17 replies to this topic

[Ra v]

Hello everyone,

 

We are facing problem in our cooling water centrifugal pump discharge pressure. The pump design details are below  

 

Motor Description Pump Description         Frame No ND 160 M     Ref HCAM 0966 HYD Pr 24 Bar Efficiency 86 Machine No 13090675 KW (HP) 11  (15) Capacity 600 LPM RPM 2920 Pressure 3.5 bar Voltage 415  ± 10% RPM 2920 Bearing DE-6309; NDE-6209,2RS Year 2009 Amp 20 Material CS     Head  35 Meter

 

Present situation:  Discharge pressure is  1.7 bar

we want to increase the pressure upto 3.0 bar, so what changes we need to do ??

 

Is it recommended to increase the flow in order to increase the pressure ??

What happens if we decrease the head ? Can it effects the pressure and flow ??

 

 

Regards

A Ravi

[ColinR33]

Ravi,

 

The first thing you need to do is brush up on the basics of centrifugal pump characteristics/design/operation so you understand what is happening in the pump and how it works, then take a look at the pump curve, this will tell you a lot about where your pump is operating and what will happen when you start to change the flow.  Increasing the flow to increase discharge presssure will not work.  Differential head on a centrifugal decreases with increasing flow, how much depends on the pump, some have fairly flat curves (gradual decrease in diff head with increasing flow), while some have very steep curves (diff head decreases rapidly with increasing pressure).  If your discharge pressure is much lower than design, it could mean a damaged impeller, or your pump is operating close to the end of its curve, pushing out more flow but less head, and this can be due to a lack of backpressure in the system.  Have you checked the actual flow?  If the pump was designed to operate at a certain discharge pressure and the actual pressure is much lower due to lower than expected system losses, then the pump will "run out".  Try partially closing off a valve (or valves) in the system to add resistance and the pump should move back up its curve to the proper flow and discharge pressure.  If you are intentionally running at a higher flow than the pump operating point, then your discharge pressure will be lower and you will need a new impeller and possibly motor to get the additional head (assuming the existing casing has room for the larger impeller, if it does not, you will need a new pump).  If the pump is operating at its correct flow and adjusting system resistance does not improve operation it is likely a damaged or undersized impeller.

 

Hopefully my ramblings make a modicum of sense and others can chime in,

 

Cheers,

Edited by ColinR33, 04 March 2014 - 10:16 AM.

[gegio1960]

Hi Alamuri.

Colin answer is almost covering all the possible aspects relevant to your questions... anyway, you could obtain a more specific answer if you add some details....

Scope: why do you need to increase the discharge pressure?

What are the currently discharged flow and the one you want to obtain?

How much is the currently absorbed electricity?

Have you available (could you attach) the pump curves?

good luck!

[Ajay S. Satpute]

Hi,

 

Can you share the pump curves including P.O.R. values by vendor?

 

Regards.

 

Ajay S. Satpute

[Ra v]

Thanks ColinR33 for the reply.

 

And gegio1960

 

Unfortunately i do not have the pump curves with me for the pump and the name plate on the pump was also little bit confusing. it was damaged , so i was unable to get the information.(such as vendors name).

 

Discharge flow as per pump design is  600 LPM, but we don't have any rota meter to tell exactly.

 

Absorbed power : 600 LPM = 10.02 LPS*35(head)* 1(SG)/102 = 3.43/0.80(eff) = 4.56 KW

 

Actually the process demands for the cooling water flow in circuit is around 2.8 bar, and now the pressure is 1.7 bar.So we are trying to install a new pump with increased capacity in parallel. But if we put pumps in parallel then its nothing but flow rate added to the system ,but not the pressure. and if we go for series , it is the head added which means decrease in pressure.

 

Increase in RPM is the case to develop the pressure so we have to buy a new motor with more power.So we want to increase the discharge pressure of the pump. But the question is how??

 

Rgs

A Rav

[ColinR33]

Ravi, I do not understand your statement "and if we go for series, it is the head added which means decrease in pressure" 

 

Head = Pressure!

 

Adding a booster pump is a common way of increasing differential pressure (which means higher discharge pressure). 

 

What concerns me is you have no pump curve for the existing pump and no flow meter on the system so you don't know what you are actually flowing, therefore how can you characterize the system in order to add a new pump or even properly control the one you have?  You really need to understand what you have and how it is operating.  If the existing system demand is such that the pump is running down on its curve (ie. higher demand than the pump was designed for) then the pump will actually be operating at a higher capacity than the nameplate, with a correspondingly lower differential pressure = lower discharge pressure.  If this is the case, adding a booster pump with the same nameplate capacity as the first may not solve your problem because it would end up doing the same running down on its curve as well with a lower than expected differential pressure and you still don't get the pressure you want!  You'll end up buying and modifying pumps on a hit and miss basis which can be a very expensive proposition.  The other possibilities are as previously stated - damaged impeller or insufficient control which is allowing the pump to run out on its curve.  Understand what you have and what the system needs and then you can properly specify new pumps.

[aroon]

Hi Ravi,

 

As mentioned by Collin .... there are several possibilities and probable solutions, for which you have to study whole system thoroughly. Here I am assuming that you have cooling water network for your facility. The following things you will have to do as primary analysis:

1. Calculate the flow required for each and every user (all the equipment where you are using cooling water).

2. Sum those requirement and compare with the present capacity.

3. If sum is high then requirement of additional flow can be met by following two options

 

Option-1: increase impeller diameter to its maximum possible provided your motor should meet the power requirement.

Option-2: If option -1 is not meeting your requirement then additional pump in parallel is mandatory.

 

If sum is less than present capacity then below noted point should be taken care, which may help to meet the capacity with present pump itself.

 

Note the following:

- Cooling water header hydraulic is very much important to find out the reasons for low discharge pressure of the pump. Carry hydraulic study based on piping.

- There should not be any sort of channeling of the flow such as minimum flow circulation line or utilization of cooling water at low elevation will cause low pressure drop route and maximize (channel) the flow through them, which subsequently result again in low discharge pressure of the pump.

- Try to optimize the requirement at every user. Many times cooling water utilization is in manual mode and manual valves are at their full open conditions, which is some time not required. In this case you are not using your utility effectively. This you may cross check by measuring cooling water outlet temperature at every user (equipment). If it is less than maximum allowable cooling water return temperature then you are not using your utility effectively provided your equipment (mainly heat exchangers) should be design to take that LMTD.

- All the user should be at their optimum capacity.

 

Conclusion of all the above input is that you should study your system in depth before taking any final decision. Better, you should contact pump manufacturer to make the curve available for that model if those are not available in any of the your department because performance curve is very much important without that you can not check the capability of the present pump.

 

Hope this may be useful for your study. All the best.

 

Regards,

Aroon 

[gegio1960]

Hi Colin. Thanks for your post #6 that has helped me to maintain the calm...

Dear Ravi, I don't know where is your plant and what is made for but, for sure, here in Europe, you couldn't have any responsibility on it.

I don't intend to blame you but to highlight the risks that your way of going-ahead can create.

.

[fallah]

 If the existing system demand is such that the pump is running down on its curve (ie. higher demand than the pump was designed for) then the pump will actually be operating at a higher capacity than the nameplate, with a correspondingly lower differential pressure = lower discharge pressure.  If this is the case, adding a booster pump with the same nameplate capacity as the first may not solve your problem because it would end up doing the same running down on its curve as well with a lower than expected differential pressure and you still don't get the pressure you want! 

 

Colin,

 

Yes, it may not solve the problem, but by selecting a pump with proper performance curve matching with that of the existing pump, it may solve...!

[gegio1960]

Fallah, how Ravi could make that? He didn't find the curve ;-)

And when Ravi will have the curve, he shall be made other steps of system analysis before adopting that solution.(booster or replacement "in kind").

[fallah]

gegio,

 

Yes, for a classical study to approch the required working point by the pumps in series we should have performance curve of the existing pump, but i juat wanted to say it might be possible by a little bit operational adjustment.

[Ra v]

Hi everyone , Thanks for the responses. 

 

Collins , my statement " if we go for series, it is the head added which means decrease in pressure" Decrease in pressure means "discharge pressure " of the pump would might change ( low )

 

I have a doubt on discharge pressure and Head?? What happens if Head is decreased or increased?? how will discharge pressure of the pump varies??

[ColinR33]

@ Gegio, thanks for your comments

 

@ Fallah, yes, if the pump curve was available this would be a fairly simple fix, but as Gegio stated, without the pump curve we can't match it! ;-)

 

Ravi, your statement "if we go for series, it is the head added which means decrease in pressure" does not make sense.  How can adding a pump in series lower the pressure?? (unless you install it to run in reverse sort of like an energy recovery turbine....).  I am worried you are trying to design this without a proper understanding of how a pump works.  As I stated earlier, Head = pressure!  A centrifugal pump converts rotational kinetic energy (supplied by the motor) into hydrodynamic energy (pressure).  Pumps are characterized by the Total Differential Head (TDH) at a given operating condition.  The differential head is the PRESSURE INCREASE across the pump, if the TDH decreases (ie, increased flow) the discharge pressure will also decrease (assuming a constant suction pressure).  If the TDH increases (reduced flow), the discharge pressure increases, again assuming a constant suction pressure.  Again, you need to look at the pump curve to see how the differential head varies as the pump flow varies - I understand you don't have one for this pump but Google is your friend - there is a lot of information out there and a lot of typical curves for you to look at. 

 

Aroon has made some very good suggestions for what you need to do to characterise the system and determine the pump capacity.  In addition to what he has suggested you would also have to generate a system curve, this involves hydraulic calclations to determine the pressure drop through the system (piping, elevation changes, control valves, users, etc).  Combined with the capacity calculations outlined by Aroon, this will help you select the proper pump as capacity is only one part of the equation, the pump TDH is also very important.

 

I think you really need to get in touch with someone who understands pumps and pump systems and get their assistance.  Proceeding as you are without a proper understanding of how pumps work is a recipe for a lot of wasted money and effort.

Edited by ColinR33, 06 March 2014 - 10:42 PM.

[Ra v]

Collins Thanks, now i understood about the TDH and Discharge Pressure. Thanks for your suggestions and i googled and i will try to understand the curves and pumps terminology.

 

Thanks to everyone.

 

Regards

A Ravi

[kalpana]

thanks colin sir for you valuable information sharing with us.. i got lot of knowledge about pumps from your discussion 

[Padmakar Katre]

Thanks ColinR33 for the reply.

 

And gegio1960

 

Unfortunately i do not have the pump curves with me for the pump and the name plate on the pump was also little bit confusing. it was damaged , so i was unable to get the information.(such as vendors name).

 

Discharge flow as per pump design is  600 LPM, but we don't have any rota meter to tell exactly.

 

Absorbed power : 600 LPM = 10.02 LPS*35(head)* 1(SG)/102 = 3.43/0.80(eff) = 4.56 KW

 

Actually the process demands for the cooling water flow in circuit is around 2.8 bar, and now the pressure is 1.7 bar.So we are trying to install a new pump with increased capacity in parallel. But if we put pumps in parallel then its nothing but flow rate added to the system ,but not the pressure. and if we go for series , it is the head added which means decrease in pressure.

 

Increase in RPM is the case to develop the pressure so we have to buy a new motor with more power.So we want to increase the discharge pressure of the pump. But the question is how??

 

Rgs

A Rav

 

Hi,

Once you add a parallel pump to existing and I presume the total flow requirement is same then your header pressure should go up as individual pumps flow would be reduced.

[Bhaulman]

Greetings! Good software will allow you to analyze your situation fully. Changes to your pump operation totally depend on the system that it exists in and therefore an analysis of the pressures and flow needs must be considered. Your pump may be able to run at different speeds or have another impeller size used - it all depends on your pump's type, current configuration and, again, your system's layout. So analysis is essential. Try finding HYDROFLO software (free Academic version) and see if your system can be modeled with it. Then you can play with the system's design, change the pump's operation and so forth. Good luck!

[Alfred210]

The pump type needs to be confirmed first, per the flow rate, could be a smaller centrifuge pump. Different strategy can be used after that.