VFD ESP application

I think maybe you do not understand how an AC induction motor works.The motor can NEVER spin at the exact same speed as the frequency applied to it. You must always have slip (the difference) in order to create torque,

Motor nameplates give a rpm rating based on a no load condition. Once the motor is under load it will not spin at the rated rpm of the nameplate at the rated hertz. The drop is fairly insignificant.

ie 1700 rpm @ 60 Hz no load

1680 rpm @ 60 Hz load present

I assumed when you said output frequency is 3450 you were referring to the motor's rpm not the actual frequency.

There are some things you left out. How much of a difference is there? How are you measuring the motor rpm? Did you preform an auto tune with the drive? If so, was the motor under a load during tuning? Did you double check the nameplate info you programmed into the drive? You know the usual suspects.

I always understood that rpm ratings were at rated load, and slightly less (by percentage of slip) than synchronous speed.

For instance, synchronous speed of a 60 Hz 2-pole motor would be 3600 rpm. At full load, it might be 3500 rpm. Unloaded, it would run at something like 3590 rpm (load never being exactly zero, on account of friction.)

Thanks for the reply! The drive is set for 60hz, when auto tuned, converts 60hz to motor rated speed. Auto tune to nameplate rpm (3450), full load amps, # of poles, voltage rating, and kw. No rotate tune. Drive display scaling is set to 3450. Set rpm at 3450. With display scaling set in rpm's, drive converts hz to rpm. 60hz should equal 3450rpm. Start pump. output frequency read from drive (U1-02) shows 3450. Motor speed from drive (U1-05) shows 3324. Not sure how drive gets this reading. Pump depth at 1200'. 1200' of drop cable have anything to do with it? I have PC pumps, motors on surface, convert hz to PRS(polish rod speed). Set to 300prs, motor speed shows 300prs. Thats why I'm confused.

I think you're confused but you're not the first one I've heard who had the same line of thinking. The motor's rpm or speed is not the same as frequency. Check your motor's name plate again. It should have frequency of 50 or 60Hz. That's the figure that you should put into your drive not 3450.

The motor is rated for 60hz, this is plugged into the drive during the auto tune. I do a no rotate tune. Plug in nameplate voltage, FLA, hz, poles, rpm, and kw. I have 3 options for display scaling, hz, rpm, prs. I set to rpm. The drive then converts the hz to rpm. I think 60hz should equal 3450rpm. This might be were I'm confused. It then shows my output frequency at 3450, but gives me a motor speed of 3324. My motor speed is always lower, not sure which reading to go off of. I have PC pumps, motors on surface. Read PRS. Output and motor speed are the same on these. Thanks for the reply, if you have any other ideas or advice please let me know.

You are misinterpreting.

When you change the VFD display to say Hz or RPM etc., you are just changing the scaling of the display, which is based on the basic motor speed formula: RPM = 120 x Applied Freq./ Motor Poles.

TO THAT, you have t apply the slip percentage to get actual no-load speed, then that varies with load as well. So if you WANT the motor to run at 3450RPM unloaded, you would set the output frequency to 3600.

In a vector drive, which is what it appears you have, it is measuring the slip based on feedback information from the motor's current and voltage vectors, so it then tells you what the calculated speed of the motor is under load.

So why on my PC pumps are out put frequency and motor speed the same?

They are probably not the same, unless driven by a synchronous motor. This is possible, but not likely.

Surely if you set the slip to achieve the no load speed when loaded, you set 3600rpm and get this in both cases +/- tolerance??

..and vector drives measure the current, (knowing the output voltage as this is generated by the VSD), so the slip is estimated according to the load calculation in the vector mathematical medel therein? Estimated speeds are only as good as the information and the VSD modelling techniques, but these are normally just readouts of frequency (scaled or not as you said) before or after slip is applied in the open loop case.

If you have set the VFD correctly including slip speed and you are still short of speed, either the torque calculation is wrong, so adding the incorrect amount of slip (Hz) or the motor is slipping more because of reduced motor voltage increasing its slip at the same load. So, the motor may have a 60Hz slip speed of 3450 but if presented with a lower than nameplate voltage (cable volt drop) you will get reduced speed (increased slip). So, you accept this and rescale either your setpoint or slip or voltage to get you a more predicatble result.

What do you have connected to the VFD output to allow such a long cable length? If you're using an output choke OK for this distance, you will have a volt drop across this in addition to cable volt drop.

I'll try tuning to 3600. I have #4 copper from drive to pump. 10hp motor FLA 19. 480volts coming into drive. The motor is rated for 460volts. Would this create large enough drop to increase slip? How do you figure motor slip? Should I try tuning with load? Sorry if these questions are preschool level to you all. I'm not an engineer, nor electrician.

The 480 volts is one of the nominal voltages furnished by utilities in the US and other places. Motors for this are rated at 460 volts to account for typical voltage drop (wire resistance) from the service to the motor.

Slip is another issue. To pick an example, a 3500 rpm motor operating on what would otherwise be synchronous speed of 3600 rpm is running at (3600-3500)/3600 ≈ 2.78% slip. This is quite normal. The torque produced by the motor (and the current it draws) are proportionate to the slip. If the motor is lightly loaded, the torque, current, and slip are less, and thus it runs closer to synchronous speed. (In my previous post I used the example of 3590 rpm). If heavily loaded, the torque, current, and slip are more, and the speed slower.

Most VFDs can display the frequency they are putting out, but I don't know if they can tell what the motor is actually turning. If so, that's pretty neat. If you set a VFD to 3450 rpm (57.5 Hz), the normal motor slip would give you a little over 3300 rpm at the motor itself.

I'm not familiar with #4 but I guess it may be similar cross section to 6mm² and the volt drop would be in the region of 40V for 360m distance (1200 feet) so you will be a bit shy on voltage wich you could compensate for by programming motor voltage higher to see if this helps e.g. try 500V 60Hz 19A 10HP 3450rpm as motor data. You can look up volt drop calcs for your cable locally, what current does it actually draw?

Can you also monitor actual output Hz to see if slip is added correctly for your load e.g. at 19A load (some part is extra earth leakage so reduce switching frequency of VFD to 2kHz, say) you should expect +150rpm for the slip = 2.5Hz equivalent. So if you asked for 3450 this is 57.5Hz synchronous which would actually mean running at 60Hz but full load slip is 150rpm giving 3450 output.

In real life, motor slip is normally about 70% of motor data at this power, so motor nameplate is 3450rpm but it probably runs at about 3500 at full load at rated voltage.

So, is it Hz or volts its short of?

I take it you have no choke between VFD and motor as you did not reply to this point. I would normally recommend a choke at this cable length to avoid overcurrent tripping and afford the motor some protection to the volt doubling that the insulation will experience.

Good luck and hope you achieve a decent result. Hope the info is several posts have helped.

Dear Guest. It would help if you'd included your name, however, I would point out while you are worrying about the speed of the motor, you are not looking at the pump curve.

Many other contributions have pointed out to you how you calculate the actual motor speed, and how a VFS/VSD controls the RPM of a motor.

I am assuming that as you said "esp motor" you are using a submersible pump from a known supplier, and it would have the relevant documentation, pumps curves, test results etc, etc. Looking at any pump curve for a downhole pumps, you must consider the size of pump and its production rate, either gal per min or barrels per day.

So from the pump curves you now have the flow rate of your pump at a certain RPM How do you know its producing as per the pump curve? You can either measure the flow at surface or with a downhole monitoring tool, measure the discharge pressure and adjust the RPM on your VFD to give you the BEP of the operating range of the pump.

I assume you have knowledge of the Laws of affinity, if not I have included a part from Schlumberger Technical Service Manual

60 Hertz Versus 50 Hertz Operation

A submersible motor rotates at a nominal speed of 3500 RPM at a frequency of 60 Hertz. At a frequency of 50 Hertz, it rotates at a speed of 2915 RPM. For each individual pump series, the diameter of the stages cannot be changed so the only variables available to vary performance within a series are the different types of stages, the number of stages and the speed (RPM) at which they rotate. The rotational speed can only be changed by varying the frequency. When this speed change occurs, it varies the pump's operation according to mathematical relationships known as the affinity laws.

Affinity Laws for a Speed Change

50 Hertz Capacity Flow = 50/60 X 60 Hertz Flow = 0.833 X 60 Hertz Flow

50 Hertz Head = (50/60)2 X 60 Hertz Head = 0.694 X 60 Hertz Head

50 Hertz HP Pump Load = (50/60)3 X 60 Hertz HP Load = 0.579 X 60 Hertz HP Pump Load

50 Hertz Motor HP Load = 50/60 X 60 Hertz HP Rating = 0.833 X 60 Hertz HP Rating[p/] 50 Hertz Motor Voltage = 50/60 X 60 Hertz Motor Volts = 0.833 X 60 Hertz Motor Volts

50 Hertz Motor Amps = Same as 60 Hertz Motor Amps

50 Hertz Pump Efficiency = Same as 60 Hertz Pump Efficiency

Note: These laws only apply to pump behavior - not to well or system performance.

Now before anyone comes back and asks about TDH losses, first lets not get to complicated and second TDH calculations take into consideration the required stages to lift fluid to surface and therefore the motor HP, please lets take this as a given

Guest.... Just remember that what you see on paper is NEVER what you get in practice.

and while I think about it you must also ensure that your motor speed in relation to your pump curve is not so low that you do not achieve the correct motor cooling.

An ESP motor is ONLY cooled by the velocity of the well fluid passing over it as the pump lifts the fluid to surface, so be careful of what you adjust the minimum speed to and what size pump you install. If the pump is to big then you will burn the motor PDQ as you lower the speed

Revisit the programming of Ur drive.

Dickson

Hi Guest,

I´m not and expert in electricity, but i know something about controlling those kind of Motors, i have had some troubles controlling them.

I have just one question for you,

- when you say that the speed of the motor is not the same of the VFD, what is measuring the speed of the motor?

Comment: most VFD send a feedback of the speed of the motor to the motor controller throught a 4-20ma signal (that is the one you configure to show in the HMI or display). if your VFD works like that you should make a loop test to be sure that when the VFD is stopped (0 RPM) it´s sending you exact 4mA, if this measure is over or below this (according to your comment it could be below), you should fix this, maybe grounding the shield of the cable can help you. or changing the cable with shield, beldfoild and braid shield could help you.

hope my comment can help you.

Excuse my english, it´s not my foreing language, Like celia said " My english is not very good looking",

Yohan Axel Moreno S.