analog signal interference from VFD output

[Pitt123]

We have recently found that when we run a new 480V motor from a VFD were are seeing a distortion on a nearby analog signal. These particualr shielded analog cables although run in seperate conduit do pass through a pull box that has the 480V motor feeders in it. This pull box is only a couple of feet. There is also a couple off feet inside the drive enclosure where these two cables are in the same proximity?

I would appear that the distortion on the analog signals is being caused by the VFD ouput on the motor cables. My quesion is though since these two different cables are only in the same proximity for a couple of feet in two places is this enough to cause this distortion with this minimal exposure? Whay type of coupling could be occuring here?

Secondly, besides seperating these cables at these locations, is there anything else that can be done to eliminate this noise on the analog input? Are there additional circuit technicques that can be used to get rid of this noise?

 

[Jraef]

The distortion is most likely caused by relatively intense RFI from the VFD cables. Think of a very powerful FM (Frequency Modulation) radio transmitter antenna inside that box!

Did you properly ground your signal cable shields, i.e. only at one end? A common mistake is that people ground both ends, which defeats the purpose.

Is the VFD cable shielded? Another good reason to use VFD cable. Not absolutely necessary, but it would have helped to prevent this.

Does the signal cable break out to a terminal block in that box? If so, make sure the shields continue through but do not ground, unless that is the only ground on either end of the two cables.

More good info on this topic, click here.

 

[ELA]

What are the parameters of the analog signal?
0-5V, 4-20ma?

When you say you are seeing distortion what does that mean?
The reason I ask is to be able to quantify the level of improvement achived when you make some changes.

Another possible interference issue can be shared grounds (or ground loops).

Does the VFD have a control signal that shares a common with the analog device? Separate dc power supplies or common?

One thing you could try is to apply a fixed resistance ( or substitute analog transducer) directly at the analog devices inputs (removing the remote cable feed).
Is the analog device now stable? One would expect this to be the case but it is worth trying just in case the problem is not radiated interference but instead conducted.

 

[Pitt123]

Jraef very informative article.

I have confirmed that the analog cable shield is indeed only grounded at one end. However I was told that when we ground both ends of the shield the noise on the analog signal goes away. We are going to try to move the cables however I'm still intersted in alternative solutions.

A couple of other pieces that the article mentioned that I have on the drive system are the output ferrite core reactor, as well the drive is connected to a high resistance grounded isolation transformer. The drive is a very large drive (700hp) and therefore has (2) power units with each power unit having all three phases going through a ferrite core reactor before it it tied together on the bus going out to the motor. Should the equipment grounds also be run through these reactors?

The input to the drive has an RC filter of some sort which I belive is an EMI filter.

From the article it looked like maybe we could run the analog cables through some sort of air core (ferrite) reactor to help with the problem if it still persists?

ELA

The distortion we are seeing is a spike on the analog signal.

Can you explain what you mean when when you talk about adding a fixed resistance at the analog input?

 

[Cmdr_Suds]

I suspect what your fighting is a "common mode noise" problem. This occurs when some of the high frequency power is getting into the ground system and finding a number of low impedance paths back to your drive. This often causes the electronics inside your analog transmitter to get wiggy(very technical term) Try simulating the analog source with a hand held simulator that has no ground reference. If the signal is then stable, a loop isolator maybe what you need. Another thing to try is to use a shielded drive cable. This helps by providing a very low impedance ground path back to the drive for that stray noise instead of ground system.