Connecting a second device to computer affects sensor

[Ubiquitous]

Hi All,

I'm using an Arduino to read an analog voltage from a sensor (Aurora scientific load cell). I can get a reading of the load cell that seems fine. I'm using my computer to read the result through USB.
However, when I connect an additional device (Newport motorized stage) to my computer through a USB port, the measurement goes immediately to almost zero and doesn't change anymore.
All devices are connected to the same power supply, so same ground.

I am absolutely confused with what might cause this.
Any help or suggestion will be very useful.


Some other details - The connection to the second device (the motorized stage) is through a USB to RS232 connection.
The voltage reading on the Arduino is through a connection to a Coax cable (BNC connector that I cut) and through a voltage divider. The same happens without the voltage divider as well.
The load cell is connected through an amplifier and I am reading the output of the amplifier.
The same happens even if the stage is not powered. Just the connection to the power outlet, even when it is turned off causes the same (hence my suspicion of the grounds).

Thank you very much in advance,
U

[jremington]

There is not enough information in your post to determine what the real problem is. Please read and follow the directions in the "How to use this forum" post.

The same happens even if the stage is not powered. Just the connection to the power outlet, even when it is turned off causes the same (hence my suspicion of the grounds).

This sounds like an AC ground loop problem, which could involve your computer. Plug everything into the same AC socket.

[Ubiquitous]

Thank you for the suggestion.
All are already connected to the same socket.

I have solved the problem as far as measurement goes and would like to update. I would still appreciate any insight, as I don't understand the effect that I'm seeing.
The load cell amplifier outputs both negative and positive voltages (something like -7 to 7V). When I connected to the Arduino, I flipped the signal and ground, so measuring the ground in an analog input and connecting the signal to the ground. It worked fine before connecting the additional stage device.
I don't get the same effect when I connect ground from the load cell to the Arduino ground and the signal to the analog input.
As far as my practical problem goes, I simply connected the signal through a voltage divider thus mapping the [-7,7] to [0,5] (See attached diagram). So other than losing some resolution, I can perform my measurement.

I am still very curious as to why connecting the additional ground to the amplifier caused the output voltage to be close to zero. I'll try to research it too, but suggestions or explanations are more than welcome.





Regarding additional information - I'm using the Aurora scientific 403A device

Newport ESP301 stage with 1 motor connected to it.
Arduino Uno R3
Attached are the rough values I used for the voltage divider.

Thank you all again,
U

[wvmarle]

As far as my practical problem goes, I simply connected the signal through a voltage divider thus mapping the [-7,7] to [0,5] (See attached diagram). So other than losing some resolution, I can perform my measurement.

With good resistors you don't lose any resolution.

You can improve your measurements by bringing the voltage down to 0-1V, and using the internal analog reference. More stable than using Vcc as reference.

[Wawa]

You can improve your measurements by bringing the voltage down to 0-1V, and using the internal analog reference. More stable than using Vcc as reference.

Maybe not in this case.
OP is using pull up (3-resistor voltage divider) to also measure negative voltages,
and that pull up is still connected to the 'dirty' 5volt rail.

Could use the much cleaner/stable 3.3volt supply as reference (assuming Uno).

Connect the Aref pin to the 3.3volt pin.
Make sure you set Aref to EXTERNAL in setup(), otherwise you burn out Aref.
Calculate the voltage divider for 3.3volt, and connect the top resistor to the 3.3volt pin.
Leo..

[MarkT]

The load cell amplifier outputs both negative and positive voltages (something like -7 to 7V). When I connected to the Arduino, I flipped the signal and ground, so measuring the ground in an analog input and connecting the signal to the ground. It worked fine before connecting the additional stage device.

Firstly 7V is outside the voltage range for the arduino, and -7V is right out.

You needed to properly level-convert the analog signal, not play around with reversing the polarity.  Its
possible you've caused damage to the Arduino pin.

The voltage range +7 to -7V can be mapped onto the 0..5V range with three resistor divider/level
shift circuit (assuming the amplifer is low impedance output).

Common the grounds - that's a given.

If you connect an analog pin via a 10k to +5V, via a 14k to the amplifier output, and
via a 35k to GND,  then it will map +7V input to +5V on the analog pin, -7V input to 0V on the analog
pin.

[Wawa]

@MarkT
The .PNG attachment in post#2 shows such a 3-resistor divider.
Leo..