4-20 mA Shielding 1

mgopalan · Feb 3, 2004

I have an application where I have a 4-20 mA transducer at the end of a 250 ft long cable. Plus, this entire setup up is in a nasty environment electrically (lots of RF and generator noise, think oil rig with gigantic generators, motors and nasty power).

I am using a 4 conductor (2 twisted pairs, one pair not used) cable with a foil shield and a drain wire and I have connected the shield at one end only (it is floating at the sensor side) and have connected the other end of the shield to my Analog Ground. I am also using a 400 Ohm sense resistor and a unity gain follower before I get to the A/D.

This setup works great in the lab and fails miserably at the field location. The DAQ system goes wild with noisy (as in white noise up to the rails of the A/D input levels) data...

I have tried... Common mode chokes on the 4-20 mA lines, I have made sure that I am not connecting my Analog Ground to the Case Ground (Earth)..

so the question I have is this...

Do I connect the shield to the Analog Ground of the 12V DC power supply (and therefore the AGND of my sense resistor, Buffer Amp and the A/D as well) or to the Case Ground which is tied to the safety Earth of my AC power supply and the EARTH/GND pin of my 110VAC connector..

The shield itself only protects against Capacitively coupled noise right? The common mode chokes should clean up any common mode noise...so why am I getting crappy data? This is driving me nuts...

Any help?

MG

pablo02 · Feb 4, 2004

What did you do with the unused pair? Typically, I will also ground any unused pairs at the same end as the shield is grounded (don't know if that helps any, it's just a practice I have always done...) Don't have a lot of experience in handling the rough environment you have, so I can't provide any other help...

ScottyUK · Feb 4, 2004

mgopalan,

What is the route your instrument cable takes? Long parallel runs alongside cables carrying heavy currents will cause problems. Ideally you should have separate instrument and power routes on their own cable racks.

You will find it is easier to sceen E-field interference (capacitively coupled) than B-field interference (magnetically coupled). The screen of your cables should cut out the bulk of the E-field stuff if it is grounded. It will be next to useless at screening B-field interference. The best solution to reduce magnetically couple dinterference without spending a lot of money is to get as as far away as possible from sources which have high magnetic fields: high-current cables, generator stators - especially around the ends, transformers, large motors.

Good luck, you have a challenging time ahead of you.

maximka · Feb 12, 2004

Go fiberoptic with converters at both ends.

hacksaw · Feb 16, 2004

you might try a smaller sensing resistor (100 ohms) and since you have an extra pair, paralleling your wires (lowers loop resistance).

common instrument shielding being mylar with a very thin foil is only good for electrostatic shielding. it does very little good against induction effects. you can also play with gounding the un-used pair - treating it as a partial shield for the active pair.

JesperMP · Feb 17, 2004

1. Use a cable with foil + braid (+ drain wire). Foil only is not enough.

2. Experiment with grounding both ends of the cable shield. (This can stir up a debate, I have seen recommendations for: Only at provider, only at consumer, and at both ends).

3. Try to lay the cable away from noise generating equipment or cables.

4. Do you have noise on other PLC channels when the sensor is NOT connected ?

5. Check up on the ground connection in the cabinet. There must be a prominent earth wire on the ground bus where your cable shield is connected.

hacksaw · Feb 17, 2004

grounding the cable at both ends is not a good idea given the interground potentials that exist in most operating plants. we have seen differences in ground potentials of 150 vac between different areas of the plant. the causes are many and cannot always be fixed, many times they are transient in nature, as when a motor is started, but not always.

capacitive coupling tends to dominate in high resistence circuits, induced currents in low impedence circuits, with many variations between. use of a 400 ohm current sense resistor is unusal for most 4-20 ma systems. never the less if one is used it will aggravate noise pick up due to capacitive coupling. we all agree foil shields are not always effective, but then neither are braided one.

proper cable routing is always a good idea, but not always an option. Agree on checking adjacent channel noise and the requirement for a proper signal ground.

good luck

obg · Feb 18, 2004

Just a thought 400 ohm @20mA =8v from a 12v supply are you sure your sensor can work from 4v - the losses in the cable. you may be on the wrong track with the rfi issue.

JesperMP · Feb 19, 2004

400 ohm ?
Its more likely to be 4 ohm with 2 x 250 feet wire distance. Of course it depends on the cable, but it is not 400 ohm.

But I didnt notice the 12V supply voltage before. Most 4-20mA trasnmitters expect something in the range 15-30V. Please check if the transmitter can work from 10V (to take into account the voltage drop due to the cable).

JesperMP · Feb 19, 2004

obg,
I didnt notice the "400 ohms sense resistor" that mgopalan mentioned in the original post (OOPS). You are quite right, this cannot work.

mgopalan,
if you MUST use a voltage input in your DAQ, then do yourself a favor and use a real signal conditioner (for example 4-20ma/0-10V) in stead of some do-it-yourself circuit.

jghrist · Feb 23, 2004

Grounding the shield at one end provides shielding for capacitive coupling but not inductive. If the noise is high frequency, you could ground the field end through a capacitor that would be a high impedance to power frequency but a low impedance to the noise. This would provide better inductive shielding at the noise frequency. Another option would be to put the cable in steel conduit.

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

4-20 mA Shielding 1

mgopalan

Mechanical

Recommended for you

pablo02

Electrical

ScottyUK

Electrical

maximka

Electrical

hacksaw

Mechanical

JesperMP

Electrical

hacksaw

Mechanical

obg

Electrical

JesperMP

Electrical

JesperMP

Electrical

jghrist

Electrical

Similar threads

Part and Inventory Search

Sponsor