Retrofit high service pumps with VFDs 1

[djcbgn]

Hello Everyone,

We are installing VFDs on our High Service Pump Station pumps at the WTP. There are currently 3 large (duty pt.:6,000 gpm @ 360' TDH)and 2 small(3,000 gpm @ 360' TDH) constant speed split case centrifugal pumps. The pumps have been running out on the curve causing excessive maintenance and repairs due to cavitation. Therefore VFDs are being installed to run the pumps closer to the BEP, reducing costly repairs and energy costs. VFD will be initially installed on 2 large pumps and 1 small pump. I need to develop a functional description for operating these pumps. This project was developed by another engineer at our water system who has since left and the VFD installation is almost complete so there are no other options to consider for the original problem(like impeller trimming or control valves).


WTP staff prefers to operate the pumps based on elevated storage tank levels versus pressure or flow. I am going along with this for now. With varying demands, typical scenarios will be one small pump running, one large pump running, one large and one small pump running, and possibly two large pumps running. All these scenarios will have pumps running at less than full speed (still need to work this out, though). The tricky part is running a small and large pump together in parallel(could have the large pump on VFD and one small constant speed pump combination). I see almost endless combinations. Any suggestions on speed control using tank levels and how to approach the varying pump combinations.

Thanks in advance.

 

[77JQX]

You need to develop an operating plan for inappropriate technology using uncooperative operators? I'm glad I don't have your job. I don't see how this project will reduce repair or energy costs.

If you have any hope of success, you need to understand how and why they currently operate the five installed pumps. Attempt to mimic this in your solution, or there will be resistance to it.

In theory (I suggest you not do this) you could build a table of tank level (or static lift) on one axis, with pump combination on another axis (to estimate friction), and lookup values of TDH, and use the THD to arrive at the required VFD speed. That's a klunky solution, and requires you to think of every possibility and come up with (model) the correct solution before hand.

If it was me, I'd insist the pumps be controlled using pressure or flow.

 

[bimr]

There is an expression that you "run with what you brung".

What you are now thinking about should have been done prior to ordering the pumps. Now, your job is to see if what was purchased will work.

With a static dominated system, a VFD will not allow you to operate close to the BEP to save operating cost. It may help to stop the pump from running out on the curve.

"For static-dominated systems (injection against constant pressure, lifting against constant head), the energy savings are substantially less, and the pump surprisingly operates substantially off-BEP position, not significantly different from a valved flow control case"

http://www.pump-magazine.com/pump_magazine/pump_articles/article_44/article_44.htm

 

[1gibson]

So they are running fewer pumps out on the curve, instead of running additional pumps at the correct flowrates?

As soon as you install the VFD's, the operators will just overspeed the same fewer pumps instead of bringing more on. Cavitation will be worse because instead of running out on the curve at rated speed, they will run out on the curve at max speed the VFD can command, which will increase NPSHR, and the extent/frequency of damage to the pumps due to cavitation.

If someone doesn't know how to safely operate equipment, you don't add features that make it possible to operate even further beyond the design conditions.

 

[BigInch]

You need motors rated for VFD operations. A VFD will not keep operation at BEP. You must control pump head and system curve. A VFD will only control pump speed and vary the pump's head vs flow curve. It will not control your system curve. You must still adjust flow to match the BEP of the pump at any given speed. Otherwise you will just continue to operate somewhere at the intersection of the pump head curve and system curve, except then the pump head curve can vary with pump speed too.

All it appears that you need is to install a discharge pressure control valve. That will in effect change your system curve and be able to keep flow regulated so that it matches BEP at full speed.

I hate Windowz 8!!!!

 

[Artisi]

djcbgn; How about posting in one forum - not 2.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[QualityTime]

"....This project was developed by another engineer at our water system who has since left and the VFD installation is almost complete so there are no other options to consider for the original problem(like impeller trimming or control valves)...."

You still can trim the impellor and run at full speed on the VFD. Much simpler. Just check to make sure that you are on the good part of the pump curve for different combinations and you are not shutting in any pumps