IR Test Liquid Resistance Starter 5

[freetown]

Hi all,

We recently installed a Liquid Resistance Starter in replacement of grid rotor resistance for 2 motors 11kV/2700kW.
With grid rotor resistance, we could make a rotor to earth insulation test using Megohmmeter insulation tester (we test rotor+cables+resistances connected together). The value to earth we find can go from 500 M/Ohm to 1G/Ohm (with 500V test).

With liquid Resistance Starter when perform the same test (cable+rotor) we only find around 200 K/Ohm. We thought that there is a problem on cables or rotor, but when disconnect cables of Liquid Resistance Starter contactor and testing cable+rotor we find the normal value (500 M/Ohm to 1G/Ohm).

The motor is running without problems, according to the manufacturer the value we found (K/Ohm) when testing is normal. It’s because of liquid and moving electrodes.

My question is it normal we find this small value between rotor to earth and what’s the exact reason for this value?
Every time when want to test rotor, we have to disconnect cables from liquid Resistance Starter contactor, is there a way to test the rotor without disconnecting cables as for grid rotor resistance?
We want to measure rotor current, is there a device allow us to measure current and torque of the rotor in real time?

Thanks.

 

[Gr8blu]

With the grid resistor method, there is a switch (contactor) between the rotor side of the circuit and the resistor bank. This remains open until closed, so a simple insulation resistance test at the rotor will only test the rotor, not the grid.

The liquid rheostat ALWAYS has the liquid in the tank (unless there is a BIG problem!). That means there is always a direct (tenuous) connection between the rotor side of the circuit and the (typically) grounded tank wall. In terms of an insulation resistance test with the liquid rheostat in circuit, you're really measuring the resistance through the liquid, more than the insulation in the machine.

There is no good way to test just the rotor insulation without disconnecting from the liquid rheostat. You could always install another contactor/breaker between the two to help isolate the rotor side for testing purposes, but that quickly becomes another potential point of failure. Most folks don't bother, because they test relatively infrequently.

Converting energy to motion for more than half a century

 

[Gr8blu]

freetown My apologies - I didn't see the last couple of questions in your "to do" list in the original post.

Yes, there are ways to measure the rotor current and torque in real time. Are they cheap? No. Easy to use? Sometimes.

Direct torque measurement usually means setting up some sort of strain gauge system at either the shaft or coupling. It can be done at the motor itself, or on the process side of a gearbox for example. If done on the process side, you then have to account for both the torque amplification from motor to load (which may be increasing or decreasing) and any inefficiencies in the gearbox itself. On the other hand - what you're really interested in is the torque to the process, so there's that argument too.

For measuring current - depending on rating, dedicated current shunts and/or Hall effect device (Rogowski coil) should do the trick for an individual phase. Might have to do a bit of mathematics to get the total 3-phase effect, if your data acquisition system doesn't do it for you.

The final thing to consider is that your data acquisition occurs in small enough time steps to be useful for what you're trying to accomplish. For a steady-state value, it's pretty easy. For recreating the performance under actual start-up conditions, the time increment between data samples is a lot smaller.

Converting energy to motion for more than half a century

 

[waross]

If an external shorting contactor is used, CTs may be used to measure the rotor current.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[freetown]

Thanks for the help

I was thinking to install a manually operated switch-disconnector between the contactor of the liquid rheostat and the rotor but as you said, it will become a test point and a potential source of failure.

The problem we do not have any monitoring on the rotor of the motor. When something happens on the rotor, we have no measure to understand what’s going on.
From your experience, what is the best tool we can use to follow-up the rotor? Using CTs and remote the current to our scada system?

 

[Gr8blu]

Well ... what kinds of failure modes are you experiencing (or at least think you might experience?). Some things are identifiable by a number of means, others by only a single method.

I find it hard to believe that if your machine is large enough to warrant a liquid rheostat starter/controller, there is no monitoring of any type on the rotor. As a minimum, there should be vibration probe(s). There is possibly also some sort of brush wear indicator mechanism, even if it's as simple as a colored piece of plastic (i.e. when the color disappears it's time to change the brush).

Most likely failure modes in the rotor of a wound rotor induction motor are going to be: 1) vibration-induced fatigue of current-carrying components (e.g. brush shunts, coil-to-coil connections, coil-to-collector connections), 2) thermal-induced fatigue of those same components, 3) insulation degradation in the winding due to thermal cycling, 4) insulation degradation in the winding due to mechanical abrasion (one coil against another, or one coil against a lamination, or even impact from a foreign object), and 5) insulation degradation from moisture or other contaminant (which may or may not be actually abrasive). Additional failure modes include mechanical fatigue of the shaft or shaft-to-core structure, and corrosion of the exposed surface(s).

Converting energy to motion for more than half a century

 

[freetown]

We use 2 or 3 cables for each rotor phase.
Failure modes could be low insulation resistance of cables or the rotor itself.
We have vibration probes on motor bearings.

 

[edison123]

Slipring rotor voltage rarely exceeds 2 KV. Most of the rotor faults happen during starting when the rotor sees full voltage and there is a flashover across sliprings due to insufficient ring to ring clearance, accumulation of carbon brush dust etc. Such faults are picked by stator overcurrent protection. Any rotor winding earth fault/ phase to phase fault is also picked up by the stator overcurrent protection. I have not seen any rotor fault protection in any slipring motor since the rotor voltage is practically zero at full speed.

Measuring the rotor current with normal CT's won't work since the rotor running current is at slip frequency. You need hall effect clamp-ons to measure such low frequency currents.

Electrolyte salt solutions have low IR values by nature and there is nothing to worry about. I have seen violent arcs in the LRS during starting and then the motor runs without any issues as the electrodes are closed to a short.

Muthu

http://www.edison.co.in

 

[freetown]

We had rotor winding earth faults before during starting and stator overcurrent protection reacted but the rotor is damaged or has a low insulation resistance in some cases after that.
The only thing, that we don't have data on the rotor to analyse what happened because no measurement (specially current) exists.
That's why i asked if it's normal for such motor or we use in general some type of measurement to observe the rotor.

 

[edison123]

OP

How would you put in an electrical protection in a rotating component?

Except for OV/OC/OS/OT, which are early warning systems, other electrical protections like EF/Differential etc. will not stop a winding fault from happening. They are there only to prevent further damages to the motor.

A well-made machine with good maintenance and operation practice will give many years of trouble free service.

As a rewinder, most of the winding failures I see are stator related and only a few are wound rotor related. Guess why.

Muthu

http://www.edison.co.in