Control.com - Nerds in Control

Dear all,

I need your help, we have 3 flowmeters and one level transmitter which are around 500m far from PLC (AI/AO module).

Are there any tricks or tips to connect the devices and get good signal from field inst. to AI/AO module?

Thanks for the help...

Unless you are dealing with wiring difficulties, you should have no problems getting a 4-20mADC signal back from those four devices. Even if they are two-wire devices, you should be able to go somewhat further than 500m.

One of the important things to do, first, before you spend any money on buying new products, is to check the wiring and determine why you are suffering from signal loss.

One thing you could do is to use a current isolator retransmitter (Moore Industries makes them, so do many other people, like Devar, AGM, Invensys Action Instruments, etc.) about half way, so that the loops then become only about 200-250m distance.

Alternatively, you could try looking at a point to point radio link. Adalet, Omnex, ProSoft, Synetcom,Cirronet, and others make them-- this transmits the 4-20 link via a point-to-point wireless connection. This would be more expensive than the current isolator retransmitter system.

Walt Boyes
Editor in Chief
Control magazine
www.controlglobal.com
blog:Sound OFF!! http://www.controlglobal.com/soundoff
_________________

Putman Media Inc.
555 W. Pierce Rd. Suite 301
Itasca, IL 60143
630-467-1301 x368
wboyes@putman.net

It depends on the quality and type of the field wiring. But this turn up to be a very expensive solution.

Because to the anticipated voltage drop across the cable ends, due to the some-how long distance, you only may transmit CURRENT (0/4-20mA Signal), not VOLTAGE Signal (0-5/10v).

In order to transmit a current loop signal (i.e. 0/4-20mA), you need to consider the voltage drop when sizing the cable. The voltage drop across the cable length should be just enough for the signal to drive the AI/AO accordingly.

A cheaper way than a good quality and an over-engineered field cabling, is for you to use either:
1) 2 PLCs of reasonable price like the Twido PLC from Schneider
or
2) Remote Analogue I/O Blocks, like those of Moore Industries.

Anyway, The PLCs or the Remote I/O Blocks will be used to mimic the signal on the input side of one end and send the same signal to the other the equpment at the other end.

The linking media can either be either "RS-485/RE422 twisted pair shielded cable, or via a Wireless Link.

You need to weigh up all the solutions to come up with the solution most convenient for your application as well as for your budget.

Regards,

Ashraf Zaki

At the outset, before posting this type of question, please ensure that you provide sufficient information for people to comment. What is the PLC you are using, what are the flow meters you are using etc., which gives readers a scope to give you the proper & viable solution. This is my advice not an order.

Let me try giving some information.

Please check if the Analog meters are HART compatible. If they are, please check with your PLC vendor to provide a serial communication card into a spare slot.

He will be able to guide you from there onwards.

Second solution: If your PLC is Siemens, go for a combination of Profibus DP & Profibus PA. I am not sure if is cost effective.

Fellow Professionals, please comment on my answers. I am also in the phase of learning.

Assuming that the output of each instrument is 4-20mA or perhaps 4-20mA with a HART signal superimposed, use shielded twisted pair copper wire. Be sure to ground the shield at an instrument ground, not the electrical earth ground on one end.

500m distance is not unusual, there are 100,000's of loops that run at this distance. It is testimony to the suitability of a DC current signal to do so, worldwide.

Carl

Make sure you use cable with good insulation. Particularly in damp locations, leakage can be a significant and highly variable error term. Otherwise, 500m shouldn't be a problem with a current loop.

Regards
cww

In response to Curt re: underground, I agree a good cable to use for long distance is one of the direct burial telephone cables these are typically filed with petroleum jelly. The smallest size 0.5mm has a resistance of only 91 Ohms/km (I assume 180 for the pair), that's only 4 V/km drop so a typical transmitter should be good for 2 km at 24V 8km at 48V. Cross talk shouldn't be a problem for 4-20 mA. Don't do underground joints though, I have had bad experience in the past with water shorting. Use all the regular communication hardware, punch terminals, etc.

Roy

Dear Gentlemen,

Our design is using ABB DCS and the field instrument will be E+H and Rosemount, we are not using any protocol, only hardwire using single twisted pair cable...

From some reference, I got that the 4-20mA is still good until 1000 feet, but I still afraid of the path loss and drop.

Please advise me, is it still possible to apply the design, or better using comm protocol such as HART or Profibus? Or maybe put repeater? Since the cost is one of big matter for this case...

And another matter (seems I have many problems, right?), we also have some limit switch on offshore facilities, but after few months, the contacts is broke due to corrosive, are there tricks to solve the problem???

Thanks a lot for the advice...

There isn't any path loss except leakage and there isn't any drop since the current flowing out of a wire has to equal the current going in. Voltage drops are handled by the compliance of the driver and any reasonable gauge of wire won't exceed this in 1000 ft. This is why people use current loops.

Regards

cww

Regardless of system or devices the limitation is ohm's law

If the DCS provide 24 VCD and the device requires 12 VDC to operate then the maximum drop along the wire allowed is 24-12=12 VDC. In order for the device to deliver 20 mA output the maximum LOOP resistance is 12/0.02=600 ohm. Typically the DCS has a 250 ohm resistor but some models have different values. If you add other indicators or recorders they also have resistance. Anyway, lets say 250 ohm then this leaves 600-250=350 ohm total for both wires and any poor connections. AWG#18 cable is 44 ohm/km loop. 350/44=8km which is very much longer than you need. Therefore the limit is not wire length. It is resistance due to other devices you put in series such as converters, indicators, recorders, limit switches, isolation amplifiers etc.

The other threats include low supply voltage and resistance due to corroding terminals.

Actually resistance must be lower because transmitters use 21 mA as a rudimentary way of indicating overrange and failure etc.

The problem with current loop is that if the signal is limited due to resistance increase of voltage sag, or distorted due to ground loop, it is almost impossible to detect because one valid signal (maybe 19 mA) just becomes another valid signal (maybe 18 mA) and the system cannot detect this. This is why it is necessary to use digital communication.

Control systems are moving to fieldbus, and one reason for this is to avoid this and other issues. SIS (Safety Instrumented Systems) are now going smart, combining HART with 4-20 mA to detect problems on the analog signal. Even WirelessHART is completely digital just like fieldbus.

So use fieldbus for control, HART for safety, and WirelessHART for monitoring where running wires is impractical.

Once you have a digital plant architecture, you can do so much more.

Cheers,
Jonas