Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

Signal Noise - How to Optically Isolate DIO and AIO

Hello,

 

I'm having a bit of trouble with noise on the analog inputs of my 6259 DAQ card.

I've figured out that the noise is coming from the control lines of a servo motor controller that are connected to my DAQ card.

 

Manual for servo motor and controller I'm using are here:

 http://www.orientalmotor.com/products/pdfs/opmanuals/HM-5073-2E.pdf

 

My system is setup in the following way:

 

- DI/DO of the 6259 are connected to the servo controller. The digital lines read or write 3.5V logic to control CW/CWW rotation, start stop motor etc.

- An AO of the DAQ is connected to the servo to control rotation speed

 

The method I'd like to implement would be to optically isolate the control Digital and Analog control lines. I've found some optocouplers but I'm not sure how to do the electical setup. I'm even more lost with how to optically isolate the analog signal.

 

Does anyone know of commercial boards available that can do this?

 

Any help would be appreciated

 

best

SS

 

 

 

0 Kudos
Message 1 of 7
(3,837 Views)

SS-

 

Typically try to eliminate the sources of noise first, before isolation. Have you followed the suggested wiring practices stated on pages 29-30 of the manual you posted?

 

-AK2DM

~~~~~~~~~~~~~~~~~~~~~~~~~~
"It’s the questions that drive us.”
~~~~~~~~~~~~~~~~~~~~~~~~~~
0 Kudos
Message 2 of 7
(3,834 Views)

Yup, checked the list twice.

 

The analog channel that I am monitoring is from a separate instrument but goes to the same DAQ card. It is this channel that the noise is a problem.

 

I'm pretty sure that the noise (in the AI channel mentioned above) is coming from the control lines of the servo which I need to connect to the DAQ for control purposes.

 

When I remove the servo control lines from the DAQ (while the servo controller is still powered up) the noise is removed. The change in the instrument's noise level is subtantial - goes from 5nm to 40nm (peak to peak) once you plug the control lines back in.

 

best

0 Kudos
Message 3 of 7
(3,827 Views)

SS:

 

Glad to hear you followed the manual 🙂

 

For analog isolation, it typically use units from Dataforth (one of many manufacturers). Do you know what range of bandwidth you need for the signal? Here is an isolator with a 3KHz bandwidth and +/10V in to +/-10V out:

 

http://www.dataforth.com/device.view.aspx?deviceid=164

 

 

 

Opto isolators are pretty easy to use, what make/models do you have on hand? They typically invert the logic level, you can use a transitor or inverter gate so it does not invert the logic.

 

-AK2DM

 

 

~~~~~~~~~~~~~~~~~~~~~~~~~~
"It’s the questions that drive us.”
~~~~~~~~~~~~~~~~~~~~~~~~~~
Message 4 of 7
(3,802 Views)

AK Thanks for the link, I beleive that was exactly what I was looking for.

 

I also came across the following digital isolators (looking at the ADUM6404) from Analog devices so if you have any thoughts that would be great.

 

http://www.analog.com/en/interface/digital-isolators/products/index.html

http://www.analog.com/static/imported-files/tech_articles/57062351iCoupler_vs._Optocoupler_Details.p...

 

 

All the different products are beginning to confuse me though. Not sure if these isolators will work as well as optocouplers. What do you think?

 

 

I just need one for digital signals and then I should be good to go. Presently I'm thinking of this one:

http://www.fairchildsemi.com/ds/6N/6N137.pdf

 

Thing I don't like about the 6N137 is that I would need to attach a capacitor between pins 5 & 8. I'm not an electrician so I'm partial to the idiot proof stand alone modules...like the analog one you linked above.

 

Will be cool to see how noise free I can get the signal to be.

 

thanks for the help...

 

best

SS

0 Kudos
Message 5 of 7
(3,780 Views)

Actually, I wonder if this one could work for both the digital and analog channels.

 

http://www.avagotech.com/pages/en/optocouplers_plastic/plastic_wideband_analog_video_optocoupler/hcp...

0 Kudos
Message 6 of 7
(3,778 Views)

Looks like the Analog Devices ADum only come in a surface mount package, that may be a little hard to wire up to if you are not experienced or have the proper equipment.

 

The 6137 looks like a good choice, a cap is standard practive when using high speed components.

 

I did not delve too much into the Avago analog isolator, I do know it can be tricky getting good linearity when using an optocopler in analog mode.

 

Hope this helps.

 

-AK2DM

~~~~~~~~~~~~~~~~~~~~~~~~~~
"It’s the questions that drive us.”
~~~~~~~~~~~~~~~~~~~~~~~~~~
0 Kudos
Message 7 of 7
(3,753 Views)