VFD Application 1

[jes1983]

I am trying to get more PSI out of a lube oil pump. The pump is running at 78 PSI. The motor is a 480 v, 47 FLA 40 hp motor rated at 3600 rpms. It is operating at 3525 rpms and 38 amps. I need approx 89 psi out of my pump, based on my calulations I need to drive the motor at 3772 rpms and 47 amps to get this psi. My question is, can I do this with a VFD? Will I damage my motor running it at this RPM?

 

[LionelHutz]

I don't know if you can run the frequency/speed up that much while still supplying 480V. If you used a 600VAC source and VFD and set it up "pretending" the motor is a 600VAC, 75hz motor then you should be able to get that amount of overspeed out of if. You are talking a 7% speed increase which should be OK for that motor. If you don't go over the rated current of the motor it should be OK electrically wise.

 

[dpc]

Running at 3772 rpm probably won't bother the motor, but it will still be only a 40 hp motor. If you need more than 40 hp at the shaft of the motor on a continuous basis, you'll probably need a bigger motor.

 

[CJCPE]

The motor will have no problem with the RPM. The potential problem will be increased torque. You stated the rated RPM as 3600, but that would be the synchronous RPM. The rated RPM marked on the nameplate should be about 3500 RPM.

If the motor is running at 3525 RPM, it is operating at about 75% of rated torque. Rated torque would be 60 lbs-ft and operating torque would be 45 lbs-ft. If the speed is increased to 3772 RPM by increasing the frequency without increasing the voltage the motor will be capable of operating at about 40 Hp without being overloaded. At 3772 RPM, that would give you 55.7 lbs-ft of torque.

Assuming that you are driving a centrifugal pump, an increase is speed will require an increase in torque about proportional to the square of the speed increase. That would mean that you need about 51.5 lbs-ft of torque. Since that is 92.5% of the torque you should be able to safely produce, you should be OK. The motor should operate at not more than the rated FLA.

 

[DickDV]

jes1983, the procedure is simple. First, change the speed or frequency limit in the drive up to 3800rpm. Under manual control, run the motor up over 3600rpm watching the motor amps as displayed on the drive keypad. You can keep increasing speed until the motor amps reaches the nameplate value of 47 amps.

At that point your load horsepower and the motor horsepower are equal and no further increase is possible. I am assuming that the pump flow rate is the natural flow rate of the system when you do this test. Depending on the pump design, lower or higher flow may throw this equalibrium point off so be sure the flow is where it normally would be.

You will not hurt the motor to do this within the limits we set above. In fact, you can probably go as far as 4500rpm with problems but you will likely run out of hp long before that. The motor will not overheat as long as the amps stay at or below nameplate.

This actually is becoming a fairly common method of optimizing motor/load behavior and is one of the nice benefits of operating with an inverter.

 

[electricuwe]

Buy a VFD with Space Vector Modulation. Such a VFD will be able to increase the output voltage a little bit so that you can run the motor with slightly higher speed and voltage. If the increase in voltage is enough depends on line voltage tolerance you have to consider. If the speed is increased without enough voltage available the motor will not be able to supply the required torque.

Using a drive with Active Front End will help if this is not enough. Such units usually run at a higher DC-Link voltage than VFDs with diode Frontends.

LionelHutz proposal would be one option but it is quite expensive and exposes the motor to voltage stress it was not designed for.

One other common method is to reconnect a star connected motor in delta configuration but unfortunately the ratio you will get does not fit for your application. You would end up with the current beeing to high.

 

[waross]

How about a set of 0-480-600V autotransformers connected in open delta?
This will boost the three phase voltage to 600 volts. You will be able to use a 600 volt VFD, and maintain the V/Hz. ratio.
Depending on the load, this would give you a little more horsepower than running at 480 volts. The motor should see about 514 volts at 3772 RPM. Not a problem as the V/Hz. ratio will be correct.
respectfully

 

[jraef]

Buy a VFD with Space Vector Modulation. Such a VFD will be able to increase the output voltage a little bit so that you can run the motor with slightly higher speed and voltage.

Yes but, you cannot increase the voltage above what is being supplied! You would need to boost the line voltage as per waross' suggestion. if you can get a simpler buck-boost transformer that only gives you the 514V as waross said, it may allow you to stay with a 480V class VFD instead of going to 600V class. Most 480V VFDs can withstand +10% which would be 528V. If you boost beyond that you will need the 600V class drive which will cost more.

http:/

http://www.jraef.com

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[DickDV]

Good grief guys! In my opinion, you are murdering the subject with ridiculous complexity.

First, a 40hp NEMA motor is almost certainly wired delta internally so changing from wye to delta is not going to be possible.

Second, by the time you buy autotransformers and a 600V drive, you've got to be awfully close to buying a 50hp 480V drive and motor and doing it the usual way.

Third, if I am reading the original post correctly (not always a safe assumption!) the poster was wondering what more he could get out of his EXISTING equipment since the motor was not at full load at base speed.

Finally, 460V 60Hz motors are, these days, under considerable stress to develop normal operating lifetimes as they are. Increasing the V/Hz curve upward to 500+ volts and correspondingly higher frequencies is probably going to cause short motor insulation life.

 

[itsmoked]

DickDV; CJCPE seems to have answered the OPs question clearly and succinctly. (Yes it should work. Do it![My opinion.])

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[quark]

Lube oil pumps are generally positive displacement type and you can't increase the pressure by simply increasing the speed.

 

[waross]

Hello quark;
Anyone with an oil pressure gauge in their vehicle will not agree with you.
respectfully

 

[thewellguy]

When you state tht it is an oil lube pump, could you be more specific? eg an oil cooled pump, and oil lube line shaft turbine. The reason I ask if that pressure could be increase by keeping the pump at the same speed and restircting he output. Also I would be concerned about the bearings in this application. Are the rated for the increased speed?

 

[itsmoked]

Hi thewellguy; +200 rpm is not, in any way, going to matter for bearings.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[quark]

waross,

Thanks. I made that comment instantaneously(by intention) but have been trying to find the logic behind that. Couldn't get any till date. Equally, I am not sure whether the oil pressure increases only due to acceleration in your example. I will look for the possible answers but I am lazy at the moment.

This is not to support my earlier post but to caution the future readers about my half baked idea :-(

 

[waross]

Hi quark
A constant displacement pump, by definition, delivers a set volume with each revolution. The pressure is determined by the viscosity of the fluid and the restriction of the flow path.
The higher the flow through a given restriction, the higher the pressure. The higher the viscosity, the higher the pressure.
In an automobile, and many positive displacement pump applications, there is a pressure relief valve. When the relief valve opens the restriction is reduced and the pressure is limited at the pressure set by the valve.
As the fluid (oil) heats up, the viscosity reduces and the pressure drop across a given restriction decreases.
I appologise for my terse reply earlier. I have more time today.
respectfully

 

[electricuwe]

I have to admit that jraefs comment to my post is correct. Space Vector Modulation will not solve the problem. The only chance to get more output voltage from an inverter (with diode frontend)than supplied at the input is going into overmodulation.