I don't know what drives the design one way or another. In my very very limited knowledge my thought is:
If it is a multi-stage pump let's say 5 stages... mechanically it would be very difficult to overhung and therefore usually between bearings. For single stage pumps I have seen both between-bearings double suction, and overhung.
=====================================
Eng-tips forums: The best place on the web for engineering discussions.
I think electricpete is pretty close to the point in his statement - but I would add, that it is usually much cheaper and easier to manufacture a "back pullout" than a split case pump when it's a single impeller unit.
If you only need a single stage, there's no need to go to a between bearings pump. Actually, I'm not sure if anyone even still makes between bearing single stage pumps. Multistage pumps are built between bearings (with some exceptions like downhole pumps or vertical turbine pumps) to compensate for the longer shaft being more flexible. Cost and maintenance will be higher with between bearing pumps because you have two sets of radial bearings and two mechanical seals, and the seals are much harder to replace because you have to pull the bearings off first.
without any doubt, a back pull-out end suction single stage over-hung impeller design is the only way to go. This design is available from hundreds of maunufactures in as many materials of construction as you can imagine.
Your size of pump (35 kw), can be big trouble as an overhung unless you know what you are doing and get the right pump.
Radial Thrust
Not sure what kind of pump you are looking at, but at that 35kw kind of power you MUST be looking at either multistage type with diffusers, double volute, any pump with diffuser, or "modified concentric" to escape fast destruction of the driver bearings due to radial thrust, unless you are using pump ONLY at or near BEP.
Simple Volute overhung is a bearing killer when you pile up the following additive "crimes"
1 - Specific Speed Ns >1800 - 2000 or so.
2 - Pump operates well off BEP for long durations.
3 – High Suction Specific Speeds Nss > 9000 or so AND pump operates below BEP for long periods.
4 - Large diameter impeller, larger the diameter, worse the problem of thrust.
One crime and you get down the road a few years,
Two crimes and you get to the cliff a little sooner,
Three crimes and your shaft bearings meet their maker very soon.
Actually, I like single volute overhung for their high efficiencies, simplicity, and low cost to both purchase and service. Hard to beat all that. But if you cross thresholds resulting in fast failures due to the problems these pumps have in larger sizes, all the benefits are gone.
We will not use overhung single volutes above 20hp (15kw) at all unless we can operate pump ONLY at or near BEP.
If you use Double Volute especially, you can go a long ways with these pumps in large sizes, at least through 100 hp (75 kw).
Pumpdesigner, not sure I agree with your thinking ... I do agree with your statement they have to be built right, but 35 kW isn't that bid a pump, (mind you, I'm used to refinery installations), that's barely even a respectable ANSI pump. That said, there's nothing wrong with going to between bearings at that size (if you can find one) but I think it's overly conservative. I've put single stage overhung API 610 OH-2 pumps as large as 300 HP with no problems, and I've seen others built larger than that. Double volutes I agree with, and multistage may be required depending on the specific speed of the service, but between bearings for 35 kW sounds excessive. Even a single volute at that size shouldn't cause too much problem as long as the shaft L3/D4 factor is reasonably low.
I must also add my 2 cents worth here as well - 35 kW -some mornings I wouldn't even get out of bed for anything under 100kW.
I too have applied end suction o/hung impeller pumps upto and exceeding 300 kW. one interesting application I recall - a series of 8x6x21 pumps operating at 4 pole 50Hz with installed motors of over 200 kw on black-liquor recovery - not many applications harder than that - have been running now for more than 10 years and haven't missed a beat.
Don't forget the large sewage pumps - end suction single stage o/hung single volute 30" - 40 " are common - slurry pumps of the same sizes with 300 - 400 kW drivers - these can only be of this configuration.
In my 20+ years the number of between bearing pumps I've selected - sold or installed - which exceeded 35kW - that's not counting the "toys" ie, bore pumps - high-rise boosters etc wouldn't account of 5 % of the total - of these quite often they were specified by others or there was a very specific reason to go with between bearing pumps.
It all comes down to good design - nothing else and selecting the best economical unit for the application.
Artisi and Scipio,
I was tired when I wrote that, having some fun, but I do stand by all of it.
On your large pumps, where were they operating, at or near BEP or way off BEP?
If Specific Speed is low, single volutes do better because they exhibit less reaction to off BEP flows.
Good design and engineering you are correct about, select the pump correctly, install it correctly - no problems. "Correct" however is hard or not possible for many if they lack the knowledge. Some pumps are forgiving -some are not.
The information you guys gave is good stuff. Here is one small informal history that has taught me.
Over the past 10 years we have sold a specific pump in many applications. 30hp, single volute, overhung, close coupled, end suction. We have "noticed", actually tracked how long the motors on these pumps have lasted. Some last only a year, some 6 years and counting. In every case without exception, the back bearing failed. This confused me about the back bearing, but then I realized that on JP/JM motors, the front bearing is huge in anticipation of the shaft load, while the back bearing is very small. Radial forces vibrate the shaft and the front bearing handles it well because it is designed to, but the vibrations transmit through the shaft down to the back bearing and kill it, usually driving or wobbling it within its bearing race.
The key diagnostic is this, where the motor bearings failed early, 1-3 years, the pump was being used over a wide range of flows for long durations well off BEP.
So in this case, we feel that if the single volute pump is operated at or near BEP much of the time, pump motor bearings predictably last a long time If long duration runs well off BEP, short life.
A double volute, modified concentric volute, or pump with a diffuser is not vulnerable to this problem.
Therefore, if we have control over how the pump is used, and that use does not include long duration runs at flows far off BEP, we use the pump and like you, we receive good service no problems.
If we cannot control the pump's use, of if the pump will have to work long periods away from BEP, we go with a pump that cannot be hurt so easily.
If you're talking close coupled pumps (motor pumps) in my vernacular, I would agree that you could end up with problems as I always considered them a compromise - usually a good hydraulic end mounted on a marginal set of motor bearings.
-but then I always steered well clear of them anyway, to my mind there isn't any place for them in heavy industry - well not the industries I was in where the operators expect the pumps to run to destruction if needs be rather than shut the plant before they had time to prepare a premature shutdown.
Of course they have a place in industry - but need to be selected well, as you have said to make sure no problems down the track a bit.
Artisi,
You are correct.
I just realized that close coupled is my focus, came back here to clarify/correct myself, but alas, there you were with your years of experience faithfully to correct me.
Frame mounted overhung is different, L3/D4, bearings, everything can be played with to withstand radial forces, close coupled cannot.
I get tunnel vision, focusing only on what is in front of me.
I go home soon and let the wife humble me some more.
It was only when I reread your previous posting where you mentioned the motor bearings that the penny dropped and I realised that I was and I assume Scipio was also on a slightly different track.
When you get home - just your wife to explain the differrence NPSHa and NPSHr - this should level the playing field a little.
Artisi,
Yep, you assumed right, I wasn't thinking of that type of pump construction at all - like your experience, motor-mount impeller pumps don't show up very often in my industry, I was on a completely different track.
