Diesel Driven Vertical Turbine Pump (875HP)

[TheCairn]

We are going to install a vertical turbine pump (14,000 USgpm, 180 ft) driven by a diesel engine (Cat, V12, 875 HP, C27) as a backup effluent pump. As far as the mechanical drive goes, there will be a coupling on the engine, a ~5' long drive shaft and then an angle drive on top of and coupled to the pump. Pump suction will be ~18ft below surface. Engine will turn at 1800 rpm, pump at 1200 rpm. I am wondering about coupling selection and whether I need to be doing any sort of dynamic analysis (polar moment of interia, spring equivalent, etc) or is this a standard application that can be done with selection of a decent coupling?

If required, do engine suppliers offer this analysis?

Also, I have been told that a clutch is not typically used for this type of application. Comments?

SDS, P.Eng.
Canada

 

[Artisi]

Is the drive shaft a rigid shaft or a spicer universal jointed shaft?
Clutch or no clutch would be your decision, however a clutch gives you the flexibility of running the engine for maintenance etc.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[nondimensional]

In the US, if the engine is the prime mover then the engine will have to rated for Tier IV emissions. Check with local emissions requirements.

I would start with dynamic lateral and torsional analysis for the whole pping unit as a requirement since there are too many moving parts. A rubber boot or tire type coupling would be a good start. See Hydraulic Institute document on Couplings. It gives some good information on coupling types and applicability.

 

[JJPellin]

We have a number of installations like this. Basically, every cooling tower has a diesel driven emergency back-up pump. Most of them use drive shafts with universal joints. Some of them use other types of couplings. They all have angle drives and vertical turbine pumps. The manufacturers of this equipment build these all the time, all over the world. This is not a particularly unusual or innovative machine train. The drive shaft or coupling manufacturer could quickly answer the question. Is this a plain-Jane, run-of-the-mill application for you and no additional analysis is needed? Or, is this an unusual application for you that needs additional analysis? I would be very surprised if they didn't choose the first option.

If the pump manufacturer built this as a package and sold it to you, they can answer the question. If you are buying the individual parts and assembling them yourself, then the risk is greater and the coupling manufacturer can give you guidance.

Johnny Pellin

 

[TheCairn]

At this point neither the drive shaft nor the coupling are selected. I suspect we would choose the rigid shaft as it would be simpler. While a universal shaft may allow us to move the engine down toward ground level, if we moved it all the way down we would be talking about a 33" drop over 60" between engine and gearbox. Seems pretty steep to me and I'm thinking this would add a pulse to the speed/torque. Yes/No? Is this a problem? Maybe not in the context of the engine pulse. If I elevate the engine and align to the shaft line, the universal is not a great choice as lubrication at the u-joints becomes an issue.

Thanks for the comment on the clutch. Something for us to think about. I guess we could always drop the shaft or coupling if we needed to run it for maintenance. This should not be often as it will typically accumulate about 20 hours per year during weekly test run and under normal circumstances the engine can be run with the pump on. Side note: I plan to install a standby lube pump and coolant heater.

Emissions requirements are met.

nondimensional: Thanks on the Hydraulic Institute document. Any links or references you can suggest for digging into torsional and lateral dynamic analysis?

 

[nondimensional]

Are the engine and the right angle gear on the same baseplate with machined mounting surfaces? If not, I suggest you go with a flexible coupling instead of rigid shaft connection.

For lateral and torsional analysis, HI 9.6.8 is a great start. Also, see API 610 section 6.9. With so many rotating parts and various high frequencies involved - gear mesh, vane pass, bearings, engine-pulsations, etc. I would suggest going through a lateral and torsional analysis as a requirement in your specification. Put the onus on the system supplier.

 

[EdStainless]

Are the speeds actually 1800/1200?
Often with engine driven pumps the speeds are actually odd ratios in order to make sure that you don't aggravate resonant frequency issues.
I would ask about vibration and critical frequencies.
Often there are also issues when a pump takes a long time to come up to speed and spends too long at 'wrong' speeds.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[TheCairn]

non-d: original plan was to mount pump separate however we could put it on the same skid structure as the engine if there are benefits. And just so we are clear, I may have used the term rigid shaft incorrectly, the shaft end connections would still have to accept some misalignment and axial delta however not as much as a Spicer style.

Ed: The pump spec'd would turn at 1200. The engine produces a constant peak 875 hp from 1800 to max rpm at 2100. Ratios available from the angle drive maker that fit these constraints are 1.5 and 1.75. People had the opinion that it was better to run the engine at the lower rpm for engine longevity which would give a 1.5 ratio. This opinion is not necessarily correct and if you have a different perspective, please let me know. I thought that it might be better to have the higher reduction as it would require less torque from the engine including lower engine starting torque.

 

[Artisi]

Johnny P has summed it up for you, don't overthink what is a very normal installation.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[TheCairn]

Artsi: Thanks for reeling me in. I'm tempted to agree with you and Johnny P. However API 600 recommends at least an undamped natural freq analysis for pumps driven by diesel engine >335hp. We are considering building this ourselves so I'm trying to gauge how deep the rabbit hole might be. In real terms the risk of just building it and having a vibration issue to fix after the fact is low as this is a backup machine and there would be lots of opportunity to make changes. However it never looks good and as they say around here "you never want to be famous".

I'll go lean on the coupling supplier as JP suggests.

 

[Artisi]

Correctly sized Spicer# universal jointed shaft or similar (not sure what's available in your market)is the way to go, just ensure the pump / driver are off-set to each other and the joint angle each end is the same - problem fixed.




It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Artisi]

Attached a photo from about 30+ years back of a 36" inclined axial flow irrigation pump, diesel driven with a double joint universal shaft, this is installed 100 miles up the Musi River in Sumatra, Indonesia, dragged off a barge by a bulldozer, installed by bulldozer and unskilled labour - no API 600 involved, or vibration analysis, no problems - could still be running for all I know. .

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[nondimensional]

Artisi,
With all due respect, I have just one word - Boeing. They all worked, until they didn't.

 

[CouplingGuru]

This doesn't sound like a very complicated application. I am sure your coupling supplier will be able to provide a solution for you once you provide them the application specifics.

When it comes to couplings we are always here to help.

http://WWW.PSCCOUPLINGS.COM