Pump RPM to achieve flow 1

[nucleareng78]

I am a fairly new engineer and have a pump replacement project. The pumps we are using are Hayward Gordon screw centrifugal pumps and I need some help on the pump curve.

After reading the pump curve (matching the design flow and head), I found that it should operate at 750 RPM. After getting a quote from the vendor, they set me up with a standard 900 RPM motor...

Question, do we need a VFD then in order to get the desired flow rate and head needed for the design? We are not wanting to add margin to the system with additional flow or head. I'm so confused about this and need help!

Below is the link to the pump curve. We are wanting to operate at 3750 GPM and 20 ft WC. The current pumps we are replacing say "Variable RPM" on their pump curve so I'm assuming they also used a VFD???

http://www.haywardgordon.com/documents/variableXCS12-C.pdf

Thanks for the help! I'm still trying to learn being a new engineer...

 

[Artisi]

simple answer - do you need VFD is NO.

The curve you have posted, is this the curve the pump vendor supplied with their quote, or is it the curve for the installed unit?

Are you operating at 50 or 60 Hz.

The curve you supplied indicates 750 not 900RPM for the duty specified.

At first look I would say the pump is oversized and you should be looking at a smaller unit running at a higher speed, unless of course the appliccation dictates a slow running pump.

Why are you asking in here, surely the pump vendor would be the one to ask as to WHY.

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.)

 

[nucleareng78]

It's a 60Hz, and the curve shown is for the new pumps. The reason why lower rpm is hood is that this impeller is very large since its a screw centrifugal pump. Are you telling me they should have a 750 rpm pump? I heard the 3one phase 60hz motors only come in certain RPMs, but correct me if I'm wrong.
Why would the pump curve indicate "Variable" then? I much rather ask here first before making myself look dumb in front of the vendor...

 

[Artisi]

You post is somewhat confusing, are you replacing pumps with the same pump brand / model or are the Hayward Gordon screw centrifugal pumps a complete new selection?

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.)

 

[nucleareng78]

The haywards are a completely different pump and motor. The Hayward does have very similar horsepower to the old Weir pumps. Flow rate and head must be fairly identical as well. I can't tell if the old brand had a VFD..

 

[Artisi]

We already have the flow and head requirements, 3750 GPM (US or UK gallons)@ 20Ft head - what are you pumping?

Why are you looking at centrifugal screw pumps - any good reason?

Why can't you tell if the original pumps were VFD?



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.)

 

[TenPenny]

When you say the Hayward is set up with a 900 rpm motor, is it direct driven? If so, then the pump willrun at 900 rpm, so you'd want a vfd. If it's belt driven, or through a gearbox, it may be set up for 750 rpm operating speed.

 

[Artisi]

As with most post there is a wealth of information not supplied, luckily we are mind readers.


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.)

 

[nucleareng78]

The pumps are handling condenser cleaning balls (1-2" sponge balls), thus with our given flow rate/head we needed a screw centrifugal pump.

The pump is directly driven due to space constraints. I figured we'd need a VFD, but wanted to get others opinions to make sure (thanks TenPenny).

The spec they sent me said 750 RPM but in the motor description it says 900 RPM....thus I would need a seperate VFD. It's just that most of the time for such a small motor it isn't worth the cost of having a VFD. The VFD isn't located on the pump but on the breaker, which is why I'm having a hard time confirming where it's at.

On the other hand we could also adjust a control valve to get the hydrualics we're trying to acheive. Operating at 900 RPM may give us too much flow and head.

 

[Artisi]

"The spec they sent me said 750 RPM but in the motor description it says 900 RPM" Did you ask the pump supplier WHY -- always a good place to start. (as per my first post)

60Hz 10 pole is only 720RPM with probably a lesser speed in operation - so duty is not achievable.
60Hz 8 pole is 900RPM also a little less in operation.

As a mind reader / psychic - possibly they are offering 60Hz 8pole speed with reduced inpeller -- again, ask the pump supplier

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.)

 

[nucleareng78]

It's a 60 Hz 8 pole speed motor. Since it's a screw centrifugal impeller, it may be quite difficult to reduce the size of it.

If you're trying to insult my intelligence, stop. I post on here to learn and get advice and not to be made fun of.

 

[Pumpsonly]

The pump curve is stated as variable speed because for this type of pump, the impeller is always supplied in full size. see attached picture of such pump.
You either use VFD, belt drive or gear box to get the RPM to meet the performance you need if the required RPM can not be met with the standard motor speed.
For your required flow of 3750 GPM at 20 ft, you need to run the pump at 750 RPM.
For 60 Hz supply, a 10 pole motor has a synchronized speed of 720 RPM only. Therefore you have to use a 8 pole motor which has a synchronised speed of 900 RPM. Actual RPM will be less to to slip.
At 900 RPM the pump will give about 35 ft head at 3750 GPM or about 5700 GPM at 20ft head.
If space is not a constrain you can use a higher RPM motor and reduce to the pump speed by belt drive of a speed reduction gear box.

If you can not perform discharge valve throttling and also can not accept higher flow rate, you will have to use VFD.
To check what will the flow you will be getting without valve throtlling, you need to plot the system curve on the pump curve.

 

[psv1990]

You can not get a motor that will operate at 750rpm. To get your speed control look first to a belt and pully system. This will allow you to match the pump speed to your application and allow you to change the pump speed reasonably easy. A gear drive option followed by a VFD are your next options. If you use a throtling valve, you may be short of motor power and have to upsize the motor as you are operating the pump 20% faster.

 

[nucleareng78]

In order to get a lower flow couldn't we adjust a butterfly valve downstream of the pump? We may not have the room to have a pulley system for the pump motor.

Once the pump is operating it will not be using the entire 40 HP. The design flow requirements is 3740 gpm but the pump must be able to produce 4200 gpm if the system is "dirty". We're pumping small sponge balls here.

 

[psv1990]

You will have to watch the wear on the downstream valve as they are not designed to throtle flow on a continuous basis. The wear goes up with more material you are pumping in the fluid. More information for your consideration.

 

[nucleareng78]

psv1990, good point. I am going to see if these valves are currently used to control flow or not.

The pump's impeller cannot be trimmed or modified as it is a screw impeller (huge). I'm thinking we can either have the control valve adjusted or run a hydraulic calculation to confirm that the additional flow/head is not a huge deal.

 

[GHartmann]

I wouldn't think you would want any type of throttling valve downstream your pump because of the "sponge balls" you are pumping.

It doesn't seem that impractical to me to add a VFD. Many times the VFD drives are installed in the MCC lineup close to the motor starter circuitry (see if you can find out from your electrical folks). There are many units where the VFD and start-stop "bucket" are located in the same enclosure.

I don't think anyone was picking on you about contacting the vendor. Asking questions about the vendor's bid is a good way to address any of your concerns. Most vendors a more than happy to answer your technical questions. Maybe you have a policy where yoou need to involve the purchasing folks in the discussion.

Good luck!

 

[nucleareng78]

Yea a throttling valve could get eaten up by the sponge balls. I also don't think the VFD is impractical but our electrical guys says he likes to stay away from them because they sometimes give you maintenance issues. But for this system, being able to control the flow may be beneficial is there is a clogged sponge ball collector at the suction of the pump.

The vendor just told me "if you plan on using a VFD, it will have to be supplied by others." I just don't get why they would point to me on the pump curve where I want it to operate at but then give me a 900 RPM pump, leaving me out to dry. The purchasing people at the plant will not be too found of a VFD due to their cost and additional engineering cost to evaluate and spec one out. One thing would be to ask the vendor if VFDs had been used on these motors before, because some motors overheat when hooked up to a VFD.

thanks for all the help so far.

 

[BigInch]

Any alternatives here?
Has it been determined that this pump must not be operated at rated motor speed?
Is cleaning a temporary situation, or is cleaning a full time operation?
Can it not be operated at a higher head and flow for short periods of time while cleaning, then turned off?
Will it product too much head, or too much flow, or both?
Why can you not clean at a higher head or flow?
Why do you need a valve in the main discharge line? Is it possible to use a bar-tee and valve to recycle back to suction to reduce head and flow while keeping the main discharge line clear.


"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[nucleareng78]

From the P&ID it appears there is a line connecting the discharge and suction bit I need to look up what type of valves are between, it may be the head adjustment im.looking for. With an increase in flowit shouldn't negatively affect the system but they may fire back and ask to evaluate this increase or flow on the fiberglass piping we're connecting to.
Our system runs full time but if it were to shutdown it would not be a big deal. These balls clean the condenser tubes.

We're also adding additional piping to the suction line to prevent aviation and flashing that is occurring. A 5 ft run will help prevent this, which having additional flow may be somewhat necessary.