API Horizontal Split vs Horizontal Ring Section (stacked) Pump

[pkhenry]

I am looking at purchasing a 1600'TDH 350 USGPM flow pump. The quotes cover horizontal split, at a much higher initial capital cost and horizontal stacked pumps. I understand the stacked units are higher in maintenance down time which may off set the initial capital. Are there any other issues I should be aware off. Basically the units are coming in the same regarding trim, # of stages and material. It seems only an issue of casing style.

 

[BillBirch]

More potential leak paths with the stacked casing? Also by nature of stacking, the "build-up+ dimensional tolerances would be additive, necessitating greater clearances and lesser performance.

 

[quark]

Phenry,
One basic thing is for split casing pumps thrust load on the bearing is nullified by entering liquid frm two sides. Moreover the efficiency of split casing pump is higher than even back pull out type pumps.If budget permits it should be the option. The efficiency of multistage is poor because of losses at each stage.

 

[tstead]

If I offend anyone with this post, my apologies but two of the above posts hit on pet peeves of mine and I have to speak up.

If you purchase a low-quality horizontal ring section pump, then Bill might have a point but a good quality pump from a company that has been building ring section pumps really is nothing to be scared about. I've seen ring section pumps in service for decades with no more regular maintenance than other standard pumps. Remember, tolerances are plus OR MINUS and a plus tolerance on one stage may balance out with a negative tolerance on the next. With ring section pumps that do not use O-rings or gaskets between the stages, this tolerance is very, very tight and you do not have those problems.

Be careful to generalize all pumps into a perceived pitfall because you have to look at each pump seperately. Also, efficiency doesn't mean SQUAT - that should NOT be your deciding factor in which pump to buy. What really matters is how much it will cost you to operate the pump. Sure, on paper a pump with higher efficiency equates to lower operating costs AT THE DUTY POINT but you have to look at the OVERALL cost of owning the pump and decide if a higher efficiency is worth the extra cost.

If you compare two different pumps, they will undoubtably have different Q-H curves. It is possible, and I have seen it, where pump A is more efficient at the duty point than pump B but pump B actually costs LESS to operate than pump A because of where they will operate the pumps. This can happen if the duty point is to the right of BEP and the oeprating range is further to the right and efficiency drops off quickly with pump A. Pump B would have a lower efficiency at the duty point but operates to the left of BEP and consequently has an increase in efficiency as flow increases.

While you will probably have around a 200hp motor on this pump, the cost of operation may be significantly different between two competing pumps - but don't take someone's tag line about efficiency as Gospel - put it into real dollars because that is how we pay our bills.

Also, ring section pumps use balancing discs or drums which reduce the load on the thrust bearing. If you have an adequately-sized thrust bearing then where is the issue? Split-case pumps tend to have large rotors which loads the radial bearings. So, where do you draw the line? B10 bearing lifetime calculations. If the bearings have the same B10 lifetime than chances are that there would be no inherent benefit of one or the other (from a reliability standpoint).

My point is to analyze each pump against the other and not rely on over-simplified generalizations about this design or that. Look at the pumps themselves, ask specific questions, and compare them apples to apples.

Regards,
Tim Steadham, P.E.

 

[quark]

Tstead,
I agree with you in principle. This response is with sheer academic interest and not personal. So I request back your comments.

Pumps can be tailor made now a days(atleast impellers) and almost any duty point can be brought into the best efficiency range. But is it possible with multistage pumps?
Further I have always found that actual duty point is always deviating from design duty point because of design redundancy.

In this context what is the general practice you apply when comparing pumps of two different kinds?

Your response will be of very much help to me.

Regards,

 

[tstead]

Without re-designing the impeller geometry and vane profile, the only way that you will hit the BEP is out of sheer luck.

You start with a duty point and look for a pump whose published performance will meet what you want. You may have to trim the impeller a little bit, but you cannot always hit the BEP all the time. Moreover, the only (or best) pump that you can find to do the job may be operating to the left of BEP (preferred) or to the right - it's just a matter of the luck of the draw.

Of course, if one wants to redesign the impeller and match it around a certain duty point (which is almost never done, but I have done it a few times - safety-related nuclear pumps come to mind).

So, what you have are designs that you are stuck with where the only thing you can practically change is the diameter of the impeller and a little backfiling/polishing to improve efficiency and head (respectively).

With multistage pumps, there are various options. The physics is still the same (except most use diffusers and not volutes, but I digress) and so you are still left with a crap shoot on where you are in relation to BEP. You can trim impellers in a multistage pump but not nearly as much as a volute pump because of the diffuser. So, if your duty point falls in the range of BEP of a multistage pump, you just lucked out like it's horizontal cousin.

Your point may be the trimmability of a multistage vs. single stage pump. As discussed above, multistage pumps can only be trimmed a little bit because the impeller tip/diffuser clearances will become too wide and you will just kill performance and reliability. In that respect, multistage pumps are more limited and unless you luck out on the duty point, you are kind of stuck. So what do you do? You design and offer wider ranges of impeller hydraulics for a given pump size.

While horizontal pumps generally have just one or two impellers that can go in the casing (hydraulic design) because the pump mfgr has decided not to go too wild in hydraulic choices there, multistage pumps do sometimes offer 2, 3 or even 4 impeller designs for any given pump size. In this manner, you have a wider range of BEPs than a single hydraulic so you open yourself up a little more.

This is why you have to look at each pump seperately from a generalized case. While some mfgrs may offer only 1 or 2 hydraulics for a certain pump size, others may ofer more.

As to actual duty point - it never operates at design because of what you said. System designers come up with a calculated TDH from friction loss and static height. The static height is pretty cut and dry but unless your fluid is laminar (which it never is except in a few esoteric cases) the friction loss is just a guess - a SWAG if you will. To account for future pipe corrosion and degradation, a little more head is often included in the TDH calc - usually 20% or so of the friction loss. There is no need to add 20% to 5000 ft of TDH when 4975 of those feet are to overcome boiler/reactor pressure.

Because the TDH is overestimated, and this is anticipated, the desire to have a pump whose duty point is to the left of BEP is the norm. That way, when the pump runs down its curve to the actual operating point, it is moving closer to BEP and not away from it.

I hope this answers your question. Please pardon any spelling errors - it's late and I have to go home

Regards,
Tim Steadham, P.E.

 

[quark]

Tstead,
Thank you very much for your prompt and satisfying reply. Till times I was also not knowing about the fundamentals of pumping operation to the basics. Last year casually I had an oppurtunity to check the performance of the pumps in our factory, despite it's not my daily work.

To my own estanishmnet(a little bit exaggeration, isn't it?) I saved the energy wastage by 3 times my annual salary by just spending 1/10 th of monthly salary on trimming of the impellers. Till then I was very eager to study each and every basic of the pump. Your comments are very useful.
(Ironically, I have a dislike for Split Casing Pumps reason not knowing so far, but lot many experts suggest them)

Any comments pro split casing are very much invited.

Regards,

Quark