Hello all, looking for some answers regarding specific questions involving 4-20mA circuits as they apply to microprocessors such as Arduino and others. I have about an 8th grade level of knowledge in Electronics, although that's debatable...it might be as low as 1st grade level.

1) Converting 4-20mA to voltage (DC). I've got the concept, pass the current through a resistor and you get voltage. I already know that using a 250 ohm resistor gets me 1-5vdc which works great with an Arduino. However, in playing with the formula V=I*R it seems there is no limit to the voltage you can produce and I don't understand that. For example, .020 * 10000 ohms equals 200 volts. I can't wrap my mind around this. If I'm only feeding 12vdc to the sensor, where does the 200v come from?

2) I will be using an 8 channel multiplexer (74HC4051) to support up to eight 4-20mA analog sensors. My original thought was to feed the current from each sensor through it's own dedicated resistor (converting to voltage) prior to the multiplexer (20mA -> Resistor -> Multiplexer -> ADC). However, because the multiplexer can only provide access to one sensor at a time there would be 7 less parts and a simpler design if I could pass each sensor through the multiplexer first, then through a common resistor (20mA -> Multiplexer -> Resistor -> ADC). The multiplexer will be run with VCC of 5vdc but the sensors will have VCC of 12vdc. The 74HC4051 datasheet indicates it will handle a maximum of 20mA (must be more than coincidental) so seems like passing the current through the multiplexer and converting to voltage afterwards will work...but I'm skittish about this because of my voltage conversion confusion as stated in #1 above and not wanting to pass more than 5vdc through the multiplexer. Does this mean that the current is 4-20mA regardless of sensor VCC and so voltage is not relevant to the multiplexer in this context?

3) Using a 4-20mA current output with an RC filter. Someone with extensive electronic experience once told me that using a 10k/.1uf RC "noise" filter on a 0-5 voltage (DC) headed to an ADC on an Arduino type microprocessor was just good practice, cheap, and would provide more stable readings. I didn't question the advice and started implementing it without even testing to see if it was effective or not. In terms of a 4-20mA output sensor, I understand I need to use a 250 ohm resistor to convert to voltage for the ADC. But if I convert to voltage and then pass that through an RC filter, doesn't the resistor in the RC filter change the voltage again? Won't it be the same as having a 10,250 ohm resistor and bump up the voltage to 20?

 

1) No, it doesn't go all the way to 200 volts; with your sensor supplied with 12 volts, that's as high as it will go-- most likely, around 9 to 11 volts or so.

2) Use individual 250 ohm resistors, one for each channel. Do not attempt to pass the 4-20 mA through your multiplexer because with only one channel ON at a time, the others have no path for their 4-20 mA output and this may cause them to malfunction.

3) the 10 kohm resistor in the filter doesn't change the voltage because it is not in the DC path for the 4-20 mA, which goes, as before, through the 250 ohm resistor to circuit ground.

 

1) Converting 4-20mA to voltage (DC). I've got the concept, pass the current through a resistor and you get voltage. I already know that using a 250 ohm resistor gets me 1-5vdc which works great with an Arduino. However, in playing with the formula V=I*R it seems there is no limit to the voltage you can produce and I don't understand that. For example, .020 * 10000 ohms equals 200 volts. I can't wrap my mind around this. If I'm only feeding 12vdc to the sensor, where does the 200v come from?

What drives the loop is a current source obviously. Current sources have a compliance limit,
which means they need a certain minimum voltage across them to function as a current source.
So they have a power supply giving them some min V. The loop, when you keep raising the load
R, slowly subtracts from that V souces until the current source essentially craps out. So unless
the current source has a compliance of 200+ volts (in your example) it will stop regulating and
the V ultimately settles close to that 200V. But most current loops are << voltage, so that is why
you do not get to the much higher V's.

Regards, Dana.

 

Thanks folks, and thanks for being gentle with me. I thought the formula would take supply voltage into account somehow (magically perhaps?). Your explanations make sense and are appreciated.

2) Use individual 250 ohm resistors, one for each channel. Do not attempt to pass the 4-20 mA through your multiplexer because with only one channel ON at a time, the others have no path for their 4-20 mA output and this may cause them to malfunction.

Question about this. What's the difference functionally, to voltage. The multiplexer will still only have one channel "on" at a time if I feed it voltage and the other sensors will have no path for their output voltage.

Another question if you please, just at a very basic level, why is a resistor necessary anyway to measure voltage in a 4-20mA output? I'm missing something basic again. Current can't flow without voltage, so obviously there is some voltage pushing the output current. Why is that not measurable directly from the output wire? What exactly does the resistor do in this case? I know in the general sense that resistors restrict the flow of current.

 

Why is that not measurable directly from the output wire? What exactly does the resistor do in this case?

The output current is driven by a voltage, but the sensor signal is proportional to the current, not the voltage.
The voltage is whatever it takes to generate the current.
If there's no resistor to absorb the current, the circuit will still try to generate the required current, with the voltage then saturating at the maximum the circuit can generate, for any signal level.
There will thus be no significant variation in voltage with signal level.

You need the resistor to convert the current back to a voltage proportional to the signal.

 

Question about this. What's the difference functionally, to voltage. The multiplexer will still only have one channel "on" at a time if I feed it voltage and the other sensors will have no path for their output voltage.

Yes. And as I said before, for some sensors, this would cause them to malfunction.

Another question if you please, just at a very basic level, why is a resistor necessary anyway to measure voltage in a 4-20mA output? I'm missing something basic again. Current can't flow without voltage, so obviously there is some voltage pushing the output current. Why is that not measurable directly from the output wire? What exactly does the resistor do in this case? I know in the general sense that resistors restrict the flow of current.

4-20 mA sensors DO NOT produce a voltage output proportional to the measurand; they produce a CURRENT proportional to the measurand, and that current is independent of the resistance the output is feeding. This has the benefit of preventing wiring resistance (such as with long wires between sensor and the measuring device) from affecting the measurement.

 

Perfect. Finally makes sense. Thanks to all.