×
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

Pumps in Parallel Project
4

Pumps in Parallel Project

Pumps in Parallel Project

(OP)
I have a parallel pump projet where the application is to supply a packaging plant with general water. The system in place is a parallel systme with one 15kw pump in parallel with a 22kw pump without any monitoring. These pumps are started manually. I have gone though a couple of threads like thread407-191266 and have discoved that pumps in a parallel system is not as straight forward as put out to be. I had the idea of using identical pumps and swiching between them using a signal VSD when ever one fails the other would come online. But with the fluctuating demand, depending on how many machines are running, and only having the option of monitoring the pressure on the discharge side of the pumps to regulate the VSD the pressure transmitter would react to slowly to regulate the VSD - this would cause hammer? How would i go about selecting the best option for this application?

RE: Pumps in Parallel Project

2
It sounds like it will work.  If the pumps are identical (and stay that way), it is as simple as the textbooks say.  If you do not use two different pumps at the same time, it is as simple as the textbooks say.  You move to the complex mode if the pumps are not identical and operate at the same time.  Different pumps running at the same time, must give the same pressures at the points where their piping is common.  Two different pumps operating at the same time and attached to a common suction header and a common discharge header have to produce the same differential heads to get the same header pressures, otherwise an ambiguous situation is created in the headers; where you can get two different pressures at the same point.  Hydraulic mathematics doesn't like that discontinuity and pumps tend to continuously adjust their heads as flowrates drift back and forth with the pressure each sees in the headers at any given time as the two different pressures try to balance themselves by adjusting local flowrates. If they do that well, the system is stable.  If they don't do that well, pumps tend to oscillate about a central flowrate, or drift into unstable modes with one pump running at higher load than the other.

Two identical pumps generally work well operating together, if the headers are also hydraulically balanced.  Don't attach a small diameter header and place one pump on the end and the other in the middle of the headers and things should work out just fine.

You may not need a VSD, if your flowrates do not vary much.  If the process flowrate varies 50% or more, a VSD may be the right solution, otherwise a pressure control valves may offer more benefits.   

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

I run different size pumps in parallel all the time.  You can use a pump control valve or a "constant pressure valve" on each pump.  As long as the set pressure of the valves, which is also the header pressure, is at least 10 PSI lower than the shut off head of any pump, there will be no problems.  In this way each pump is only bucking it's own back pressure, not the pressure of any of the other pumps.

You can still alternate if you want but, staggering the pressures to set up a lead and a lag pump is a much simpler set up.  Alternating usually wears out both pumps at the same time.  Staggering will leave less hours on the secondary pump, which makes it more dependable as a spare.

RE: Pumps in Parallel Project

valvecrazy, does this pump have control valves?  Sounds like its got nothing right now.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

Doesn't sound like these pumps have any controls other than a mag starter.  If these pumps are never running at the same time, there should not be a problem.  If they both run at the same time, then their can be problems without the right controls.

RE: Pumps in Parallel Project

(OP)
Thanx for the feedback, i was hoping for a comment from you two guys.  You have posted some interesting stuff.  I have come to a decision to buy identical pumps and run them one at a time over a single VSD (PID controlled) the pressure at the discharge will be monitored by a 4-20mA pressure transmitter to control the speed of the pump. If one trips the other comes online.  The pumps will be hardwire interlock and software interlocked to prevent simultaneous running of both pumps.  Non return valves will be installed at the discharge of each pump to prevent the impeller of the standby pump from rotating due to pressure differences. Thanks again.

RE: Pumps in Parallel Project

The reason you need non-return valves on the discharge is firstly - so that flow does not return to the source via the second pump and secondly - so that it is not running in reverse if called to come online.

RE: Pumps in Parallel Project

"If one trips the other comes online."

With two pumps and only one Drive, you still have all your eggs in one basket.  It is more likely the Drive will be what trips out instead of a motor.  You can use an Across The Line bypass for the Drive.  However, when the Drive malfunctions and you are left running on bypass, you will have no control over the discharge pressure.

With an Across The Line starter and a control valve for each pump, you would have a dependable primary system and a dependable backup system.  Ether valve would be capable of controlling the downstream pressure while any or every part of the other pump system is removed for repair.  Centrifugal pumps can be so dependable when running ATL with Valves as controls, that the entire system can last 20 to 30 years with no problems.  

"From past experience, small drives typically fail in eight years and are
replaced rather than repaired."

http://www.powerqualityanddrives.com/payback_analysis_vfd/

I personally find smaller drives to have more like a five year average life.  Therefore, you will have replaced the Drive from 3 to 6 times instead of the system having lasted 20 to 30 years without any problems.  I don't think a Drive could ever save enough energy to pay out in a case like this, not to mention cost and headaches of the down time, and all the other associated problems that go along with Drive control.

There are those that would have you believe that a Drive is a "perpetual motion machine".  I believe a Drive is a "perpetual maintenance machine".

RE: Pumps in Parallel Project

Another advantage of using a drive on this project is that there are special pump protection VFDs available that can shut down the pump or trigger an alarm if it detects a process upset.  They can detect no-flow or dry run conditions that can damage the pump.  These pump protection VFDs can save the user money by helping to avoid pump repairs and downtime.  It can pay for the VFD on the first repair that is avoided.  It is just another way that VFDs can be a good thing in the right application.  This could be one of them.

RE: Pumps in Parallel Project

There are lots of pump protection relays out there that do the same thing, without the added complications and expense that go with varying the speed.  Personally, I like to use a simple low pressure cut-off switch.  This will protect you if you have a break in the discharge line, or the pump is running dry.  In this way there are NO electronics in the entire system, which is what I believe makes a pump system more reliable and longer lasting.

RE: Pumps in Parallel Project

Bravo!  A VFD for pump protection is like asking your neighborhood child molester to babysit.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

(OP)
The water is pumped from a buffer tank and the pumps are interlocked with the low level.  The discharge line in fitted with a flow sensor.

The drive has its own main contactor and overload.  After the dirve each pump has its own contactor and overload.  Under the condition that the overload of one pump trips and not the drive then the standby pump will start.  

Artisi
Why does the flow return to the source and the impeller turn the wrong direction... because there is a pressure difference but i'm only electrical guy.

RE: Pumps in Parallel Project

Artisi, if I may,

High pressure in the discharge header will tend to reverse spin a stopped pump.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

You already have a level switch and a flow sensor, centrifugal pumps are very dependable, the Drive is your weak link in this chain.  I predict the Drive will fail or at least hiccup in short order.  The pumps will still work but, you will have no control over pressure.  I have lots of systems out there like this that have a pressure reducing valve for the backup system.  Usually the system ends up running on the bypass contacts and the bypass control valve most of the time.  Then when they get tired of replacing or working on the Drive, they will finally decide to just leave it running Across The Line with the Valve as the main control.  Then everything becomes dependable.

RE: Pumps in Parallel Project

packlbn, Valvecrazy is givin' you some good advice here.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

packlbn,

You might want to check the valve / VFD argument in thread237-205555: Harmonics and VFDs regarding the pro's and con's of each. There are some sound arguments from both sides, plus some quasi-science from those with a vested interest in selling their products. Draw your own conclusions.
 

----------------------------------
  
If we learn from our mistakes I'm getting a great education!

RE: Pumps in Parallel Project

Well .. I'm not selling valves or VFDs.  

Very simple rule for VFDs VSDs..

IF FLOW VARIES BY 50% OR MORE

     A VFD might be more efficient, but ONLY IF
     required head varies appx with flow squared.  

IF NOT

     You don't need to be thinking about VFD.
     You will have a more efficient system using
     a properly selected pump & control valve.

Be very suspicious of anybody who won't agree with the above.  You may find an occasional example that breaks this rule, but you could probably count them all on one hand after you retire.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

BigInch-
"BigInch (Petroleum) 6 Mar 08 3:34  
Artisi, if I may,

High pressure in the discharge header will tend to reverse spin a stopped pump.
http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain"
 
No argument from me on this one- actually I missread / interperated the posting from  - packlbn - I was just pointing out that a NRV has two functions in this particular installation, the main one being to stop flow back to the source and to prevent reverse rotation of the impeller which is realy only a problem if the unit is called to start while running in reverse (as in this installation)  
This is what was said by packlbn anyway just my poor reading  - bad hair day.  

RE: Pumps in Parallel Project

Not to hijack this thread, but some VFDs offer more than the efficiency advantages they can bring to some systems.  They offer the pump protection I already mentioned, and can also detect closed discharge valves or other system problems that can quickly burn up mechanical seals.  They can also offer features like precharge modes where they operate the pump at a slower speed for a set time in order to fill the pipes in the system in a more controlled manner.  Some even allow you to get rid of troubleprone flowmeters by calculating flow internally, within 5%.  Modern pump specific VFDs are much, much more than the VFDs of even 5 years ago.

Pump Protection VFDs can add add a whole new level of control, and can provide opportunities to catch small system upsets before they become big and expensive pump problems.  If you haven't looked at the latest pump specific innovations several manufacturers have added to their VFDs, it is worth your time to look at them.  

And yes, VFDs are not appropriate for all applications.  

RE: Pumps in Parallel Project

Thanks BigInch!  I can't say that myself because, having the word valve in my name, I am automatically given a 16' of head disadvantage, and am accused of "burning energy" and melting the polar ice caps.

Packlbn....If the "buffer" tank you are drawing water from is below ground causing a suction lift, and you use the same suction line for both pumps, you should put the NRV or check valves on the suction side of the pumps.  Otherwise the pump that is running could draw air from the mechanical seal of the pump that is not running.

RE: Pumps in Parallel Project

So the VFD’s of just 5 years ago are now completely obsolete, just like the ones produced now will also be obsolete very soon.  This same cycle has been happening with Drives for decades, and will continue to happen for decades, because there are things mother nature just won’t let you fix.  You are right that the pump specific Drives have more features and are easier to set up than Drives of a year or so ago.  However, they still don’t save enough energy or last long enough to justify their use on most pump applications.  Drives do allow people who do not understand pumps, to band aid a badly designed system.  Fresh water pump systems have been known to last 20, 30, even 50 years without problems.  This will never happen with a Drive controlled system.  I can’t help but think, that is exactly why pump companies and many other people are pushing Drives like they do.  Job Security at the end users expense.

All those protection features can also be had with many different brands of pump protectors.  Even with Across The Line controls, flow rate can be calculated by power consumption.

I think inviting a child molester to baby sit is a very good analogy for a Drive on a fresh water pump.

RE: Pumps in Parallel Project

Ya. I don't know where that came from, but it'll probably  wind up being the very best analogy I'll ever think of.  Well, it'll make the top ten for sure.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

I've said it before and I'll say it again- it depends on what your pumping. Discharge control valves may be good for clean fluids- particularly with high static/low frictional heads- but they don't last very long on abrasive slurries and aren't the best fit on systems with high flow variation (with associated large changes in frictional pressure loss).

Just because a system has a reasonable range of operating conditions- doesn't necessarily mean it's badly designed.

RE: Pumps in Parallel Project

Can't argue with the specific cases, only in the general and trivial manner that VSD seem to be sold these days as the Snake-oiled Holy Grail for all pump problems.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

RE: Pumps in Parallel Project

ROFL, that sounds like the way valves are being presented by certain people in another thread...

VFDs are mis-sold into unsuitable applications, valves are mis-sold into unsuitable applications. Selective truths are presented by VFD people and by valve people as 'evidence' that their solution is 'best'. As long as salesmen make their living from commission based on sales this will continue. Both technologies have their place, depending on the application.

Caveat emptor!
 

----------------------------------
  
If we learn from our mistakes I'm getting a great education!

RE: Pumps in Parallel Project

Valvecrazy said: "You are right that the pump specific Drives have more features and are easier to set up than Drives of a year or so ago."

Kudos to Valvecrazy for showing he is keeping up with some of the latest progress in drive design.  That is step in the right direction.  Or perhaps towards the dark side, depending on your perspective.    

RE: Pumps in Parallel Project

Biginch

I think your point is interesting although i accept your reservations regarding certain applications - however i would like to know if your opinion towards use VFD is stricly economical, control optimization or a combination?

Best regards

Morten

RE: Pumps in Parallel Project

Somebody got close to saying it but....

Drives need to match system requirements; and controls need parallel system responses.  

A high differential pump required because the system has a high static head or contains spring loaded in line components or even thermal control type devices do not make good candidates for VSD's.  And a turbo machine that has a "flat" head to flow curve is not going to fit a circuit with a high dynamic component operating over a broad speed range.

A high flow circuit where most of the required pressure difference is dynamic or frictional; along with a pump having considerable slope to the V to h curve, can do well on a VSD providing slowing it pushes it toward the sweet spot (BEP).  And the pump is full of impellor to begin with.

Fans often perform a lot like the latter. The characteristics of pump circuits vary a whole lot more.  

$0.02 & done.

RE: Pumps in Parallel Project

Lotta good stuff on pump-zone.  Be sure you scan it clean.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

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