4-20 mA isolator

V0N_hydro · Feb 9, 2022

I have a 4-20 mA loop that runs parallel to a 230 kV power line for less than a kilometer and the analog input on the PLC keeps getting smoked. There was a surge suppressor on it but no isolation.

Can anyone recommend an isolator that will provide +24 out to the sensor and provide isolated 4-20 to the PLC?

something like
https://www.acromag.com/shop/signal...-o/4-20ma-isolators/340i-isolator-dc-powered/

James Mcquade · Feb 9, 2022

here are my thoughts.
with a distance like that i would use shielded cable, up a size of wire for every 200 ft, a kilometer i would use #8 due to voltage drop(a guess on my part),
on the sensor end - strip the shield and wrap it around the outside a minimum of 5 turns, then heat shrink it. ground the other end to the plc panel ground.
you might consider ferrite beads as well. i believe you are getting a ground loop induced charge in the wiring. i have been known to be wrong. someone else may have better ideas.
james

TWS · Feb 9, 2022

This will provide loop isolation for the PLC input: https://iotmart.advantech.com/Data-...l-Conditioning-Modules/model-ADAM-3014-AE.htm.
You still need surge suppression on both ends of the cable. https://rspsupply.com/p-32467-polyphaser-ix-2dc24-dc-line-surge-protector-ix-2dc24.aspx.

GaryS · Feb 10, 2022

in this case I would not run any conductors past the high voltage the induced voltage will be a problem with any signal
4-20ma were never conceived or designed to run that distance under the best of conditions in the first place
I would convert the ma signal to a digital signal preverbally ethernet in separate panel. Then run fiber optics between the 2 panels

danw · Feb 10, 2022

GaryS said:
4-20ma were never conceived or designed to run that distance under the best of conditions in the first place

4-20mA came about before my time, so I can't confirm or deny its conceptual \design intentions, but I've seen dozens of one mile long 4-20mA signal cables in the steel mills.

The problem here is running parallel to a large conductor for a long distance, not the length in and of itself.

V0N_hydro · Feb 10, 2022

GaryS said:
in this case I would not run any conductors past the high voltage the induced voltage will be a problem with any signal ...
I would convert the ma signal to a digital signal preverbally ethernet in separate panel. Then run fiber optics between the 2 panels

Hi Gary,

I think the best solution is to run AC power to the remote site with isolation transformers at either end and then the remote sensing site has a battery charger, and battery for its own local DC system to power the sensor and ethernet switch with fibre to transmit the signal back to the main facility and the PLC there gets the sensor reading digitally over the fibre.

There are still issues with ground potential differences that can cause a safety hazard by the cable armour connecting the sensing site to the remote ground

diat150 · Feb 11, 2022

Id swap to a wireless radio shot to pass the signal. Id imagine that would be cheaper than re-running cable or fiber.

mendonsy · Feb 11, 2022

Maybe you should consider running the signal over fiber optic cable?? Here's a link to one supplier: https://www.fiberopticlink.com/prod...alog-signal-420ma-fiber-link-system/#diagrams

ArcReactorKC · Feb 12, 2022

I would definitely second the idea of moving to an RF signal. Water telemetry sites have been doing it for decades. Technically the 4-20ma will be much faster but is that speed necessary for the application?

danw · Feb 12, 2022

Most times wireless depends on whether there's power at the far end, the field sensor end. The advantage of 2-wire, loop powered 4-20mA transmitters is that the power comes from the 'control' end, not the sensing end and 4-20mA can run long distances.

With wireless, both the radio and the sensor need power. ISA 100 and HART battery powered wireless is very marginal (and expensive) at kilometer distances. Banner has thermocouple and RTD battery powered 900MHz field radios (in the US) that will operate at a kilometer, but the other process variables sensed by loop powered transmitters (pressure, flow, RH, pH) drain batteries like crazy using 4-20mA signals.

Bit_Bucket_07 · Feb 13, 2022

James Mcquade said:
here are my thoughts.
with a distance like that i would use shielded cable, up a size of wire for every 200 ft, a kilometer i would use #8 due to voltage drop...
james

That sounds like extreme overkill to me. In a 1 km loop (.5 km cable length) using #18 wire the total wire impedance would only be about 21 Ohms. Voltage drop is not really an issue, since the transmitter regulates current, rather than voltage, which Ohm's Law insists is the same throughout the loop.

GaryS said:
in this case I would not run any conductors past the high voltage the induced voltage will be a problem with any signal...

I agree. Rather than patching at the problem with surge protection and isolation, I'd try to get rid of the problem. Either re-route the signal wiring away from the high voltage power line or use fiber optic cable or radio.

GaryS said:
4-20ma were never conceived or designed to run that distance under the best of conditions in the first place...

I disagree with this. The distance isn't the problem. Induced voltage is the problem.

Bit_Bucket_07 · Feb 13, 2022

Steel jacketed armored twisted pair cable could be a solution to the problem, although it would be a heavy cable to run overhead on poles.

A copper or aluminum shield will never eliminate induced voltage on signal wiring to the extent that a ferrous metal like steel conduit will.

4-20 mA isolator

V0N_hydro

Member

James Mcquade

Member

TWS

Member

GaryS

Member

danw

Lifetime Supporting Member

V0N_hydro

Member

diat150

Lifetime Supporting Member

mendonsy

Member

ArcReactorKC

Member

danw

Lifetime Supporting Member

Bit_Bucket_07

Member

Bit_Bucket_07

Member

Similar Topics