hi
I have a problem with my 4-20 m amp transmitters that there is a noise with the signal wich causes unstability to the plc input cards.
what is your suggestions, please

 

Hi,

First you should check if the DC volt is under the normal range. If it is ok, I guess the reason why the noise generated may be no shield for cable.

 

Try replacing the connecting cable with a twisted pair cable. This will solve your problem. A shielded cable will also work, as long as the shield is properly earthed on one side, but to be secure the twisted pair cable will be more effective.

 

As Li says, it's possibly something to do with the shield, either not shielded or the shield is grounded at two points. Check at the PLC termination cabinet.

Lift the transmitter shields and check it to ground with an Ohmmeter it should be very high, 1 MegOhm or more. If it's OK there re ground and go to the transmitter end. It should not be grounded there but you should get a circuit 1- 100 Ohms (depending on the distance) from transmitter to termination cabinet and back through ground.

Sometimes it only takes one shield grounded incorrectly to effect all the inputs.

A few more possibilities:
Is the power supply Negative grounded? I have seen a similar problem if the power supply is floating and the common mode voltage goes outside the PLC input range.

If the supply is ungrounded, try connecting a small 24V incandescent lamp between negative and ground, it shouldn't light but may change the noise. Is it the power supply itself causing the noise? If you don't have access to a CRO, try measuring the output with your meter on AC Volts range.

If you can post a typical loop drawing we may spot something.

How much noise do you see compared to the input range?

Good Luck,
Roy

 

We always use a shielded twisted pair for our 4-20ma signals. The shield is grounded at the transmitter end only. We rarely have problems. If you have a separate power supply for the transmitter, check that with a scope to see if it is clean. If the power source is the loop, you can check that with the scope also. Don't forget to consider the possibility that the process is actually noisy; and that the transmitter is telling the truth!

Gerald Beaudoin

 

Hello,

I would ensure your shield is grounded only on one side. The ground potential should be for control systems and not building ground. They will have different potentials. Thanks.

Bill,

Check out http://www.powerplantforum.com

 

I have seen a problem with transmitters that ground one lead and because of this, upset the readings for all transmitters connected to a PLC analog input module. Worst case situation to correect this was to place a signal isolator in the loop.

A better solution is to avoid the problem by careful selection of a transmitter that provides an isolated output but in an existing installation, an isolator may be the fastest and least expensive solution.

Often, its a single transmitter that causes a problem. If you have the ability, you may want to lift the wires of one transmitter at a time (take both wires off to eliminate the possibility that one side of the transmitter is grounded) as see if the rest of the signals settle down. That may allow you to isolate the problem and narrow down the causes.

Regards,
Russ Kinner
SSOE, Inc.
Phoenix, AZ

 

Gerald,
That's interesting, I have seen the odd document that suggests grounding the shield at the transmitter end is best but I have rarely seen it done.

As an Instrument designer I am curious:
Is your plant large or small?

Do you have junction boxes in the field with home run cables?

Do you run a wire from a local ground point to the transmitter or rely on equipment/pipe ground?

What do you do in the case of a 4 wire loop e.g. Mag meter where you have the grounded pipe and an electrical ground?

On loop drawings I will show the the shields cut and taped at the transmitter. Nine times out of ten at startup I find many of the shields nicely connected.

Roy

 

Oops... I stand corrected... the grounds are connected at the PLC end and not at the transmitters. Sorry for any confusion that I might have introduced. We do have a fairly large facility and we try to make sure that junction boxes are kept to a minimum. We always try to keep the cable runs with no joints and if we must have a junction point, keep that ground junction point isolated from any local grounds that may be in the junction box, while maintaining ground continuity back to the PLC. It took awhile, but the field guys are finally used to the fact that one end will be taped off.

Gerald Beaudoin