×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

VFD retofitting for pumps
5

VFD retofitting for pumps

VFD retofitting for pumps

(OP)
I would like to know how difficult is it to replace the induction motor for an existing pump with VFD motor. what are the things which we have to change. Regards,

RE: VFD retofitting for pumps

It depends on a large number of items.  Is the pump close coupled or long/frame mounted?  Horizontal or vertical?

If it's a standard horizontal mount motor, long coupled, all you need to worry about is matching the frame size of the motor, and putting the coupling hub on the motor.

If it's close coupled, where the impeller is on the motor shaft, it's more involved.

More details will make for a more useful answer to your question.

RE: VFD retofitting for pumps

Why are you looking at a VFD?  Is this pump called upon to run outside of its recommended range often?

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection
http://www.trent-tube.com/contact/Tech_Assist.cfm

RE: VFD retofitting for pumps

(OP)
We need VFD because the present requirment for head is much lower than the design one. The alternative is impeller trimming where we have to trim too much which is not recommended. It is horizontal with coupling.

RE: VFD retofitting for pumps

It's not difficult.  Some induction motors can be used as they are, some may have to be replaced.  The VFD will just replace tour stsrter.

RE: VFD retofitting for pumps

What is the fluid being pumped and what is the fluid temperature?  If it is just cool water, you can simply valve it back for the same effect.

RE: VFD retofitting for pumps

2
It might be easier and less complicated to replace the whole pump unit with one that is suited to the application.

RE: VFD retofitting for pumps

(OP)
The flow is the same which is 90 m3/hr, however the required head dropped from 28 Bars to 14 Bars due some process change parameters.

RE: VFD retofitting for pumps

So then at half the head coming from a 25% speed reduction you get to a "natural" pump flowrate of 67 m3.  The problem with VFDs is that a slower speed tends to reduce head and flow too.  OK not totally unmanageable.  I don't like sticking VFDs on to anything that moves, but I have to admit that they can (at east temporarily) lend themselves to a quick-n-easy solution when system flowrates or heads have had to change for one reason or another.  It would be a better flow match to change out the pump, but it appears a VFD could be an alternative possibility if holding close to that 90 m3 is not essential.

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

If you have an isolation valve on the discharge of the pump, you can manually throttle to maintain 14 bars for an easy test.  Throttling will achieve the end result, and drop the amps the same as a VFD.  Then you will see that a simple pump control valve is all that you need, and you won't be inheriting all the problems that come along with the VFD.

RE: VFD retofitting for pumps

Can you achieve the duty needed with a speed reduction and still maintain reasonable hydraulic performance. You need to generate a number of reduce speed H/Q curves to see where the pump will operate in relation to BEP and then calculate the running costs (power consumption) over the life of the installation. You might be surprised the increase in annual running costs for a drop in efficiency of a few points.  

RE: VFD retofitting for pumps

What many people do not understand is that the power consumed will be almost exactly the same for throttling with a simple valve as it is for slowing with a VFD.  You should weigh the cost of a few points lower efficiency to the cost, maintenance, complications, downtime, and other headaches of using a VFD.  It usually turns out that a little lower efficiency cost less overall than adding the expense and side effects of a drive.

RE: VFD retofitting for pumps

True. But many people also do not understand that throttling can destroy an expensive pump (seals, shafts, wear rings, bearing failures, etc). This is a significant reduction in pressure.  And how would you throttle and maintain the same flow on a centrifugal?  

I don't buy that throttling will save you the same amount of money over time as a VSD in this application.

I would go with Artisi's suggestions. Hopefully you have room to increase impeller diameter and can slow the pump down.  But it needs to be a good fit on the variable speed curves.

RE: VFD retofitting for pumps

Here are some numbers using a Goulds 3316 3x4-11H as an example, assuming 70 degree F water and 28/14 Bar differential:

Fixed Speed at 938' TDH
400 GPM @ 938' TDH
10.75" Impeller
3550 RPM
Efficiency 66%
144 HP

Same Pump, Variable Speed @ 469' TDH
400 GPM @ 469' TDH
10.75" Impeller
2650 RPM (25% turndown, do-able)
Efficiency 68%
70 HP <--- BIG savings

No way a valve can do this.  Simple first law thermo.

RE: VFD retofitting for pumps

VSD and valve solutions each have their place.  
Keep in mind that this is a bit of an unusual case.  How many times does flow remain the same with head reducing by 50%?

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

No argument there.  I just don't agree with Valvecrazy's power consumption claim.  It is terribly inaccurate in this case.

RE: VFD retofitting for pumps

Well... he doesn't call himself "crazy" for nothing. smile

In most of the typical cases, he's not going to be too far off from his claim.  You just have to realize that VSDs arn't usually a viable solution when heads don't change in  proportion to the afinity laws.  Where they do, VSDs usually can be made to work, where they don't, valves usually work.  VSDs work best when heads and flows both change proportionally to the affinity laws, flowrates vary widely and time for each flowrate is short AND tanks are expensive, can't be accomodated or the process MUST have variable flows.  If flows and heads experience some variation, but are mostly constant, valves can work well.

In lots [not all] of the everyday typical practical applications that VSDs are being pushed to fill these days, slight benefits of VSDs can be totally eliminated by extraneous considerations and may really be disadvantageous when consideration of such things as shaft currents, adverse power supply harmonics and maintenance costs are included.

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

Each application is different.  It is irresponsible to imply that throttling will give you the same result as reducing the speed in each application.

I hesitate to apply the Affinity Laws to every pump.  In my hypothetical above, the head is reduced by half and the flow is constant. Yet the same pump can accomplish this (with the same impeller diameter) by reducing speed. The pump is just running further to the right on its curve.

Only a representative of big oil could look past the energy savings a VSD could offer in this application :) Although I agree with you completely that VSD's are NOT a fix all.

RE: VFD retofitting for pumps

I don't represent anybody except myself and prefer to arrive at the best solution to any problem by optimization of as many relavent variables as can be defined, whether its water, oil, or beer that's in the pump or the pipeline.  In this case, I have agreed that a VSD is a possibility, but do propose that the OP evaluate it as one of the possible solutions, not the only one.  I have generally found that petroleum companies also usually prefer similar solutions, unless deliverability within the project time constraints is not possible, in which case overall project economics can take precedense over what may be the relatively microeconomic factors of pumping economics.  As the net result is more energy delivered to the consumer than could ever be saved from installing the most efficient systems, there is some logic to the madness there too.

And .. valvecrazy is, I gather from past posts, not involved in big oil, but in the lower pressure water distribution applications, where his generalizations are usually more representative of typical situations and economics are much tighter, not to mention the effects of the other VSD associated operational problems that can wreck havoc for a whole lot of people in a very short time.  I don't think its irresponsible to talk in the general sense, since there can be many variables, other than cost or variables that are not convenient to reduce to cost, that can influence the ultimate selection of any one solution over another.  In that respect, he does draw attention to several factors which could have a bearing on the ultimate decision for this OP or many others.  

As for any individual situation is concerned including this OP, for any one application you can produce showing energy savings with VSDs, I could give you two scenarios where  valves would perform better in the overall project, but might even possibly use more energy in the process (maybe not), so there can be many facets to consider other than energy cost.  (I've been forced to choose electric motors as drivers when diesel would have been better just to defer the emission "cost" to the central power supplier.)  So, energy use or cost may not represent the objective for all projects and the rest is 6 of 1 and half-dozen of another, so I see the only solution is to look at your overall project requirements and make selections accordingly.  What else can one do?

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

Can I get an Amen for Biginch.

Clicketyclack is right however.  If this is an application where the flow remains the same but the head is twice what it needs to be, then a drive will reduce the power required considerably.  The drive will reduce the power required by the pump to 70 HP.  Add back in the parasitic and other losses contributed to the drive and you will be pulling 75 HP or more.  But again the drive is not saving any energy.  Because if this is the only design point, then a correctly sized pump will probably only pull slightly above 60 HP.  

So the drive is actually wasting power over what really needs to be done, which is change out the pump.  Like Biginch said,  "slight benefits of VSDs can be totally eliminated by extraneous considerations and may really be disadvantageous when consideration of such things as shaft currents, adverse power supply harmonics and maintenance costs are included".

I am glad you said that a VSD is not a fix all, because that is what is driving me crazy.  Each application is different.  What is irresponsible is to assume that VSD will always save energy.  There is nearly always a better or at least as good a way of saving energy as a VSD.  Correctly sizing the pump is the best way, and many times a simple control valve can be just as good or better.

RE: VFD retofitting for pumps

I like the way a VFD fits a long liquid pipeline. At flowrate = X and pressure drop = y, then at flowrate = 2x pressure = about 4Y.

On a pump at speed = a, flowrate = x and head = y.  At speed = 2a, then flowrate = 2x and head = 4y.  Nice fit.

RE: VFD retofitting for pumps

Right...but as (I know) you know that's only without significant static head to overcome, which is the ideal case that many specific pipelines of course don't have (see BTC pipeline with two mountain ranges to 2800 m (9186 feet) elevations.)  That can have drastic affects on the suitability of VSDs.  Boiler feeds are another of what must be many similar systems.

In fact, even VSD have alternates.  Other non electrical options.  Diesel engines are the original pipeline VSDs, although the VSD benefits were not specifically intended, just a natural result of their modus operandi.  They don't require major electrical demands, don't require 3rd party electric power furnished to often remote locations, no power supply harmonics, no shaft currents, have reasonable reliability, but true they arn't as efficient as electric drivers.  The number of those installations still in use or still being designed just goes to prove that maximum energy efficiency at the pump-driver isn't always the ultimate consideration.  

...but I digress.

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

Sorry Big Inch, I didn't mean to offend you if I did. It was meant to be playful jab.

I would also consider the other benefits that a well programmed drive can offer.  There are some out there that offer dry run, dead head, min flow, run out and cavitation protection.

I run hundreds of pumps in my location and about 20 (which only one has caused me any issues) have drives on them.  So I fully understand and agree with you both on your valve position.  

It sounds like Sam has already purchased a pump since he says trimming the impeller is not an option.  Otherwise he would just have another pump selected.  My point is simply that in this situation, when you need to reduce pressure from 28 bar to 14 bar, if it looks good on the variable speed pump curve, you should seriously consider a drive.  

RE: VFD retofitting for pumps

The bigger pipeline pumps we normally look at have a maximum rpm of 4000.  With the VFD we can access the added capacity with off the shelf 3600 rpm motors by just uping the output frequency to 67 hz.

There are also automatic transmissions out there that can handle 1000 HP drives with 95% efficency.

RE: VFD retofitting for pumps

I thought so. I saw the smiley but the best defense is a complete explanation in any case.  

Ya, I agree its probably not going to be a waste of time to consider a VSD for this case.

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

(OP)
This is an old pump. It is around 20 years old and running Naphtha. We have done a change recently, so the discharge piping will go to a lower pressure vessel. So, we end up with a control valve closing 95% and producing very high noise. If we have to trim it then we have to do so much that may cause suction recirculation problmes. When need 78% of the maximum impeller diameter. So, I thought installing VFD is the easiest which will avoid any changing in the piping which is required with new pump installation. My main cerncern do I have to go for the motor supplier to confirm if his old motor can accept VFD. Or I have to buy new motor at is designed for VSD applications.

Regards for all contributions.

RE: VFD retofitting for pumps

We can't tell.  As a general guide, AC induction motors are designed for use at one speed.   Why not call an electrical engineer and let him have a look at it.  A standard induction motor when run at reduced speeds may have inadequate cooling by the motor fan. Over longer time periods, the motor may get overheated and ultimately fail. Also VFD output power to the motor inherently has power frequency harmonics.  These harmonics are likely to present higher voltages at the motor winding for which they are not designed to withstand with the result being failure of the motor winding.

Is your motor standard?

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

A VFD vendor can tell you.  The motor needs to have a certain insulation class to be VFD rated.  Also, it's a good idea to keep the drive as close to the motor as possible, less than 50 feet is suggested by GE, so the VFD enclosure is a consideration. Another consideration is your power quality.  Drives don't like "dirty" power.

RE: VFD retofitting for pumps

To run on a drive the insulation class of the motor must be increased.  Most newer motors already have this type insulation.  Motor manufacturers have gone to higher class insulation on all motors because they do not know if it will be used with a drive or not.  They know that running on a drive, a 480 volt motor will see spikes of more than 2,000 volts so they just plan on it.

If the motor is older and still has 600 volt insulation, it will fail very quickly on a drive.  If you have to replace the motor to use a drive.  It would be less expensive to just purchase a new pump and motor of the correct size.  Then you would not need the drive and you would not have to deal with all the negative side effects of VSD, which 2,000 volt spikes it just one of.

RE: VFD retofitting for pumps

New 75 HP drive $7500
New 75 HP motor $8000
Baseplate modifications $2000

OR

New pump assy about $15,000
Piping changes $2000

ValveCrazy makes a very good point.

RE: VFD retofitting for pumps

Thanks Clicketyclack.  But, I think the new pump he needs is about 75 HP.  To use the existing pump, he would need a 150 HP drive and new motor.  That would make the difference even more favor a new pump.

RE: VFD retofitting for pumps

If the new application only draws 75 HP max he could use a 75 HP drive on a larger rated motor. It just depends on the amp draw.

Of course this is an estimate anyways based on my Goulds example.  With an older pump he may need a 100 HP drive / motor combination, depending on the pump characteristics.

If the pump is very old and you have to buy a new motor and drive anyways, I would go for a NEW fixed speed pump that better fits the application, even if I had to make piping changes.  Why spend 75% of the cost of a new pump assy to bandaid an old pump? Sounds like something an accountant at corporate purchasing would do.

RE: VFD retofitting for pumps

Aw, but save 15% on the KwHr and that's $5000/year or NPV10 = $30,000.

RE: VFD retofitting for pumps

Not including power cost incrementation, but if I can get it up and running 1 month sooner without the VFD, I can make 175,000 clear profit during that time.  Great. At least now everybody's thinking about evaluation rather than personal preference.  OK, maybe we should add the VFD later when flowrates become variable or... drop.

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

dcasto - I agree, unfortunately the power calculations always seem to be forgotten and should form part of any good engineering study.
In many cases the savings in power costs can soon pay for new equipment - but then we have to fight with the "bean-counters" who consider this as capital expenditure for which we don't have the funds - but burn as much power as you like as this is operating expenditure and is your problem.

RE: VFD retofitting for pumps

And if this were a 7500 HP motor, the local power company may say no to starting it without a VFD. Or if it was the only motor and you had a demand meter that captured starts, you'd save too.

RE: VFD retofitting for pumps

Around this area N. CA motors over 50HP generally are required to have some form of reduced voltage starting. (Really it's reduced current.)

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

RE: VFD retofitting for pumps

300 HP is a large pump for me.  I don't know much about anything larger.  However, for the size of pumps and motors I deal with I like the old Auto Transformer soft starts instead of the electronic ones.  Sure it saves on the demand charge to use a soft start but, you don't have to use the same system to control the pump once it is up and running.  Dealing mostly with submersibles, the auto transformer style gets it up to the minimum speed quicker which is needed for the kingsbury type thrust bearings.

I have almost no problems with the auto transformer soft starters.  No electronics, just a couple relays, transformers, and a timer.  I can't say that about the VSD's and electronic soft starters I have used.

RE: VFD retofitting for pumps

  Why would you have a problem with an electronic Soft Starter?  

  Soft Starters cost less than Auto Transformers and provide a bunch of other features too!(like safety monitoring)

  Most are completely out of the circuit once the start is complete as they have bypass contactors in them.  They can also soft-stop the loads too, (just ask them).  They take up about 1/6 the space.

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

RE: VFD retofitting for pumps

Electronics do have their place.  Just with the size and kind of pumps I mostly deal with, electronics can be more trouble than they are worth.  Sure the electronic soft starters are smaller, lighter, cheaper, they don't have as much meat in them.  Heavy duty mechanical relays and transformers are going to be larger, cost more, and weigh more but, I believe they are more dependable.

I have probably sold or installed an average of 1,000 pumps a year for the last 30 years.  When you are responsible for this many pump systems, simplicity and dependability are the key.  Nobody cares if it is electronic or mechanical when water stops coming out of their faucets.  They just know who to call and scream at, their pump man.

If the bypass relay of the electronic soft starter is not large enough to handle and across the line start, when the electronics fail, somebody is out of water.  When somebody is out of water, they are chewing on my butt.

I studied electronics.  The guys I went to school with, think it is funny that I use what I learned about electronics to help build completely mechanical systems.  I can add an electronic monitor of several different kinds when needed.  I can also bypass these electronic devices and get the pump running and water pressure restored when needed.  Many times the electronics in a monitor, drive, or PLC cause me more out of water conditions than the failures they are designed to prevent.  

Because the pumps I work with are correctly sized for the application, very little reduction in speed is possible.  Most of these pumps will not lift water out of a well and build enough pressure to buck the static head unless they are spinning about 90% of full speed.  This means when a soft starter is programed to get the pump to full speed in 30 seconds, for the first 25 seconds the pump is not spinning fast enough to start moving any water.  Head falls off by the square of the speed, which is so quick that a soft stop doesn't do much good either.  

If the drive or soft starter can start or stop the pump at 90% of speed and gradually go up from there, it might help with the water hammer.  However, when a pump comes on it is usually because a demand already exist.  The longer it takes to get the pump to match the demand, the more amplified the transients are.

That is just my two cents but, I have a very narrow market.  ½ HP to 300 HP pumps that supply cool, clean water on demand.  I know there are lots of other applications where electronics are worth there weight in gold but, I believe most of my applications are better off without it.

RE: VFD retofitting for pumps

That's why I'm amazed to see a lot of VFDs on, or being sold for, well pumps.  If the well pump is sized correctly (as almost 100% static head) a VFD doesn't do anything good.

http://virtualpipeline.spaces.msn.com

RE: VFD retofitting for pumps

Hmmm,  I have used VFDs on several occasions just to get 3ph out of single ph so the down hole pump can be a less expensive more dependable 3ph instead of a capacitor dependant 'one lunger'.   Then you set them up to spool up in 2-3 seconds and you can avoid the twist and slap headaches.  You can also modulate the flow rate to match the demand and avoid the start stop cycling and pressure variations.  You can also prevent over draw on a problematic well.  If I ever do another well for myself I would definitely grab for these features.  Of course I'm an EE and love water works games so I'm probably a tad more interested in the (certainly) added complexity.

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

RE: VFD retofitting for pumps

You got me on the single phase to three phase conversion.  I have used drives for the same purpose myself but, only when I must.  However, I do not have any more problems with single phase motors than I do with three phase so, I still use 15 HP and smaller single phase submersibles and make my own 15 HP single phase centrifugal pumps.  When the demand is already present before the pump starts, 2-3 seconds can still give the twist and slap headaches.  

I use Cycle Stop Valves to "modulate the flow rate to match the demand and avoid the start stop cycling and pressure variations."   I can also prevent over draw on a problematic well with the same simple valve.  These valves react almost instantaneously which eliminates amplifying transients or "twist and slap"  When a demand is there, they meet it instantly.  When the demand changes, they change the flow rate instantly.  A slow response from a drive or speed controlled valve in these cases, is what amplifies the transients.

I have even used these valves to control the pump system when a drive is converting single phase to three phase and acting like a soft starter.

If you choose the right pump, throttling with a valve will give you the same reduction in power as a drive.  Simple control is what makes a dependable pump system.

I love electronics.  I have all the latest gadgets.  I would just never use one of them to control my pump system.  I am a little annoyed when my computer, cell phone, TV, or even my electronic thermostat go south on me.  However, when water is no longer coming out of the faucet, it is more than just an annoyance.  The simpler a pump control system, the more likely water will come out of the faucet when you open it.

RE: VFD retofitting for pumps

(OP)
I have got a chance to visit a motor manufacturer last week, and I met a design engineer who is above 60 years. His reply to this question: yes you can install VFD but you have to remember the following:
1-You will have a temperature increase about 10 deg C. So, if you are running away below from its rated capacity you should not have any problem. This is because the sine wave from VFD is not perfect sine wave.
2-Verify mechanical vibration, i.e. not running through a critical speed.
3-By installing VFD you should remember that that its efficiency is around 98%. If it is installed in a substation you need additional power to remove that heat, so you should calculate these losses and additional maintenance cost and compare these expenses Vs other alternatives.

Best Regards to all

RE: VFD retofitting for pumps

All true.

I would add that you need to be careful if the distance between the VFD and the motor gets beyond about 100 feet.

Keith Cress
Flamin Systems, Inc.- http://www.flaminsystems.com

RE: VFD retofitting for pumps

Just buy a new pump suited to the application and sleep easy at night.

RE: VFD retofitting for pumps

I looked and i looked - and finally had to use "seach" for the word "fan" before i found that Biginch allready had said what i think is the most important issue relating the the actual initial question: Cooling of the engine may be a problem if you add VFD. If the fan on the motor does not have its OWN motor then cooling may be inadequate. There may be some other requirements re. temperature monitoring for the motors that should be checked.

Best regards

Morten

RE: VFD retofitting for pumps

If you guys and gals come to Sunny 'ole Europe, you can use a satanrad IEC European frame motor with inverter - just make sure you have thermistors fittted as thermal protection.

Lots of interesting discussion though, learnt alot - thanks !!

Ash Fenn

www.cdrpumps.co.uk

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources