Increasing Pump Speed

[smCHe]

Hi All,

I'm currently checking the possibility of increasing the plant capacity because we managed to secure additional feed. The circulation pump (from KSB RPH 150-630) for propylene seems to be the main limitation. I checked the pump and the installed impeller is almost the casing max. so not much improvement in capacity, also the motor is too small (5% design margin only and we already pushed it above its original design). I came up with the following possibilities

1. running two pump in parallel, this reduces plant avalibility but is acceptable as we can manage our plant load & pump PM checks. I'm yet to try it in practice and have a concern because the pump suction velocity is > 7.5 ft/sec and I'm used to max of 6 ft/sec and now its running 4.5 ft/sec. the NPSHA looks ok,

2. increase the pump speed and install a VSD motor. The pump design speed is 1790 rpm and I would need to go to 2300 rpm or so. the pump vendor did not recommend this approach and personally I never done such a thing, so my gut feeling is to drop this option

3. buy a new pump/motor, this is not a good option because of the cost obviously, need to shutdown the plant for such modification and the equipment lead time.

Any thoughts on options 1 & 2 ??

regards,
smCHe

 

[Pumpsonly]

From the required speed increase for possibility No 2 mentioned, it seems you are trying to increase the flow rate by almost 30%
Do not forget that when you are pushing higher flow through the pipe line, you also need more head / pressure from the pump.
Can your present pump provide you the new head / pressure without changes when running in parallel.

 

[smCHe]

Hi Pumpsonly,

yes, this is option-1, running the pump and spare in parallel. have not tried it but according to my analysis it should do. the suction speed at the pump inlet is a tad high at 7.5 f/sec thu

Running higher pump speed by installing a VSD ( ~ 200 hp) means I have a full pump spare, unlike operating two pumps where i have none

regards,
smCHe

 

[ElSid1]

Is this a closed system?
Pumping propylene increases head, flow, and HP. Be sure to account for all 3 changes.
VFD have their own issues. Do not use it as a crutch as mentioned above. Running the pumps in parrallel will increase flow. This depends on the pump curves that, hopefully, have been adjusted for the plants concentration of propylene.
Example:

http://www.fluidh.com/downloads/Winter2001.pdf

I make all the new engineers pay close attention to this example . Too many pumping systems are not properly adjusted as the designer just assumes water.

 

[smCHe]

hi ElSid1,

the pumps are both in propylene service, I just need to increase the flow in the same system.

today i talked to operations, will try running both pumps and check actual operation.

regards,
sayed

 

[smCHe]

just to update you guys, tried running the two pumps in parallel and things went smooth,

now I need to check on on VSD motor to see the feasibility of that option!

cheers,

 

[stanier]

you could add a new 2 pole motor and use the VSD to operate at the desired speed. The pump vendor may not want this as it does him out of a sale of a new pump. Check the vendor data and see if they offer that pump frame size with a two pole motor.

Cable to motor would need to be changed for a screened cable to avoid interference with other instrumentation.

“The beautiful thing about learning is that no one can take it away from you.”
---B.B. King

http://waterhammer.hopout.com.au/

 

[1gibson]

Faster motor on a VFD without approval from the pump vendor is just money in the bank for the parts department. You'll spend more on repair/replacement than a new unit. There are a number of things to check like shaft hp, MAWP, NPSH; some may affect materials of construction, some may affect production requirements (impeller balance grade.) If the pump isn't engineered correctly for the higher speed and you run it anyway, pump vendor's parts department might start sending you christmas cards.

Vendor said no, try to get all of the reasons. Obviously the warranty will be out the window (may be out of warranty coverage anyway) but try to get the customer service dept to explain all the concerns. Then proceed with caution if you are comfortable. There is probably a lot more to it than the pump vendor wanting a new OEM sale, trust me they'll make more on repairs.

 

[MikeHalloran]

Consider adding a third identical-ish pump to get back your plant availability?





Mike Halloran
Pembroke Pines, FL, USA

 

[VLTPumpControl]

Making a pump run faster with a VFD is simple, but as mentioned by several here already, there could be down sides to it. If you run two pumps in parallel and apply VFD to both of them, you would reduce the wear and tear on them for the periods, where you do not need the full capacity.

I always recommend to use master/follower control, it provides the best control, highest energy efficiency in the system and secures some level of redundancy. You can set the systems to automatically change lead pump or change setup in emergencies, so if one pump fails, the other pump can run at over speed until a repair is completed (prefereably in a separate setup to make this an emergency event only). If you want details on how this could be done, here's the link to our instruction manual. This does not include the emergency operation mode, but if you want to explore this, I am sure we have someone in the area, who can help you.

Since you reduce the stress on the pumps during normal operation, the risk of damages, when you are in hi-speed operation should be less. There is however a mechanical limit, where all you do is create cavitation in the pump because you are trying to move liquid faster than what the pump can physically move through the outlet or back-pressure in the system will allow.

You really need to check carefully how your pressure drops in your piping, valves and fittings change with the higher flow rate, to make sure you do not create cavitation in the pump. If pressure drops increase too much, you will have to upgrade the system, since the pumps cannot fix that.

Also the seal on the pump could be a problem. They do naturally have a safety margin, but pushing +30% more speed on the shaft and at higher pressure might not be advisable. You could check if a standard seal is used and contact the seal manufacturer directly, since they will not try to sell you a new pump

 

[ccfowler]

smCHe,

It seems that you are getting into a very messy problem. You may be able to make some relatively minor changes in the piping system that could relieve much of the flow velocity and pressure drop problems. I would pay extra attention to the pump suction piping syste making sure that each pump's suction line is as long, large, and straight as possible.

You may find that replacing only a portion of the discharge piping system with larger pipe, including gentle transitions between pipe sizes, may be sufficient to allow less costly pump change options.

In my experience, pump manufacturers are very helpful and can provide insight from their experience base, but the usefulness of their help depends on you providing very complete and accurate information including abnormal conditions that may reasonably be expected. The more thoroughly you understand your system and the potential revisions to your system, the better the outcome of your conferring with the pump manufacturer. The worst mistake you can make is withholding information from the pump manufacturer. The pump manufacturer will want to make money on this case, but they want to secure a better position for future business by having a well satisfied customer.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[Artisi]

smCHe,

2. increase the pump speed and install a VSD motor. The pump design speed is 1790 rpm and I would need to go to 2300 rpm or so. the pump vendor did not recommend this approach and personally I never done such a thing, so my gut feeling is to drop this option.

Did you ask the vendor why - also, is the vendor the pump manufacturer?



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.)

 

[smCHe]

Hi All,

Let me share more data on the VSD, the options are

1. increase one pump speed and take all the risk ! the vendor simply will not give much info on why the speed increase is risky, just stating the pump is not designed for that. This option is not wise but I have to share it with our guys as to give the full picture

2. reduce the pump(S) speed and run two and only increase one if other fails as VLTPumpControl indicated

3. increase one pump speed and install larger impeller size as to reduce the speed increase (still the shaft load and other things would be an issue)

of course, installing new pumps is an option but its not simple and very costly, this is why I'm checking all others as a feasibility study. The cost is not just a new pump/motor, but downtime cost as well.

I personally dont have VSD experiences in motors, any help on vendors and types is very much appreciated.

regards,
sayed

 

[TenPenny]

Typically, and I stress that I don't know if that is the case here, the issue with increasing the speed is that the loads on the bearings increases, and they may not be designed for that extra load. Really, only the manufacturer can say.

 

[1gibson]

If the pump vendor is not helpful (they are being lazy) then use a different approach.

"Hello, Mr. Salesman, I am interested in some new pumps. Here are my datasheets for new conditions. They must match all original dimensions and materials as my existing (serial number), unless any material upgrades are required for the new conditions. Please get me a budgetary quote for new units, but with complete technical details so we might evaluate possibility of retrofitting original units with any upgraded components for shorter overall lead time."

 

[Artisi]

smCHe: "the vendor simply will not give much info on why the speed increase is risky, just stating the pump is not designed for that" Is this the manufacturer or a re-seller making the comment, if a reseller, can you by-pass him and ask the manufacturer.

Anyway, without having all the facts and looking at the KSB pump curves, I would suggest that KSB RPH 150-630 is not capable of operating at 2300rpm, of course I could well be wrong.
If this is a major plant upgrade expected to operate trouble free long term, change out the pumps for correctly selected units capable of the duty. At the end of the day, probably cheaper in the long run and will sleep comfortably at night.

3. increase one pump speed and install larger impeller size as to reduce the speed increase (still the shaft load and other things would be an issue) What impeller diameter is fitted now, if full dia. how can you increase it?

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.)

 

[DubMac]

If you put these pumps in at 2300rpm without the blessing of KSB, you will not sleep well in the future. If something blows up, all fingers will point to you; and probably with good reason.
As Artisi was alluding to, don't take the word of a distributor or reseller; make sure you get word, in writing, directly from the manufacturer's Engineering Dept.