VFD design for centrifugal pumps 1

[pittengineer]

I am putting together a design for a VFD controlling two 40hp motors on centrifugal water pumps which are pumping water from a tank.

Only 1 motor ever runs at a time and the other motor is strictly sitting on standby as a backup. The idea is to control the speed of the motor and thus the flow of water from the tank depending on the water level in the tank. Water is continously being pumped into tank but tank is not that big so the level tends to change quickly.

A couple of questions regarding the design:

1) As mentioned was planning on using 1 VFD for running both motors since 1 is always a spare. I figured the best way to do this would be to have two contactors on the ouput of the drive which would be controlled to determine which motor should run. Anyone have any experinece with this type of setup?

2) For these 40 hp drives will there be a signifigant cost savings by only using 1 drive with contactors to control both motors or is there not enough of a savings with this setup to justify not purchasing two individual drives?

3) Any suggestions on things to look out for when sizing a vfd for a centrifugal pump application? Should I just match the hp requirement of the motor?

4) Whats a quick way to roughly gauge the cost savings by varying the speed of the motor as opposed to keeping the motor at constant speed and modulating a valve to control flow?

5) Planning on putting this drive on a devicenet network. I believe the drive should be able to distinguish commands on which of the two motors to run as well as provide feedback on which motor is running through this Dnet interface. Anyone have experience with this.

I'm working with the vendor on designing the control schematis, but just wanted to see what the experts thought about these questions.

Thanks in advance.

 

[davidbeach]

So, you're going to put in two of the reliable things and only one of the least reliably of all the things you listed. Why bother with two pumps? Are you prepared to run a pump direct on line?

 

[waross]

I suggest you use interlocked contactors. Reversing contactors would be a good choice, but there may be one two many or one two few overload relays depending on how you configure it.
If you use the overload relay supplied with the reversing contactor you will have to add a second overload relay for the second motor. I recommend his configuration. If a motor trips, you may automatically go to the other motor until the cause of the trip has been investigated and repaired.
If you use the overload function of the drive, you won't need any overload relays on the contactors. But, if a motor trips there may be some confusion as to which motor has a problem especially after a power failure.
I would use an individual overload device for each motor with a high setting on the drive as a backup.
The reversing contactor will have contacts that may be used to provide status information to the control interface. In this instance the "sealing" contacts (#2 and #3 on the forward contactor and the corresponding contacts on the reverse contactor) will not be used for control and may be used for status indication.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[pittengineer]

I apoligize but I realize now I forgot to mention a key piece to all of this in that these two motors are already connected to individual starters running across the line. I am wanting to retrofit this setup to incorporate a vfd.

Davidbeach brings up a good point in that with only 1 drive we will be eliminating the reliability of having the two motors. Yes these drives are capable of being run across the line, and maybe I need to factor that into the control scheme if something happens to the drive. I was just thinking that only purchasing 1 drive would be a signifigant cost savings, but now I'm not sure. The VFD will be located right next to existing starters.

I'm guessing since the drive will be located near the existing starters that I can use these existing starters. Maybe I now keep the starters and overloads intact as waross mentions, and somehow come up with a way to run these motors through the starters across the line if something happens to the drive.

 

[pittengineer]

Just thought of an idea to run the motors DOL if there is a problem with the drive.

Where the drive is located, or in the drive enclosure I will add a 3rd contactor which esentially bypasses the drive if there is a problem with the drive. The contactor will be on the load side of the thermal magnetic breaker protecting the drive so it could bypass the drive and supply 480V to the line side of both of the contactors. Since we are keeping the overloads in the starters as waross suggested we would have a starter arrangement that would be capable of running the motors in a DOL arrangement.

Not sure if it is all code compliant but just a thought that I had.

 

[waross]

Bypass would be safer with two contactors. Disconnect the drive before going to bypass. If you want to go all out, if your logic scheme detects a drive failure and goes to bypass and DOL, you could divert your level control signal from the drive to a control valve to avoid short cycling the pump.
Consider also adding a second breaker. A breaker trip may embarrass your plan for redundancy, a second breaker feeding the bypass will maintain redundancy.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[DickDV]

One other suggestion, arrange the control circuit for the two starters fed from the drive so that they cannot be operated or changed while the drive is running.

VFD's sometimes incur damage to their outputs when a motor-side contactor is opened under load.

Rather than spend the money for bypass contactors to cover for drive failure, simply arrange your power leads so they will reach around the drive to the output leads. That way, if the drive should fail, you can simply rearrange the power feeders to get going again.

 

[pittengineer]

Waross

Good point about adding a second breaker for redundancy on the bypass contactor. I was thinking as well that if one of the motors failed and tripped the breaker it would leave us dead in the water.

I am awaiting on pricing for just the standard drive to see how much each drive costs to see what the potential savings is by only using 1 drive. Anyone still think it is worth the cost savings at this size drive to have all the extra control complications (not even involving the DCS system yet) as opposed to having a second drive?

 

[LionelHutz]

Personally, I don't think the added complexity and hassle of a multi-motor switching scheme is worth the savings. You are contempating a VFD with a full-voltage bypass that will operate 1 of 2 motors by mixing some of the old components in with the new. The control and power wiring compexity to pull this off just creates a big maintenance and operation nightmare.

I'm not sure where the the liquid is pumped when it leaves this tank, but it's most likely that a VFD installation won't be more energy efficient. If there is no fixed head requirement then you might see some savings if you can run the pump with the speed speed reduced enough for enough of the time. I'd suggest you go over to the pump forum to discuss this more.

The other part to judge is the maintenance costs - if you are having maintenance issues that quickly add up to the cost of the VFD then might save some of those costs.

Overall, controlling the level of the tank with a VFD still sounds like a good process control application more than anything else.

 

[pittengineer]

Thanks for all the responses guys. After reading all of the respones it sounds like trying to use one drive to control both motor will be too complex to integrate into the existing control system and may only lead to problems with maintnance and operation. I am still awaiting pricing on drives.

For these 40HP centrifugal pump motors, do you guys think that I should use a 50HP drive to give a little head room, or do you think the 40HP drive should be large enough? I dont have much experience with centrifugal pumps so wanted to see what you guys thought as far as extra headroom?

 

[pittengineer]

I just recieved a quote for one of these Powerflex 70 50Hp drives at about $17,300.00. This included the drive mounted in a NEMA 4 enclosure with an HMI mounted on the front of the enclosure. It also included a 3% line reactor installed in the cabinet on the line side of the drive as well as an 120V I/O input interface, and DeviceNet interface module.

I thought maybe this price seemed a bit high but I have not purchased too many of these. Is there a large difference in cost between a 40hp and 50hp drive?

At this cost we would be saving at least $10,000.00 if I were going to stick with the single drive design and have to purchase some additional contactors and circuit breakers. Not sure if this would still make it worth it.

 

[electricpete]

Whats a quick way to roughly gauge the cost savings by varying the speed of the motor as opposed to keeping the motor at constant speed and modulating a valve to control flow?

There is currently a pretty long current discussion with some links on this topic here:

http://www.eng-tips.com/viewthread.cfm?qid=260203&page=1

User BigInch has a few thumbrules of situations where he thinks vfd won’t save money.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[itsmoked]

I think that price is about oh, um.. 11,000 dollars ridiculous!

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

MA7200-4050-N1 TECO MA7200 AC Drive, 50 HP, 460V 3 PH Input, 460V 3 PH Output, 80FLA, NEMA 1
$2,299.00
MA7200-4060-N1 TECO MA7200 AC Drive, 60 HP, 460V 3 PH Input, 460V 3 PH Output, 96FLA, NEMA1 $2,699.00
MA7200-4075-N1 TECO MA7200 AC Drive, 75 HP, 460V 3 PH Input, 460V 3 PH Output, 128FLA, NEMA 1 $2,989.00

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

In agreement with itsmoked.
Sorry, I hit the submit button prematurely on my previous post.
Here is some additional info from:

http://www.factorymation.com/s.nl;j...e3eSc3aTah0Le34Pa38Ta38MaNj0?sc=2&category=11

MA7200-4040-N1 TECO MA7200 AC Drive, 40 HP, 460V 3 PH Input, 460V 3 PH Output, 64FLA, NEMA 1 Est. Availability 12/16/09
$1,899.00

KDRULC1L TRANS-COIL AC LINE REACTOR, 3% IMPEDANCE, 40HP, 460V, 52 AMPS, CHASSIS UNIT $220.00

KDRULF2L TRANS-COIL AC LINE REACTOR, 3% IMPEDANCE, 50HP, 460V, 65 AMPS, CHASSIS UNIT $258.00

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

A "drive in a cardboard box" is a far cry from a complete packaged combination drive as described above. I do however think the price is a little excessive (I'm in that business so it would be unethical for me to quote, but let's just say you can do a lot better).

No reason to over size the drive by the way, it gains you nothing in this kind of application and in fact will cost you more for installation as well. That's because by code, the conductors must be sized based on the VFD maximum current rating, not the load connected to it. So if you buy a VFD sized for 50HP, the conductors must be sized for 125% of THAT highest rating, rounded up. So picking on that Powerflex 70, that 50HP rating is the largest size in that series and is their "Normal Duty" rating, which is likely what the rest of the world would call "Variable Torque" (I love AB, they make their own rules...). But for "Heavy Duty" applications (Constant Torque), that is a 40HP drive. More importantly though, the maximum output is 65A, even though your motor is probably closer to 52A, which is coincidentally the max rating of the next drive size down. Yet the conductors for the 65A drive will need to be sized for 81A instead of 65A, and rounded up to the next largest standard size may push you two cable sizes higher.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[DickDV]

Yes, it's time to shop around. And, a 40hp normal duty ("variable torque" as jraef mentions) will be big enough for your motor and pump.

That optional HMI is costing you a lot of extra money. Since you want a NEMA 4 drive, let the drive keypad be your display.

I wouldn't be surprised if you cut your cost in half.

 

[pittengineer]

I jsut recieved a quote for the same drive in a 40hp and it still came in at $15,960.00 which is high. I'm probably going to do just as DickDV mentioned and get rid of the additional HMI if I pursue this drive.

Do you guys think that after removing the HMI I should be paying under 5k for this combination, or more between 5k-10k for this combo. I want to stick with the Powerflex family to keep all the drives the same on our DeviceNet network, as well as keep all the drives the same for spares.

One other thought is weather or not I'd need a braking resistor for this application. Is a pump typically a high regenerative load that would justify ordering a brake resistor.

 

[itsmoked]

You should be around 5K. You may never get there with PowerFlex. "Make your bed and sleep in it", as the saying goes.

Regen on a pump? Not no, but &*%%(^&* NO!

Keith Cress
kcress -

http://www.flaminsystems.com

 

[panelman]

Hey, for $17k we’ll build it in the UK and come over and install it too.

Is the HMI you talk about a remote drive keypad?