24V distance sensor project (4-20mA)

I have a very interesting little project that came up, and I was contemplating using a simplified Arduino system to accomplish what I need, but am not sure if I can make it work.

Basically, I have a Baumer inductive proximity sensor that is providing a 4-20mA realtime position signal to a control system. The problem is, when the sensor is replaced or maintenance is done, getting the sensor back in the correct "neutral" position is tedious, since the feedback for the system is about 100 yards away in an operator's station. What we would like to do is tap that 4-20 signal, and put a small box close to the sensor location that has just use a few basic LEDs to guide the maintenance crew when aligning the sensor. I am thinking a green LED that lights up when it's in the proper centered range (12mA +/- maybe 1mA), and a red LED on either side of the green to indicate if it's too far to the left or right. The Arduino code should be very easy, once I determine how to properly read a 4-20 analog signal.

But, my issue is the power and signal itself. The Baumer sensor is a 3 wire affair: +24V out, signal wire, common wire. I need to tap the signal to detect position, but not affect the signal upstream at the controller. But I also need to tap the power from that signal, if possible - if not, running separate 24V power to the unit is possible, but not ideal.

Any thoughts on how to come at this? The idea is to build something very basic, rather than forcing a move to expensive LED bar graph types of displays, which is one of our few alternatives at this point.

You have a typical 3-wire device. The +24 volt is the voltage applied to the sensor to power it. The output provides the 4-20mA signal, which drives into typically a 250 ohm resistor at your controller (this results in a 1-5 volt signal). The ground or negative connection is the common return for supply and loop current; his actually carries the difference between the power supply current and the 4-20 signal.

A 4-20 signal is exactly that !. You can insert an extra resistor into the signal line and the transmitter will boost it's output drive voltage to maintain the correct current value. I therefore suggest you use something like a 100 ohm resistor which will have a voltage value of 0.4 to 2.0 across it in correspondence with the 4-20mA signal. However both sides of the resistance will be at some potential above the sensor ground circuit so your display system must be "floated" from both the + and - supply lines.

This really isn't a problem if your remote "calibrator" is self powered by a set of batteries. I'd suggest you use a set of 3 trigger amplifiers, each set to light the LEDs at your desired measurement values.

Alternatively you could ignore all of the above and simply insert a DVM into the signal line set to mA and read the transmitted current - much simpler and more accurate than fancy gizmos.

Why not just have the maintenance crew hook up their multimeter into the current loop and measure the loop current while they align the device, that's how we would do it in the refinery I worked in.

Lefty

Thanks for the tips, @jackrae. I'll play around with some of those ideas to see if I can get something together that looks workable.

As for using a DVM - it's pretty much the only way they can do adjustments now, other than physical gap gauges (which are nearly impossible to use). The problem is that using a DVM is very clumsy, and typically just one maintenance person is doing all of the adjusting to multiple sensors during the very limited downtime they have. And the actual access points to the sensor wiring are not able to be positioned in a convenient location for maintenance personnel. Providing a little tool that is easily visible from up 5-10m away for each sensor would dramatically reduce the amount of fumbling that is required whenever they do adjustments (which can be 1-2 times a day for each sensor).

An idea to take your project one stage further :-

Rather than having only 3 LEDs to show the reference points, you could use an LM3914 dot display chip which has all the circuitry necessary to drive 10 LEDs as a voltage display meter giving a range of 0-1.2 volts Hence each LED would equate to 120mV (1.2mA on a 100 resistor)

This chip has been around for just about as I can remember, but it's a very useful little unit.

jack

Ah, thanks for that suggestion as well, @jackrae! I could definitely use that LM3914 to make something more interesting...