long distance transfer of analogue signal 1

[briwire]

I have a transducer in a steam line near a corrugating machine that shows the steam pressure 0-16Bar on an operating desk.

It says on the transducer 0-20mA, so I assume that 0mA =0bar and 20mA=16bar.

I have been asked to get this signal sone 300metres to the boiler house for them to monitor.

I'm an electrician so instrumentation isn't exactly my trade, but if I get a signal conditioner and connect its input in series with the original signal. Then use say 24v via a resistor to drive a current through the output and a two core cable running to the boiler house. Does that sound about right?

Is 300m a problem? What cable to I use twisted pair screened? What CSA? How do I calculate the resistor size? Anyone recommend a suitable signal conditioner.

 

[danw2]

It would help is we knew the brand/model of what you have. 4 wire? active or passive output? There's no way to know.

The existing transducer already has power because it "shows steam pressure". Presumably on a built-in display?

It has 0-20mA output. Use that milliamp signal to drive a digital display at the far end. A milliamp signal will go long distances, 300m is no problem. Use any available twisted pair wire, say 20 gauge or thereabouts for your 300m run. I don't think you need a 'signal conditioner'. Your transducer has a current output, you can get a digital display that'll take a 0-20mA signal, there's nothing to condition or change.

Problem is, it isn't clear whether your brand/model pressure transmitter's output is powered internally by the power that drives the display you already have, or whether the output needs to be powered separately from an external power supply (loop powered).

If the current output needs an additional external power supply, then use a 24Vdc power supply.

You don't need a loop resistor by itself. Whatever display you use at the far end to read the pressure needs a resistor, but the display units usually build it into the unit, or provide it.

 

[briwire]

The transducer provides a 0-20mA signal to a PLC input. This is then processed in the PLC program to display it on a monitor along with lots of other data. I intend breaking into this near the PLC so that it is in series with the signal conditioner input.

I don't want to affect anything on this display and I don't really want to take a signal derived from the corrugator 24v control circuit outside of the area of the machine, hence the use of a signal conditioner. I intend using a seperate dedicated 24v for the 300m current loop.

The people in the boiler house are playing silly buggers. They want a signal there but as of yet they don't know what to do with it. So I just want to give them their 0-20mA corresponding to the steam pressure on the machine and leave them to it. It doesn't really concern me. I think they want to modulate the pressure at the boiler using this remote sense. The boiler house belongs to them so I'll leave them to it.

 

[danw2]

Ahhh, the rest of the story . . .

Yes, you should have a signal isolator, otherwise you'll more than likely have a ground loop between the source, the PLC and boiler room; and everyone will be mad at you.

But be sure you're buying an isolator, not just a signal conditioner. Many signal conditioners are NOT isolated input from ouput. Lots of choice from Phoenix Contact (many other brands as well)

http://www.phoenixcontact.com/signal-level-matching/31244_14827.htm#30529

There are also signal splitters that provide two isolated outputs that follow a single input, like Phoenix Contact's MCR-FL-C-UI-2UI-DCI

http://rd.phoenixcon.com/products/interface/db/datasheets/MCR-2UI.pdf

 

[briwire]

I have seen one in RS made by ABB (ISVR011714R1100) RS No 442-8907.

It says it is a 3 way analogue convertor. I believe that this means the output, supply and input are all isolated?

One last thing, does the supply for the output current loop come from the 24v device supply (sentance above says no) or do I have to provide a PSU for the output.

 

[danw2]

>I believe that this means the output, supply and input are all isolated?

The spec sheet says 3 way isolation:

I presume the heavy dark bars between the power supply, input and output represent the isolation on the internal schematic below.

>does the supply for the output current loop come from the 24v device supply?

Since the schematic for this model, a V to I converter, shows the I (current) output driving a 500 ohm load without an external power supply, I would concluded that the 24Vdc power needed to run the isolator also drives the current output, so, you need 24Vdc to run the isolator, but not to loop power the output.

 

[Thealanator]

I have done this type of thing before. I used an Action Instruments DC input transmitter (model 703 if I remember correctly). Its input resistance is fairly low compared to a typical PLC input (250 ohms). The output is then hooked up like a typical current transmitter (power supply, transmitter, display/input).

You had mentioned using a separate power supply for the output, which I think is a good idea.

Current loops work well over long distances, but you need to be careful about how much load you put into a loop. A typical transmitter requires 12 volts. A 250 ohm display/input drops 5 volts at 20 mA. So a 24 volt supply will allow for two typical inputs and some line resistance.

You will need to check the transmitter specs, power supply voltage and verify what kind of load is on the existing loop.

Hope this helps.

 

[briwire]

I ended up buying a slightly more expensive ABB type that can be selectred for any combination of input and outputs.

I set it up on a bench with 0-20mA input and adjusted the offset and gain to give 4-20mA output.

I'll connect it into the machine circuit next week. I'll get contractors to run the cable. It's a long way. I'm going for twisted pair foil screened. Still think I'll use a seperate PSU for the 24 power supply for the unit, just so the machine voltage stays around the machine (even though it is isolated).