than not, beancounters rule companies.....how many engineers are CEOs?

 

[stanier]

HomemadeSin,

This is a point made by many engineers but I believe that they bring it on themselves. If the engineer has any balls he will dump the whole project file on the procurement officers desk and tell him to buy what he wants and then walk away. Beancounters are just like school bulliees. Once yopu stand up to them they shrink away, the cowards that they are.

We have to remember that they have no technical skills. When push comes to shove they are smart enough to realise that they cant do it.

Simple rules:-

Dont compromise your engineering integrity
Dont sign off anything you dont believe is right
Walk away from a company that doesnt have ethics
Dont live beyond your means so that you're financially compromised before you start

If all else fails tell the client whats happening and ask him for a job. Its surprising how many clients will engage an engineer who has the guts and ethics to walk away because of what he believes in. Then you can stick to the procurement guys from the other side of the fence.

 

[RotaryG]

For the pumps of the type described by HomeMadeSin in his Nov5th posting differences in L10 life of the bearings will be statistically insignificant. This discussion is just academic and does not add any value to the practical aspects of how to select right pumps for a given duty. Even assuming that 5HP, 4 pole motor driven pump bearings are significantly better than the 2 pole versions, it does not translate in to significant savings to justify 4 pole pumps of this type. Note that most pumps do have installed spares.

The moment operating conditions become a little bit more respectful than indicated by HomeMadeSin, they would on their own start dictating the speeds – neither the engineers nor the bean counters



Regards,

Guru

"Evolution rather than Revolution"