Best way to implement redundant cabling on 4-20mA

DLMUK · Jan 23, 2019

Hi,

I have the following setup:

4-20mA sensor -> PLC Input module

What I would like to do is have a primary signal and power cable, and a redundant signal and power cable.

With the power cable I could connect into the same terminals without concern, what I am unsure of is if there is an issue passing the 4-20mA current down two cables and terminating them on the same input terminal?

I would terminate the cable shields at the same earth terminal too.

Thanks

tragically1969 · Jan 23, 2019

DLMUK said:
Hi,

I have the following setup:

4-20mA sensor -> PLC Input module

What I would like to do is have a primary signal and power cable, and a redundant signal and power cable.

With the power cable I could connect into the same terminals without concern, what I am unsure of is if there is an issue passing the 4-20mA current down two cables and terminating them on the same input terminal?

I would terminate the cable shields at the same earth terminal too.

Thanks

Is that really going to solve any problems, if the primary power cable gets shorted then so does the secondary by virtue of the fact they are paralleled ?

cardosocea · Jan 23, 2019

DLMUK said:
Hi,

I have the following setup:

4-20mA sensor -> PLC Input module

What I would like to do is have a primary signal and power cable, and a redundant signal and power cable.

With the power cable I could connect into the same terminals without concern, what I am unsure of is if there is an issue passing the 4-20mA current down two cables and terminating them on the same input terminal?

I would terminate the cable shields at the same earth terminal too.

Thanks

Do your instruments have space for two glands?

DLMUK · Jan 23, 2019

I see the point about shorted cables, yes it has space for two glands.

cardosocea · Jan 23, 2019

DLMUK said:
I see the point about shorted cables, yes it has space for two glands.

Not sure it's what you want to hear, but would a current splitter near the instrument do? It would increase the cost, add failure points but would let you run two cables per instrument.

tragically1969 · Jan 23, 2019

DLMUK said:
I see the point about shorted cables, yes it has space for two glands.

just to be clear, is this single instrument, single PLC input ?

You can sort the issue easily with Diodes/Fusing on the power, same problem with the analogues if you did connect them like that though.

DLMUK · Jan 23, 2019

tragically1969 said:
just to be clear, is this single instrument, single PLC input ?

You can sort the issue easily with Diodes/Fusing on the power, same problem with the analogues if you did connect them like that though.

Yes, single instrument and single PLC input.

damica1 · Jan 23, 2019

If it's because your concerned about a cable "Break", just add in the plc program the ability to issue an "ALARM" event if reading would drop below 4 ma. This would be a sign that maybe the cable has broke.

If you have to have redundancy then add another sensor and use another channel then monitor both channels.

DLMUK · Jan 23, 2019

damica1 said:
If it's because your concerned about a cable "Break", just add in the plc program the ability to issue an "ALARM" event if reading would drop below 4 ma. This would be a sign that maybe the cable has broke.

If you have to have redundancy then add another sensor and use another channel then monitor both channels.

My concern is data integrity not cable break detection. Therefore I need to ensure I have a redundant signal. The sensor goes into a clevis arrangement so it is not as simple as just adding another sensor. Hence why I am looking at the cabling options.

thanks

danw · Jan 23, 2019

Just 'electrically' parallel the conductors/cables.

A 4-20mA loop has a single conductor between the power supply to the the transmitter to the receiver input and back to the power supply.

If you install cables with 2 conductors and connect both conductors to the same point in the loop, you have done have achieved "conductor redundancy". Or you could use separate conductor/cables.

DLMUK · Feb 13, 2019

danw said:
Just 'electrically' parallel the conductors/cables.

A 4-20mA loop has a single conductor between the power supply to the the transmitter to the receiver input and back to the power supply.

If you install cables with 2 conductors and connect both conductors to the same point in the loop, you have done have achieved "conductor redundancy". Or you could use separate conductor/cables.

In this setup would creating a short in either set of cabling not induce a short the whole system?

I could put fuses/diodes at one end of the cables but not the other very easily.

danw · Feb 13, 2019

a short circuit from where to where?

JesperMP · Feb 13, 2019

My concern is data integrity not cable break detection.

?
What do you mean by "data integrity" ??
It sounds as if you want to have a better failure tolerance.

Exactly what kind of measurement is it ?
If this is a Pt100 or similar that is in a pocket or tube, then you can investigate if you can have 2 Pt100 sensors in the same pocket or tube. Then connect each Pt100 via separate cables to separate PLC inputs.
If measuring static pressure, you could have a tube with 2 attachment points for pressure transmitters.
Etc..

DLMUK · Feb 13, 2019

danw said:
a short circuit from where to where?

For example if you have two cables, and one gets damaged causing a short between say the 4-20mA and +VDC conductor. This will also be present on the second cable as they are essentially connected at the terminals at either end.

Therefore without putting fuses/diodes at the terminations at both ends, if one cable becomes damaged the second cable won't be of any use anyway?

VAN · Feb 13, 2019

DLMUK said:
For example if you have two cables, and one gets damaged causing a short between say the 4-20mA and +VDC conductor. This will also be present on the second cable as they are essentially connected at the terminals at either end.

Therefore without putting fuses/diodes at the terminations at both ends, if one cable becomes damaged the second cable won't be of any use anyway?

I would assume a true redundant system would have a different input (different input card or rack). Landing to the same input seems more problematic than anything else.

Best way to implement redundant cabling on 4-20mA

DLMUK

Member

tragically1969

Member

cardosocea

Member

DLMUK

Member

cardosocea

Member

tragically1969

Member

DLMUK

Member

damica1

Member

DLMUK

Member

danw

Lifetime Supporting Member

DLMUK

Member

danw

Lifetime Supporting Member

JesperMP

Lifetime Supporting Member + Moderator

DLMUK

Member

VAN

Member

Similar Topics