4-20mA Signal Repeater (Splitter)

[aljamri]

Hi,
In maintenance area of work, similar problem may offer you a good opportunity to think of a solution of your own.
4-20ma signal is one of the common languages among instrumentation devices, and making one Tx. To talk to more than one Rx requires special arrangements as it is a current loop. In our case we are using Signal Repeater or splitter which in simple: One signal in and two or more signals out, all of 4-20 mA.
I’m thinking of using my humble knoledge about PIC, with a lot of ETO Pros., to do this project, so I can decide, myself, how many outputs.
For simplicity I’ll make all of them analog, avoiding the need of converting the values into digital in case of LCD output type.
So, my questions are:
Is this project is possible using PIC ?
Which PIC best to choose for this project?
Thanks

 

[Mr RB]

Isn't that normally done quite simply using an opamp?

 

[dr pepper]

You can daisy chain so many devices, depending on the max o/p voltage of the transmitter.

But if you want to use a pic look at the analog devices AD420, its a serial in 0-20, 4-20ma o/p converter.

 

[aljamri]

Isn't that normally done quite simply using an opamp?

Thanks Mr RB,
Frnkly speaking, I've no idea, I know converting voltage into mA can be done by op-amp, but getting parallel ma signals?

 

[aljamri]

You can daisy chain so many devices, depending on the max o/p voltage of the transmitter.

But if you want to use a pic look at the analog devices AD420, its a serial in 0-20, 4-20ma o/p converter.

Thanks dr pepper,
Yes, you are right, but what I've forget to tell you is that I want to use it as isolator between two differnt circuits, where interupt in the reciving circuit will not affect the Tx circuit as if they are connected in one loop.

 

[ericgibbs]

hi aljamri.

Look at Figure #15 on the attached datasheet for the HCNR200 analog optocoupler.

E.

 

[dr pepper]

Cool circuit Eric, one for my archive.

You can buy gizmo's to do this, that fit on a din rail and can be run from any power available, but they'd cost a whole lot more than building one.

 

[Mr RB]

Thanks Mr RB,
Frnkly speaking, I've no idea, I know converting voltage into mA can be done by op-amp, but getting parallel ma signals?

Sure, the incoming 4-20mA makes a voltage across a sense resistor (referenced to ground), the opamp measures and matches that voltage on another sense resistor by turning on a transistor. That gives you a parallel 4-20mA current loop that matches the first one. Total parts; an opamp, a transistor and a few resistors.

 

[alec_t]

MrRB's method would be my choice. The PIC approach seems more complicated and more expensive.

 

[aljamri]

Sorry for late reply, I was busy with the kids exams.

Look at Figure #15 on the attached datasheet for the HCNR200 analog optocoupler.

E.

Thanks Mr. Eric, I've checked my supplier and unfortunately, they are not having this opto-coupler.

Sure, the incoming 4-20mA makes a voltage across a sense resistor (referenced to ground), the opamp measures and matches that voltage on another sense resistor by turning on a transistor. That gives you a parallel 4-20mA current loop that matches the first one. Total parts; an opamp, a transistor and a few resistors.

Aha, I'll try to do this, but what concern me is the required tolerance of the resistors, to achieve acceptable accuracy. I'll try and let you know. Thank you

 

[dr pepper]

1% ought to be fine

 

[Mr RB]

Yeah, and you can match two resistors with a multimeter too if you have a heap of them and a decent ohm-meter.

Or you could add a small trim pot to give a few precent of adjustment on one resistor, then tweak it until you are happy with the accuracy.