Stray Voltages from VFDs in Dairy Sheds

Incorrectly selected and/or installed VFDs on turntable drives and vacuum pumps are causing problems with your herd. If your cows are agitated on the platform, lean away from the bum rail, do not milk out properly in reasonable time, suffer problems with mastitis and have an abnormally high cell count, then you may have stray voltages caused by a VFD .
Solutions are simple, but you must use the right equipment and it must be correctly installed.

There are now many experiences that illustrate the effects of stray VFD voltages in cow sheds and there have been severe losses in production and animals in some herds where the problems have been particularly bad.

Measurements

Measurements in a rotary cowshed are not difficult, but you need to have a good oscilloscope that is isolated from the supply, and a terminating resistor of around 300 ohms. This is beyond the kit that the average electrician would carry. A very simple test is to measure the voltage between the bum rail and metalwork on the turntable with the terminating resistor in circuit. High frequency voltages in excess of 0.25 volts will begin to have an unsettling effect on some animals.
While there is no definative threshold voltage where problems occur, it is obvious that least is best and voltages above 0.25V do begin to cause discomfort.
Traditional measurements for line frequency stray voltages will not detect or display the High Frequency stray voltages resulting from the Variable Frequency Drives, so a null result from these meters does not eliminate HF stray voltages which are very short and peaky, much like a very fast operating electric fence.

Solutions

The stray voltage is a function of both the noise generated and the containment of that noise.
The first step is to ensure that the VFD selected is an inherently quiet (electrical noise) VFD. Ideally, the VFD will be filtered to a domestic level of conducted EMC emmissions, but as an absolute minimum, it must be compliant with the industrial emmissions levels.
There are standards and tests covering the conducted EMC emissions and these are covered by the CE mark in Europe and CTick in New Zealand and Australia.
The difficulty with VFDs, is that conducted noise emissions are dependent both on the equipment, and the installation.
*  A good VFD and a good installation will give good results.
*  A good VFD and a bad installation will give bad results.
*  A bad VFD and good installation will give bad results.
*  A bad VFD and a bad installation will give worse results.
The level of noise is often determined by the weakest link in the system, so you must have a good VFD and a good installation to minimise problems.

Ensure that:
a) the equipment has been tested to a minimum of the industrial conducted emissions standards and
b) that the installation fully complies with the manufacturers specifications for that VFD to comply with these emissions standards.

The minimum requirement for all VFDs is that the cable between the VFD and the motor must be an EMC screened cable with the screen correctly terminated in the VFD (no pigtails) and in the motor (screened gland). If there are switches or overloads between the VFD and the motor(s),these must be mounted in a metal enclosure with the screened cable glanded or clamped into the enclosure and out of the enclosure so that the screen is continuous from the VFD to the motor.
Some VFDs will require input EMC filters and these must be installed as close to the input of the VFD as possible and the earthing between the VFD and the filter must comply with the manufacturers specification. Some VFDs will also need an additional filter on the output of the VFD in order to comply.

The test certificates for the equipment will detail the levels achieved under test, and if you select the VFD with the lowest levels, you can potentially get the best results.

Example 1. Poor production and major animal health issues.
Vacuum pump wired in unscreened cable.

Vacuum pump wired in screened cable.


In this example, the oscilloscope waveforms were measured between the bum rail and the edge of the platform and are indicative of the voltage level sensed by the cow.
With the unscreened cable, the voltage levels were in the order of 1.2 Volts Pk. The same installation with the cable replaced by a screened cable yielded stray voltage levels in the order of 0.15V Pk measured the same way as the previous measurement.
The result of the change, was from very agitated animals with prolonged milking times and health issues, to very content animals and a major increase in production.
The only change made was the cable from the vacuum pump motor to the VFD output.

Example 2. Cow shed Turntable motors wired to turntable VFD in neutral screened cable with 500mm pigtails at the switchboard wired back to the earth rail in the switchboard.
The voltage measured between the bum rail and the platform edge was 0.84V Pk.
The pigtail earthing was removed and the screen was clamped to the gear tray (with paint removed) directly under the VFD and the stray voltage dropped from 0.84 down to 0.18 V Pk.

Unscreened cables with VFDs