Hello, I want to measure different circuits and they have different grounds that I can not connect. What's the best way I can measure this? Does anyone know of any kind of isolator so that I can measure this with one arduino (uno or due)?
Did anyone face a similar problem? advise is really appreciated.
Thanks
You can use DPDT relays to electrically the Arduino from the different circuits. Will be similar effect as if you used a multimeter and measured in different places with 2 probes.
If it's voltage you wish to measure then you can't! Voltage is the measure of the difference in electrical potential between to points in a circuit.
Mark
What are you wishing to measure?
How much voltage can there be between the different grounds?
You could use the flying capacitor technique. Connect a capacitor to the common connections of a two pole change over relay.
The normally closed pair of contacts go to your external system. The normally open go to the Arduino input and ground.
When you have made the connection between the + and the ground of your external system you energise the relay and connect the capacitor to your Arduino. Where you can read the voltage.
If the voltage is going to be over 5V then you need to take measures to limit it. A potential divider will not work here as you will discharge the capacitor with it. You need an op amp with a less than unity gain.
I have drawn out the circuit if you are interested:-
Flying capacitor.pdf (24.2 KB)
You need an op amp with a less than unity gain.
Since a Non-inverting amplifier has a gain of (1+RF/RIn) ,it will always be greater than one. The only way I know to get less than unity gain is cascaded Inverting amplifiers.
Why do you think that you need a non inverting amplifier or to cascade inverting amplifiers? That is what software is for.
I was responding to your comment here:
If the voltage is going to be over 5V then you need to take measures to limit it. A potential divider will not work here as you will discharge the capacitor with it. You need an op amp with a less than unity gain.
Yes but a simple single stage inverting amplifier with less than unity gain will do because the inversion can be taken out with software.
ok. I see your point.
If it is required to measure voltage, it may be possible to simply use an amplifier with differential inputs.
If it is required to measure the length of a pulse, it may be possible to use an opto-isolator.
We need more details of the requirement.
I will make an schematic this weekend to show you what I want to do, but to sum up I have some different circuits together that I want to measure and If I put all grounds to the gnd of the arduino they make a short.
So I want to turn the arduino in something like a scope with different channels that dont share the ground (if possible).
I was thinking about using stuff like ISO124P http://www.ti.com/lit/ds/symlink/iso124.pdf but they are "expensive" and introduce an offset big considering I am measuring very low voltage.
I wonder if it is possible to have for example two arduino UNO's (or analog-digital converters) measuring different points and sending the measurements to an arduino due for example so that they dont share the ground and thus I dont make a short circuit.
Thanks
Espada:
I will make an schematic this weekend to show you what I want to do, but to sum up I have some different circuits together that I want to measure and If I put all grounds to the gnd of the arduino they make a short.
So I want to turn the arduino in something like a scope with different channels that dont share the ground (if possible).I was thinking about using stuff like ISO124P http://www.ti.com/lit/ds/symlink/iso124.pdf but they are "expensive" and introduce an offset big considering I am measuring very low voltage.
I was thinking of connecting an operational amplifier as a differential amplifier but as you are measuring very small voltages that may not be a good solution.
Espada:
I wonder if it is possible to have for example two arduino UNO's (or analog-digital converters) measuring different points and sending the measurements to an arduino due for example so that they dont share the ground and thus I dont make a short circuit.
An option to consider is to have a second Arduino floating by using battery power; connection to the 'master' Arduino to be via opto-isolators. The battery would also power an amplifier to give better measurement of the small voltages.
If you want to measure a voltage in a circuit whose GND is different from the Arduino's GND you can (I believe) use the differential input capability in many Atmel chips - but NOT in the 328. It can be done with a Mega or a Leonardo or an Attiny.
You will need to read the relevant Atmel datasheet as the Arduino system does not provide any functions for it. Unfortunately the system is a bit different in each chip.
This YouTube video may be of interest
...R
PS - this is something on my long to-do list that has not yet been tried.
Robin2:
If you want to measure a voltage in a circuit whose GND is different from the Arduino's GND you can (I believe) use the differential input capability in many Atmel chips - but NOT in the 328. It can be done with a Mega or a Leonardo or an Attiny
Interesting, but the differential capability is limited by the fact that both inputs must be between ground and the supply voltage to the chip (Vcc).
Anyway, as you can always subtract the measurement of two analogue inputs, there does not seem to be much benefit from using differential mode unless either timing is critical or you use the 20x amplifier.
Use of a differential amplifier to measure the difference between two floating voltages with respect to the amplifier's ground is not going to work.
Archibald:
the differential capability is limited by the fact that both inputs must be between ground and the supply voltage to the chip (Vcc).
My understanding (and I could be wrong) is that the differential cannot be below 0v or higher than 5v - but the actual voltage could be anything - as long as it is isolated from Arduino GND, I assume.
While not directly related, IIRC some of the Atmel application notes (or whatever they call them) actually link high voltages to Atmega pins through large-value resistors - presumably to limit current flow.
Grumpy_Mike:
Use of a differential amplifier to measure the difference between two floating voltages with respect to the amplifier's ground is not going to work.
I'm not sure if this refers to my comment. I was thinking only of measuring the vltage between the two differential pins - not the voltage relative to GND.
...R
Grumpy_Mike:
Use of a differential amplifier to measure the difference between two floating voltages with respect to the amplifier's ground is not going to work.
I'm not sure what you mean but please explain why it is not going to work.
Are you saying that the circuit here at the top of page 12 is not going to work?
If V2 is the floating voltage and V1 is the "floating" voltage GND, then V2-V1 must be</= 5V or a voltage divider must be employed. In addition, where number of voltages to be measured > 1, a separate DIFF AMP for each would be required.
Given that, there is no reason why it wouldn't work.