Has anyone heard of any illeffects of running submersible pumps (100hp)(460v) on VFd's. Lead length <30'. I have only seen one setup this way and it is burning up motors prematurly. ;q

There are a lot of papers written about the long term effects of running non-inverter duty motors from inverters. The details are too long and complicated for a quick post here, but yes, it happens all the time if proper steps are not taken to address the issues. Submersibles in particular are vulnerable because the motors are typically special purpose designs specific to each pump manufacturer, so they are not necessarilly designed with inverter duty in mind.

PWM drives create a phenomenon called "standing waves" that can create high voltage pulses on the motor leads that travel down to the windings and damage the insulation. Newer motor designs are using higher voltage insulation to combat this, but not all submersible manufacturers are using it yet.

You can also have circulating ground currents in between the rotor and case, causing discharge through the bearings and races that ultimately lead to premature failure. Again, inverter duty motors are now designed to combat this, but those designs may not have caught up with the submersible manufacturers yet.

Both problems can be avoided for the most part, but neither cn be ignored. Do an internet searh on "AC electric motor standing wave VFD" and "AC electric motor bearing damage VFD" and you should find a lot of good reading.

Hello Eaglepump

Welcome to the forum

In central Canterbury New Zealand, where I am, there are hundreds of submersible pumps operating on VF drives without problems. In this region, the pumps can be several hndred feet down and range up to 400KW.
A big part of the problems that I have seen with VSDs on submersible pumps, is due to slow rmp times on the VSDs. It is very improtant with submersible pumps to get them up to pumping speed within 2 - 3 seconds. If you have a 30 second ramp, you will have problems.
The answer is to use a VSD with a fast intial ramp and then switch to a slower ramp once the pump is up to pumping speed.

There are other issues with standing waves etc, but I have not seen these as a major problem on the local market.

Best regards,

Hello Marke,

We are about to use a VFD (380 volts rating and 350 hp) for a 250 hp 2080 volts 60 hz 300m deep submersible pump.

We are planning for a sinus filter at the output of the vsd and then a stepup transformer (380 > 2080 volts ).

Now i have few queries,

a. Does the steup up transformer need to be a Auto transformer? ( what we have available is a Y-Y transformer)

b. You mentioned that the initial ramp should be 2-3 secs, but to avoid mecahnical problems and inrush current, i was planning for 15 sec initial ramp,

Can you elaborate on why higher ramp times create problem to motor?

Are you using double acc settings ? ( Acc 1 and Acc2 )?

c. What about the deceleration should it be same i.e 2- 3 secs?

Cheers,

Hello chaterpilar

a I would tend to use a delta star transformer, but a star star should be OK. If you have a sinewave filter on the output, you should be able to use any transformer.

b Many submersible pumps must get to pumping speed in two seconds. A slow ramp up to pumping speed will damage the motor.

c The motor must not speed much time at less than pumping speeds.
Best regards

Thanks Marke for your inputs..but can you please elaborate on why a slow ramp-up say of 10seconds would damage a motor (250 hp 3600 rpm) driving a multistage submersible pump?

If there is no sinus filter used, how would a Y-Y transformer at the output of the drive effect VSD/motor.?

Cheers,

Hello chaterpilar

Many of the submersible pump manufacturers specify that the pump must be up to pumping speed within two seconds.

I would talk to the supplier about the Y-Y transformer.

Best regards,

One of the main reasons why the pump manufacturers specify short ramp times is that the pumps use upthrust and downthrust bearings. The hydrodynamic lift created by the pumping action lifts the pump shaft off of its resting place on the downthrust bearing. If you don't have lift, the pump is rubbing on the bearing without lubrication and wearing it out.

One of the advantages of using a VFd is he slow ramping of pumps to reduce the mechanical stresses on the pump, but will 0 rpm to 3600 rpm in 2 secs not defeat that?

Cheers.

Hello Chaterpilar

Yes, that would, but the requirement is typically This is usually the minimum speed setting that would be used in a closed loop system. It is the speed where you have a small positive flow. On a 60 Hz machine, it may be in the order of 40Hz, so probably around 2400RPM.

Best regards,

Thank Marke for the inputs.