I know that some circuits work with current as the signal rather than voltage. I believe its called current mode circuit. Currents matter in analogue IC design I guess that is why we like to use so many current mirrors. However, here the situation is an IC called "Precision Voltage-to-Current
Converter/Transmitter". So it converts voltage to a current. Since this is an off the shelf component, it will be used as part of a larger circuit on the PCB and not inside an IC. This brings me to this question.
What is the application of such voltage to current converter ICs (this one is marked for "precision" applications)? There are a few things mentioned under Applications on the front page. However, these listed applications are rather vague. Also, in the datasheet it shows use of filter at the output. First it shows use of simple capacitor and then use of RC filter. Now, I am wondering that since the output is current and then we are connecting RC filter which will have voltage drop, isn't that a bad thing?
If you are sending an analog signal or DC value down a long wire, where the wire has good insulation (normal) but non-negligible resistance (also normal), transmitting the signal as a current proportional to the value into a known load will be more accurate than sending a voltage down a wire with unknown resistance.
I used one from Linear Tech and Analog Devices for 4-20mA current loops. Common in industrial controls applications.
An advantage of the 4 to 20 mA format is that it quickly shows if the connection is broken.
There are many precision ICs designed to use that format.
Now, I am wondering that since the output is current and then we are connecting RC filter which will have voltage drop, isn't that a bad thing?
https://en.wikipedia.org/wiki/Current_loopIn a current-loop system the receiver also senses and responds to the current, not the voltage. As long as the impedance of the circuit isn't so high that the sending device cannot put out the desired amount of current, the voltage drop won't matter. The main advantages of a current loop are its relative immunity from interference and the ability of the sending device to (almost) verify that the receiver is getting the correct signal with only two wires. The only thing that might give the sender a false good indication is if there were a short between the two wires. Some systems might even be able to flag that (low loop voltage) as a fault as well, but I'm not sure they do.
The last system I worked on was more than 20 years ago. I had used a brand new part from Analog Devices that had everything integrated into one package. Like most new ICs, they come with a lot of pain and suffering... I think in that case, I found two major bugs in the design. The stories I could tell.
That part could detect a faults, like an open or short for example.
If all else is equal (ha!), the current is the same everywhere in the loop. If you drive the wires with a certain current, that's what's going to appear at the other end. The fly in the soup would be if there was a way to inject more current into the loop, or a way to allow it to leak out. If you can keep it contained, within tolerances, it will behave.
Voltage over a long wire, though, is subject to change, because the wire provides resistance that acts as part of a voltage divider. The voltage at the receiving end will not be the same as the voltage you think you sent.
So... let's turn the voltage into a current, shall we?
One great feature of 4-20mA loops is that you can use the 4mA to power the remote sensor.
In fact the 4-20mA system was deigned with this in mind.
If the current drops below 4mA then you flag it as a short or open fault, and that's really useful.
joeqsmith, I am very interested to know your stories.
I never knew that "4–20 mA" was an actual standard.