6 replies to this topic

[marke]

It is very common practice to operate VSDs on pumps with a pressure transducer and PID loop such that the speed of the pump is controlled in a manner to keep the pressure constant.
This relies on the fact that at a constant speed, the pressure increases for reduced flow.

There are a number of pumps being used that are "Flat Curve" Pumps where the pressure does not increase much for reduced flow at a constant speed. In this case, the presure transducer approach does not reliably reduce the pump speed when the flow is reduced. Where sleep mode is employed, it is not uncommon for the pump to never go to sleep because the speed deos not reduce to minimum speed.

There are a number of possible approaches to this situation, but I would like to know if anyone has done a good analysis of these approaches and determined which will work the best?

Best regards,

Attached Files

•  flat_curve_pump.pdf   42.28KB   22 downloads

[joze]

Hello marke!

I'm sorry not to offer any analysis. Let me just point out an advanced function available in one of the latest arrivals on the VSD market which is called "advanced sleep/wake up".

This function is basicaly a booster (+ or - speed) after a certain period of time when the pump is below a certain flow limit - if the feedback is right, drive ramps down. So, one step further than sleep function which would just measure time interval and turn off if it's passed.

This of course does not yet constitute an analysis so I will be very interested in anything that might come up during this discussion. Please let me know also of other possibilities.

Best regards,

Joze

[marke]

Hello joze

There are two issues with the flat curve pump.
When the flow is reduced, the pressure does not rise and so a constant pressure system does not naturally reduce the speed bakc. This has the impact of increased cavitation in the pump with associated pump wear and energy losses, and of course, if the pump does not drop right back in frequency, the traditional sleep functions do not perform.

My major concern, is optimising the run eficiency at low flow rates. In any of the cases that I have seen, the client would be better off using fixed speed ON/OFF control as the energy savings are minimal without a good fold back in speed.

Thank you for your thoughts.
Best regards,

[joze]

Hello marke!

I still think we are speaking of the same thing:
1. for certain pumps around zero flow pressure is not so dependant on speed
2. during a progressive transition from low flow to zero flow the regulator can freeze at a speed higher than the sleep speed and the pump will never stop

So, advanced sleep function kicks in:
3. parameters allow, in the low flow zone, to query the PID regulator to verify if the regulation point is correct

I don't know if I can add an attachment on this page or if it is correct for me to post some vendor specific documentation (I work for industrial automation company) so I didn't but I think this may be a solution for you.

On the other hand - why don't you use flow rate sensors? Mechanical or cost hindrance or both?

Looking forward to some interesting answers,

Joze

[marke]

Hello jose

yes we are talking about the same thing but I want to be able to optimise the eficiency of the pump below 50% flow. This will not appen on a constant pressure system because the drive will not be controlled correctly below 50% flow. This has an impact on both the sleep function and on the operating speed and pump efficiency.
There are a number of schemes to enable the sleep function, but the major reason for using a drive, is to optimise the operating efficiency. If your flow variation is between say 100% and 75%, then the system will modulate the pump speed for an efficiency improvement, but not below 50% flow.

Best regards,

[jOmega]

Hello jose

yes we are talking about the same thing but I want to be able to optimise the eficiency of the pump below 50% flow. This will not appen on a constant pressure system because the drive will not be controlled correctly below 50% flow. This has an impact on both the sleep function and on the operating speed and pump efficiency.
There are a number of schemes to enable the sleep function, but the major reason for using a drive, is to optimise the operating efficiency. If your flow variation is between say 100% and 75%, then the system will modulate the pump speed for an efficiency improvement, but not below 50% flow.

Best regards,

Hi Mark,

I was browsing thru tonite and came upon this old chestnut from almost 5 years ago ...

The first thing that came to mind is: Where's the System Curve ?

The second thing that came to mind is: Is this an open or closed loop system?

The third thing that came to mind is: Where is the pressure transducer located; at the discharge of the pump? or at some place further downstream where a controlled pressure is required .... and if so ...how far from pump discharge is that location ?

There are other items to consider, but these will do for now to see if there's any interest in revisiting this bone from the past.

Kind regards,

[marke]

Hello JO

The systems are typically closed loop with a pressure transducer close to the discharge of the pump.
The system curve varies depending on the required flow/pressure. Commonly, this sort of pump is used to provide water to two or more pivot irrigators and needs to optimise the pump speed depending on the flow required.

Partially closing a valve reduces the flow, but does not significantly alter the pressure, so the classic constant pressure system continues to run at the same speed dissipating cinreased power in the pump.

Best regards,
Mark.