EMC Compliance and VFDs

[Marke]

I was recently at a discussion on VFDs and EMC compliance and was surprised that one supplier claimed that their VFDs complied fully with the European EMC standards without screened cables on the output. "Internal filters" block the high frequency from the output. The data sheets claim dv/dt limited to 500 - 800 V/uS. I believe this is much too high to eliminate EMC issues without screening.
Interested in your comments.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[Valvecrazy]

That was probably one of the companies that claims 20% to 50% energy savings with their VFD controls. We know that is not true so, why would they be any more truthful about the EMC problem? They are in the business of selling Drives. They will tell you whatever you want to hear to accomplish their task. Most of these problems are of such a technical nature that very few people really understand what they are talking about. "Stray Voltage" cannot be seen, so you are just supposed to forget it is there.

VFD's are what manufacturers call a "fluid system". I used to think that meant that they were used to move fluids. The following is a quote from a business meeting by one of these companies where they were comparing VFD's and other things to the Razor Blade business.

"Fluid systems can be disposable products that need to be replaced on a frequent basis. This provides a stable revenue base and a high percentage of repeat sales. In addition, in many applications there is a tendency to continue using the same manufacturer's elements due to performance history, research protocol, standard of care practice, canister design or system specifications. This can translate into a steady annuity stream of revenue for fluid system companies."

The public at large still believes that companies are trying to build quality products that help save energy. In reality, companies are more concerned about building "fluid systems" into their product line.

To answer your question, I have found that no matter how you filter, ground, or even use steel conduit, that cows at a Dairy with VFD's, will still wince from the shock as they try to get a drink. Cows can tell if an electric fence is on by just putting their noise close to the wire. When the milk production goes down 30% to 50%, the EMC problem has not been solved, no matter what the VFD manufacturer says.

 

[gepman]

I am not an expert in EMC, EMI, or EMF. I think I would consider which company is making the claim. The best that you could do however is use some of the suggestions (devices) in your last thread regarding this issue from Skogsgurra and itsmoked and see if you can measure any qualitative difference between similar installations of this brand and a different brand.

My gut feeling is that the dv/dt stated is too high also and you need to look at the carrier frequency. If they haven't changed the value of the carrier frequency to a lower level the motor leads will still be acting as a broadcast antenna (although the VFD manufacturer probably is only concerned about what is radiating from the VFD itself, not from the motor leads which is probably far greater an issue).

Now in response to Valvecrazy, I would not expect too many other people in this forum to have much dairy experience. Dairies are a small part of an energy efficiency program that I work on now and I am in the middle of the largest dairy operations in the United States and the world. The dairies here have between 2,000 and 6,000 milkers and VFDs are almost universal in application for their energy savings on the vacuum systems for the milking and on the milk transfer pumps from the milk receivers to the milk coolers. I have heard of milk production problems due to stray voltage but I have to disagree with you, it is a problem with grounding and not an EMC issue. The electric fence that you refer to probably has a peak voltage around 10,000 volts and will arc around .2". The cows put their nose near it and sometimes it will hit a whisker. They may be able to tell if their is an electric field but the voltage is much higher than you see in a VFD installation.

Sorry for the digression into cows.

 

[ozmosis]

valvecrazy
The EMC Directive and more specifically, the Adjustable Speed Drive product standard EN61800-3, will not necessarily solve EMC problem. It simply limits conductive emissions, limits radiated emissions and sets about defining levels of immunity for products in a similar environment.
The standard defines a number of categories that allow end-users to 'choose' the level of emission and immunity they require for their particular installation and network. The VFD manufacturer has to then provide a range of filters and installation advise that then ensure when the user is to install the said component (the VFD) then it will comply with the understood environment it is intended to be fitted.
The 'standard' is concerned about the impact it will make on "someone else on the same (public) supply as you". In a nutshell, it is not too bothered about what you do "in your own back yard" as long as you do not interfere with somebody else.
This is why it is virtually impossible for a VFD supplier to say "they comply with the EMC standard" What category in the standard? How do they know where it is going to be installed?
The VFD cannot comply on its own. It is a component of a system. The system is defined as a Power Drive System (PDS) and this means that compliance is making sure the VFD has the correct filter to meet the environment it is to be fitted then it should meet conducted levels accordingly. The radiated levels will be determied by the cable length and the enclosure around a drive. It is highly unlikely an IP20/IP21 drive will meet any of the radiated levels of emission unless it is mounted inside a metal box.
Having unscreened cables is again, unlikely to allow a PDS to meet levels of radiated emissions.
If someone is to mount a drive without screened cable then I would say it will not meet the requirements of C1, C2 and maybe even C3 but it depends where it is going.
Who will complain?(the cows) Managing the Directive is complaints driven.
moooving on... (ok, sorry)
Mark
I don't think there are many drive companies who will say they do not need screened cable to comply. Maybe a short cable in an environment and power network that has a whole number of galvanic barriers between it and the public.

 

[Valvecrazy]

Dairies are just one of the applications where drives give me problems. I just use the cows as an indicator of the problem. Vacuum pumps for milking and milk transfer pumps are a good application for a Drive. With enough grounding and filtering you can keep these systems from causing much of a stray voltage problem. The problems come when you add more VFD's to the system. You already have one or two VFD's in the milking room. Adding more Drives to the water supply pumps, feed grinders, etc., causes an accumulation of the stray voltage that is hard to control. It takes less than 4 volts to cause the cows to wince and not drink water as they should, which is no way near the voltage of an electric fence.

I have the same problem with radio controls, cordless phones, cordless mice, television reception, and other things in high rise buildings, schools, and residences. The cows at the dairies are just what clued me to the problem. As I have said before, there are many good applications for Drives. When even more Drives are used to control other things, which can be just as efficient and as easily controlled by other means, this adds to the stray voltage problem.

The trick is in knowing that things like centrifugal pumps can be controlled better and just as efficiently without VFD's. Many people think that the VFD is the perfect cure, and use them to control everything. Then when they start having problems with stray voltage, bearing currents, water hammer, and other things, they try to band aid the problem with extra grounding and filtering. Instead they should remove the Drives for systems that could be controlled more effectively by other means, and the problems will go away.

 

[Marke]

Here are some suggested installation methods for installing VFDs on pump installations and minimizing EMC issues.

http://www.lmphotonics.com/vsd/vsd_05.htm

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[itsmoked]

Valvecrazy;

I ask you to look at your posts.. You are basically ranting on, and on, and on. Please understand you are becoming a serious nuisance... Nuisances are not long tolerated here. Look at your post here. Someone asked about EMC levels. Your response is completely useless in this regard. You need to start being more circumspect before participating in every thread that mentions VFDs. Patience is running out.

I want you to understand that you have some good points in the valve verses VFD arena but showing up and yanking every thread into a valve verse VFD rant is doing your cause no good.

Might I suggest you write a clear, factual, data backed, FAQ - name it cleverly, and tag line it to your posts. Then give the subject a rest and DO NOT bring it into threads where it provides no value, like this thread.

Regards.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Valvecrazy]

I said I would leave you alone to debate the second best option. Someone brought up this older thread again, I didn't say another word. However, I don't think everybody feels the same way about this discussion. I guess I was encouraged by the following quote from a related post that got 5 stars.

"This is easily one of the most valuable threads in the Motor and Controls forum."
"it has been a refreshing experience to see dedicated experts discuss these matters so in-depth."

Many people who work with VFD's don't like what I have to say. I can certainly understand and I apologize but, it doesn't make it any less true.

 

[controlsdude]

If you do not have screened cables, does the manufacturer claim that you have to run each vfd cable in seperate conduit?

Is the VFD in a central cabinet with other controls and the motors are away by some distance from the cabinet?

I can see where mounting the VFD locally with the motor does cut down on the chances of emitting EMC. But it always seems when I see this that the electricians for the most part put in a short piece of flex, which is not necesarily metal. But if your cable is still shielded or screened then this does not matter.

Does this manufacturer refer to this practice?

I know most lit that I see on VFD installs point out the importance of grounding being passed back to the main panel ground and back from thier. I remember someone stating that the distance of the grounding from motor to VFD was critical in reducing EMC. But if the VFD was mounted locally would not the return to ground of the emc be shortened due to the shorter distance?

I guess I ask too many questions, but these would be what I would need answered from this manu?

 

[Skogsgurra]

Hey Dude!

What do you mean by "the electricians for the most part put in a short piece of flex, which is not necesarily metal"?

Sounds like an awful mistake. Like something being practiced in plumbing business. Are you sure it is being done in VFD installations? If so, why?

Gunnar Englund

http://www.gke.org

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

 

[controlsdude]

I believe its ok to use flex conduit if the piece is not above a certain length. Of course sometimes customers do not like this anyway depends on specs. I believe the NEC is ok with this but sure where this is stated.

 

[gepman]

Skogsgurra
The only place where I allow (reluctantly) the use of non-metallic flex is for my dairy processing (milk, butter, ice cream) clients. They continually wash down and they complain that the metallic flex inside liquidtight flex conduit eventually rusts. I say replace it every few years since it is not that difficult to pull back the motor leads but it is the standard in some food processing plants. As long as the conduit is adequately bonded on both sides of the flex and your carry an EGC (equipment grounding conductor) it meets NEC. I personally feel that the non-metallic flex gives out in less time than it takes the liquidtight to rust.

 

[controlsdude]

when I look in the nec looks like 6 feet or 1.8 meters is the standard max length on flex conduit.

 

[Skogsgurra]

OK. I thought the flex was part of the grounding. Which it obviously isn't.

Gunnar Englund

http://www.gke.org

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