Calculated flow based on pump curve and motor data

[somlioy]

Hi

The pump smart PS220 VFD has the ability to estimate pump flow based of pump characteristics and measured motor data.

The calculations requires BEP flow, BEP Power, pump rated speed and power at zero flow.

Does anyone know how this is done such that I can implement it on the PLC in those applications which doesnt use the pump smart VFDs?

 

[atc222]

Just finished an aeration control system using centrifugal blowers. Typically I use a loop with dissolved oxygen as input, and flow in scfm as output. This is an upgrade to an existing system. The existing system had surge, overload, design point, and a couple other points superimposed on an analog current meter. Plotted those points, and for the region of concern, it was fairly linear. So a simple y=mx+b works. Convert loop generated scfm target to current, and control inlet valve to blower to hit target.

This is crude, but effective.

So short answer, you need the pump curve. Current and speed feedback from drive, and probably discharge pressure to come up with a good approximation.

Of course you could put a flowmeter on the discharge side of the pump, and this all becomes academic.

Hope that helps!

 

[Tom Jenkins]

Is head measurement part of the data?

 

[Tom Jenkins]

This is crude, but effective.

This is actually a fairly accurate method if you have an inlet throttled centrifugal blower - it's been used for decades.

The method falls apart if the blower uses VFDs or guide vanes for control for a variety of reasons. Here is an alternate method that can be used for those cases.

https://www.blowervacuumbestpractic...blowers/measuring-blower-airflow-in-the-field

 

[somlioy]

Thanks. No other measurements are available other than those mentioned, and obviously data from the drive.

Please see the following video on the PumpSmart system:
https://www.youtube.com/watch?v=NFMf4kI9H3Y

 

[atc222]

Great article Tom.

 

[Tom Jenkins]

It looks to me as if the fundamental calculation is simply the linear approximation that atc222 used for his blower. At the base speed they calculate using y = mx + b, with y flow and x power. The get the slope, m, from the two points of flow and power, and then calculate the intercept, b, from the power at zero flow. They then apply the affinity laws to correct for different speeds.

Some pump curves don't go back to zero flow. In that case you can take the power at the minimum flow given and calculate the intercept.

I'm a bit skeptical about the ±5% accuracy, especially at alternate speeds, since the basic pump curve is typically ±3% by itself and for some pumps the hp to flow curve isn't really a straight line.

 

[drbitboy]

There is a fifth parameter: the pump type.

Cf. https://www.ittproservices.com/ittg...roducts/Control Solutions/PS220/PS220_IOM.pdf page 19, item C.

Other items of note:

• "complex algorithm" (pp. 18, 140)
• "SmartFlow can have an accuracy of ±5% of rated flow at rated speed. This is made possible by conducting
  a SmartFlow Tune. The SmartFlow Tune is conducted against a closed discharge valve and will take no more
  than 35 seconds." (p. 31)

So a complex algorithm, plus an automated calibration, and they can get 5% accuracy.

It might be possible to calibrate your own pump and make a model from that plus some basic affinity laws.

I am a little confused by the "Efficiency" ordinate on the right hand side of the plot in the YouTube video.

 

[Tom Jenkins]

There is a fifth parameter: the pump type.

Good catch - thanks. Based on the video I assumed a centrifugal pump. If it is a PD pump flow is simply linear with speed and the affinity laws don't apply.

 

[atc222]

Tom,

Could you elaborate on affinity laws?

 

[atc222]

Tom,

Could you elaborate on affinity laws?

Actually google to the rescue. Good Wikipedia article.

Learned something new today!

 

[atc222]

Tom,

Any chance you could post links to the articles you’ve written? I’m currently living in the water/wastewater world, and I find on occasion I’m reinventing the wheel. Successfully, but I’d benefit from your knowledge.

Regards,

Andy

ps: Folks like you and Peter N. are great resources, and make this site worthwhile! Not leaving anybody out, just citing a couple examples.

 

[Tom Jenkins]

Tom,

Any chance you could post links to the articles you’ve written? I’m currently living in the water/wastewater world, and I find on occasion I’m reinventing the wheel. Successfully, but I’d benefit from your knowledge.

Sure thing! You can access books, articles, and white papers from my website:

https://www.jentechinc.com/publications.php

(And thanks for the kind words.)