EMC Levels in dairy sheds.

[Marke]

I have been asked to hep to sort a problem in a dairy shed where cows are milked.
It is common practice today, to use VFDs to control milk pumps, vacuum pumps, backing gates and the turntable.

One of the problems with using VFDs, is that there is always a level of conducted emissions that get into the earth system and appear as stray voltages.
A cow is many times more sensitive to these stray voltages than a human and it is not uncommon to see problems in the milking shed due to the irritation caused by these stray voltages.
After considerable investigation, I have not been able to come up with an acceptable stray voltage amplitude that can applied to the dairy shed.
Does anyone have an authorative reference that gives threshold voltages that should be applied around cows?
Also, has anyone come across specific testing methods for conducted stray voltages in dairy sheds?

Reducing the emissions is not a major issue, I can give simple instructions for this, but I need a reference to show that the remedial action needs to be taken.

Best regards,


Mark Empson
L M Photonics Ltd

 

[itsmoked]

The biggest hurtle to me would be how to actually measure these stray currents or the voltages they induce in any meaningful way that you could then compare against remedial action.

I realize cows are extremely sensitive to these currents. I think it would be more than adequate to assume any measurable current to be unacceptable.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[itsmoked]

Thinking more about it.

How about creating a two legged cow out of an insulated pole with a horse shoe on each end. Then it could be picked up and set,(plopped), down anywhere. Any voltage across it would be the step potential bugging the dairy queens. You could probably, using some logic, hunt down the offending VFDs even.

I suspect another thing to check would be one horse shoe at cow length to metalwork simulating a cow leaning against a railing.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Hey Keith I like it. I think that the problem may be touch potentials rather than step potentials but I do like the idea of a skinny two shoe'd wooden cow.
Your "cow" should also work well to measure touch potentials.
Mark; It is always a pleasure to see you drop into the forum.
I remember reading a paper years ago about cows going thirsty because there was a very small voltage present on their water supply and they would not touch it.
Based on the set pints of GFIs and the extra sensitivity of the cows I would be looking for stray currents of around 1 ma.
With VFDs in the mix I would also be looking for HF currents.
Can you measure voltages and currents from motors to ground at some good installations and at some problem installations?
I suspect the final solution may be a combination of bonding to create equi-potential zones and reduction of voltage rise of the equi-potential zones above true ground. Look at the impedance of the motor grounding paths at the high frequencies associated with VFDs.
Please keep us informed of your findings.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

Front foot to rear foot can be an issue; face to foot is a far greater problem.

 

[ozmosis]

Mark
I know of no specific test procedure for cows but maybe it is the filtering itself in the VFD's that is increasing the leakage currents.
Class B filters will typically be 2 or 3 stage filter banks and the high level of capacitance could be creating the high level of leakage that the cows are picking up.
If your VFD's have a facility to 'switch out' the filter then it might be worth a try to avoid the leakage via the Y capacitor.
Alternatively, Gunnar could develop a Cow-Beppe based on the same function as the bearing currents...

 

[Marke]

Hi everyone.
Thanks for your comments.

The problem is becoming a little political and I expect that it will escalate. The major issue is that the installers believe that they know better and do not use screened cable, do not earth the motor back to the drive and have multiple earth paths. They tend to ignore the manufacturers recommendations.
If the installation is done correctly, the problem is usually not too bad provided that a good VFD is used. I usually recommend to install to the domestic category for EMC regulations and there will be no problems.
What I am looking for is some threshold or reference levels that are proven and industry accepted, that we can use as a go - nogo test. At present it is very subjective and "what do you know about these things" is the reaction that is often gained.
We have some anecdotal evidence that indicates the figures that we believe are necessary, but we lack a scientific study and report.

Have a good day,
Best regards,

Mark Empson
L M Photonics Ltd

 

[waross]

Take a look at this, Mark. It may be of some use to you.

https://www.msu.edu/user/hillman/elecshok.htm

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

Here's another one, Mark. I noticed a few things that may be significant. There were some findings that cows were more affected by or showed more reaction to intermittent voltages than steady voltages of the same magnitude. This may be an issue where a VFD motor on a "Crowder" gate runs intermittently. So far, the papers that I have found deal more with utility neutral currents travelling through the ground than with VFD induced voltages. You may be the one to write that paper before this is solved.
Another trivia, the resistance of a cow is 390 ohms and the resistance of a swine from the mouth to all four hooves is 789 ohms.

http://www.omafra.gov.on.ca/english/livestock/dairy/facts/strayvol.htm

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

Now we're getting closer to what you need.-

In Wisconsin, the ?level of concern? is derived from the 1996 PSCW docket 05-EI-115. In that
docket, the ?level of concern? is defined as 2 milliamps, AC, rms (root mean square), steady-state or 1 volt, AC, rms,
steady-state across a 500-ohm resistor in the cow contact area.

Look here.

http://psc.wi.gov/utilityinfo/electric/newsInfo/document/strayVoltage/cattle.pdf

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Scramode]

If the four legged bovines are on concrete then check NEC 547.10

 

[Marke]

Hello Bill

Thank you for your comments and research.
There is some good info there Bill, but you can probably see the issue where there are conflicting ideas as to the levels that cause problems, plus there seems to be an absense of information on the stray voltages caused by VFDs incorrectly installed.
Over here, in dairy sheds, it is common for a VFD to control the vacuum pump, another one or two to control the platform, plus one or more to control washdown pumps etc.
If these VFDs are incorrectly installed, there can be a high level of conducted emissions resulting in many volts of stray voltages. Unfortunately, many of the bovine experts believe that the emissions from VFDs are purely electromagnetic and therefore can not cause problems, and concurrently with sorting our the VFD problems, other changes are made and the VFD is not identified as a possible cause of the problem. - this is the opposite of the norm, as usually the electronics device is the cause of everything.
The general concensus over here is that the cow sensitivity is in the order of 0.3V into 1000 ohms at 50Hz while milking and this figure seems to have been arrived at over many years of experience rather than documented scientific research.
I believe that the actual issue is the peak voltage rather than the RMS, and that we would need to apply a similar peak voltage limit to the stray voltages from VFDs. These are high frequency impulses and the rms current will be low. Unfortunately, I can not identify any hard research done here. It is a definite problem that needs to be addressed, but it is in the too hard basket.
Eliminating the stray voltage is not a major issue, if the VFD is installed according to the manufacturers recommendations for EMC compliance, screened cable between the VFD and the motor with the screens properly terminated etc it is not a major issue. If the installation complies with the domestic levels of EMC installation, then there should be no problem.
Over here, even though we have EMC regulations, the installers believe that it is OK to use open flex between the VFD and the motor, and to earth the motor back to the main switchboard rather than the VFD and this creates major issues.

Have a good day,
Best regards,
Mark.

Mark Empson
L M Photonics Ltd

 

[Skogsgurra]

Hello Mark!

I recently ran a seminar on those issues (well, more bearing currents than EMI). We then had a set up with a rather long PE cable and did measure motor frame voltage against true ground. One can see the short capacitively coupled pulses reaching up to around three hundred volts during the PWM transitions. And you can feel a tickling when you touch the back side of your hand to the motor frame. But not when you grab the motor firmly.

I can imagine that as little as a few pecent of that voltage would make the four-legged creatures find putting their noses into the water less attractive.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[gepman]

Hi Mark!

Here is another research project,

http://www.strayvoltage.org/stories/index.php3?Story=20010812_researchers.inc

and another

http://edis.ifas.ufl.edu/AE019

I have seen similar problems in California but in the Tulare area, where most of the dairies are located, we have several good industrial electricians that perform a proper job of grounding and bonding.

Just don't mention this to valvecrazy, he may want to put valves on the vacuum pumps. I just recently came back to this forum after at least six months of self imposed exile after the vfd pumping harmonics thread you started a while back in which valvecrazy got involved.