I want connect analog 4-20mA output of analyzer to DCS, please suggest cabling standard to be used?

How much distance analog output 4-20mA can cover?

In my case we have distance of 850m between analyzer and DCS.

 

The analog output has a spec for the maximum resistance through which it can drive a full span 20mA signal.

If the analog output is passive, loop powered with an external power supply, then there might be a chart showing the resistances at different supply voltages.

If the analog output is an active output powered by a power source inside the analyzer, then the voltage supplying the loop can not be changed so there will be a value, like 500 ohms or 600 ohms or whatever. 4-20mA analog outputs rarely drive less than 500 ohms.

In any event, you need a total loop resistance less than the maximum value, say 500 ohms.

The first component of loop resistance is the analog input on the DCS. It has a resistance, typically 250 ohms.

The second component is 0.75mm2 copper wire has a resistance of 24.13 ohms/Km

Your point-to-point distance of 850m needs a double run of wire, so that's 1.7Km of wire which has ~ 49 ohms resistance.

Loop powered indicators drop voltage and are a load, but you didn't mention one.

So, 250 ohms on the DCS and 49 ohms of wire resistance is about 300 ohms, well below the maximum that a typical commercial analog output can typically drive.

For cable, use shielded twisted pair with an insulation and jacket suitable for the environment.

Your bigger problem than power will be a ground loop due to differences in ground potential between the analyzer and DCS. An active output powered by the analyzer will undoubtably develop common mode voltage due to the difference between the ground potentials between the sites. The analog signal will likely need a loop isolator at the DCS end.

If the analog output is isolated at (from) the analyzer and loop powered from the DCS, then it is not likely to need a loop isolator.

David

 

Dear Mr Kiyani

Distance can be up to 1000 feet using 18-20 gauge twisted pair cable
Shielding of cable can be selected depend on noise level and criticality of that signal. The total loop resistance should be taken under consideration some standard says it does not exceed 900 ohms.

 

As this is analyzer, your case comes under the passive loop where in instrument itself provides 4-20 mA and we do not need any DCS 24 VDC power supply. You may install a barrier or not as per your area classification. If installing barrier, wire according to the wiring diagram mentioned on the barrier and if wiring directly to FTA of DCS go through the input card manual. They must have mentioned where to connect the passive loop wiring. Distance of 850 m is not a concern at all.

 

Thank you dear david, Ashiq and rishi for clarification.

Best Regards
Kiyani

 

Could anyone help me in my below query:

We have one 10 pair cable between DCS and a package PLC. Two pairs are being used for DCS Analog output (4 to 20 mA signals) to the PLC. Is it permissible to use the remaining pairs for DCS Analog input signals (4- 20 mA signals) from PLC to the DCS? Only this cable is available and it is very difficult to lay a new cable. In other words, is it permissible to combine analog 4-20mA signal and analog 4-20 mA output signal in one cable? Thanks

 

There is no technical reason against running 4-20 mA analog input and output signals in the same multicore cable. The only reason not to would be if there was some local code requiring separate cables.

 

Hi David ,

It was really informative . Could you please help me in a different scenario.

I have a AB make Micrologix 1100 PLC with analog input, and that is 0-10V analog channel .

I have a Endress Hauser Promag 10W 4-wire flow transmitter (Active Output), and i want to connect this signal in to my Micrologix 1100PLC which is located nearly 900m away from the flow meter .

Is it possible to connect this in to Micrologix 1100PLC ...??

How can i connect the 4-20ma Coming from the Flow transmitter in to a 0-10V analog channel inside the Micrologix 1100 PLC.

 

According the E&H manual, the 10W flowmeter has an active (powered) 4-20 mA output that can drive 700 ohms. There's a reminder that if you need HART communications, that a miniumum of 250 ohms of loop resistance is required. That shouldn't be a problem.

The way one converts from a current signal to a voltage signal is to pump the current through a precision resistor to get a voltage drop (Ohms Law) with the resistor wired across the (+) and (-) terminals of the PLC analog input (AI).

A 250 ohm resistor will create a 1.0 to 5.0 Vdc signal (Ohms Law)
A 500 ohm resistor will create a 2.0 to 10.0 Vdc signal (Ohms Law)

It sounds like you don't have a 4-20mA input to use. If you do, be aware that many AI's use or expect a certain value resistance on a 4-20mA input and expect to 'see' the resultant voltage for that resistance. You'd have to check what the situation is for a 4-20mA input.

If you use the 0-10Vdc analog input, you want to use as much of the range as you can for the highest resolution of the flow rate signal.

500 ohms uses 8 volts of the 10 volt span (80%), producing the highest resolution.
250 ohms only uses 4 volts of the 10 volt span (40%0, producing a lower resolution than the 500 ohm resistor would.

Which resistor you use depends on the wire size you pick. A 900m distance requires 1800 meters of copper wire.

The total load the meter can drive is 700 ohms. The 500 ohm resistor leaves 200 ohms for wire resistance, the 250 ohm resistor leaves 450 ohms for wire resistance.

Find a copper wire resistance table to get values to calculate 1800 meters of wire resistance for various sizes of wire. Then see if your boss is willing to buy 1800m of wire of the size that has less than 200 ohms enabling the use of the 500 ohm resistor, which will give you the maximum resolution of the flow signal.

Zero flow is at 4.0mA but that's some elevated voltage at the analog input, depending on the resistor used (zero flow at 1V for 250 ohms, zero flow at 2V for 500 ohms). The analog input will need to be scaled to account for the fact that zero flow is not at zero volts on the input.

I assume that an A-B MicroLogix has sufficient computing power to scale an input with an offset.

The E&H manual says its current output is galvanically isolated. That's great because isolation minimizes the chance of ground loop between the flow meter and the controller's input.

 

Is it safe to run analog and digital instrument signals in the same mulicore cable? multicore is individually and overall screened.

 

Very much helpful David..thank you very much...

Rohan

>The analog output has a spec for the maximum resistance
>through which it can drive a full span 20mA signal.

---- snip ----

>If the analog output is isolated at (from) the analyzer and
>loop powered from the DCS, then it is not likely to need a
>loop isolator.

 

As a help to calculate the neccesary voltage to operate the 4-20 mA loop in its full span, check this online current loop voltage calculator: http://www.divize.com/techinfo/loop-voltage-calculator.html

The calculator allows to select cable size and length to calculate the voltage drop over the cable.

Regards,
Heinrich

 

sir,

i have to pass my sensor signal 4-20mA over a range of 300m. may i use 2 pair 1.5sqmm ducab armoured cable?

 

You should try the on-line current loop circuit calculator (in the post above), it's pretty cool.

http://www.divize.com/techinfo/loop-voltage-calculator.html

According to its calculation for

- a 24Vdc power supply
- 300m of 1.5mm2 copper wire (3.63 ohms)
- a 250 ohm receiver load

at 20mA of current the 300m of 1.5mm2 wire drops only 0.073V on each leg, or 0.15V for both legs (at 20mA), leaving the transmitter to regulate an internal drop of 18.85V to maintain the 20mA.

Assuming the calculation is correct, a single pair of wires will do just fine (for 24V supply and a 250 ohm load)

 

It may be possible. You have to calculate your round trip cable resistance and identify the other resistances in the loop (e.g. display, AI, etc.). You then look at your sensor's voltage/resistance burden curve or loop resistance spec on the sensor data sheet. If you cannot find one, talk to the sensor vendor as you will need to know the sensor voltage burden for 4-20 mA operation to do this (commonly about 13 volts). Please don't expect people on this forum to do this calculation for you. This is a basic skill for an instrument engineer. If you are a technician who does not have this skill, learn it or talk to an instrument engineer or the sensor vendor, maybe they will do it for you.

Please look at these application notes:

http://www.ni.com/white-paper/6940/en/
http://www.murata-ps.com/data/meters/dms-an20.pdf

There are other issues that may come into play such as increased exposure to EMI, physical protection, intrinsic safety if applicable, etc.

William (Bill) L. Mostia, Jr. PE
ISA Fellow, SIS-TECH Fellow,
FS Eng. (TUV Rheinland)
SIS-TECH Solutions, LP

"No trees were killed to send this message, but a large number of electrons were terribly inconvenienced." Neil deGrasse Tyson

Any information is provided on a Caveat Emptor basis.

 

Yes, it's done all the time however be aware that some digital signals such as large solenoids can cause interference in adjacent inputs using MOVs on digital outputs usually takes care of that. It's preferable to keep analog and digital at the same potential e.g. 24 VDC.

> Is it safe to run analog and digital instrument signals in the same
> mulicore cable? multicore is individually and overall screened.

 

Generally the driving voltage for a 4 to 20 MADC signal is 24 VDC. if you use #16 shielded instrument wire, you should be well under the resistance limit. The resistance of 850 meters of #16 wire is about 315 ohms, and assuming a 250 ohm resistor at the DCS input, your loop resistance is 665 ohms. 24 volts will drive around 23 MADC at that resistance. My concern is the cost of conduit and wire for that distance. if the 4 to 20 MADC signal is for monitoring purposes, the better option may be to use a RF transmitter. There are several very good ones on the market.

 

Hilton,

I used that on-line calculator, but did not check the values it produced. So when you reported a 315 ohm resistance value for 850m of 16g wire, and it reported 3.63 ohms, I wondered which value was right.

The table at this link says 1000m of 16g copper is 13.17 ohms:
http://africanenergy.com/new/wp-content/uploads/2012/08/awgtometricwire.pdf

There's some difference between 1.50mm2 and 16g (1.66mm2), and the distance is 300m, but in any event, a 24Vdc power supply should drive the guy's loop without any problem.

 

David,

I re-checked the math and was off by a factor of 10. keep in mind that we are talking about a current loop so you need to double the 850 meters in your calculation. so my rounded value should have been 31.5 ohms.

Good catch!