Author Topic: DAC stability  (Read 13293 times)

congo

  • Newbie
  • Posts: 16
  • I'm a llama!
    • View Profile
DAC stability
« on: March 26, 2018, 01:48:38 PM »
When a DAC output is set to a value, and not moving when viewed by online monitor, why would the output voltage level move around? The output is connected to a 0-10V motorised valve actuator input and a signal fluctuation of about +/- 0.02V is seen constantly. The 0.2% movement is enough to hear the valve actuator hunting constantly. Is there a way to reduce this effect?

garysdickinson

  • Hero Member
  • Posts: 502
  • Old PLC Coder
    • View Profile
Re:DAC stability
« Reply #1 on: March 26, 2018, 09:26:36 PM »
I suspect that the problem that you are seeing is the result of noise being coupled into the DAC signal.  

The amount of noise, 20 mv is pretty small, so you have a very good shot at taming the issue.

In my experience, the #1 problem with analog signals and PLCs has to do with grounding.  If you are using the PLC digital outputs to switch high current loads, you must be very careful to ensure that you use the PLC's Analog Gnd output as the ground signal for the external device.

You should use a single point ground for your PLC/system wiring.
 
If you are not using the PLC's Analog Gnd, then rewire your cabling to use this as the gnd reference for the external device.

If the valve is dithering or hunting, this will result in big currents in the wiring that supplies the power for the valve.  If you 0..10V signals are bundled in with the valve power wiring this is the most likely source of the noise. If you have done this, then I'd suggest that you:
1. Twist the power/ground wires (running to the analog valve) together to minimize their ability to radiate EMF.
2. Shield the 0..10V signals using shielded twisted pair wiring. Connect the shield at one end.  I'd suggest the PLC Analog gnd.
3. Keep the 0..10V signals away from all other wiring.

If you have a good quality DVM that can do true RMS AC measurements and can be set to show only the AC component then I'd start with a couple of simple tests:
  • Disconnect the cable to the analog controlled valve.
  • Connect the DVM between the PLC Analog ground and the DAC output that you are using.
  • Program the DAC output for 0 volts.  Measure the DC voltage and the AC voltage.
  • Program the DAC output for 5 volts.  Measure the DC voltage and the AC voltage.
  • Program the DAC output for 10 volts.  Measure the DC voltage and the AC voltage.
If you are seeing the 20 mv AC noise with no cable connected to the PLC, then the noise problem is with the PLC.  You will need  to add passive analog filtering to reduce the noise.  

If the noise is not present with the cable disconnected, then the problem is either with the cabling, routing of the cabling or your external device.
  • Connect the cable to the analog controlled valve.
  • Verify that the voltage measured between the PLC Analog Gnd and the signal ground input at the analog valve controller is 0.000 VDC.
  • Connect the DVM at the input to the analog valve controller.
  • Program the DAC output for 0 volts.  Measure the DC voltage and the AC voltage.
  • Program the DAC output for 5 volts.  Measure the DC voltage and the AC voltage.
  • Program the DAC output for 10 volts.  Measure the DC voltage and the AC voltage.
If you are getting non-zero DC measurement on the analog gnd wire then you  need to fix this first.

Now that you have fixed the DC ground issues, repeat all of the tests and see if this fixes the voltage at the automated valve.  It it does, then you are golden.

You may have to change the routing of the analog signals to ensure that they are not running close to any wiring carrying high current noise, AC wiring, AC motor wiring, arc welders, VFD drive wiring, big solenoids, ignition systems...

You may have to shield the signal wiring to minimize any noise pickup.

Once you have done all of this stuff and still have a noise issue when connected to the automated valve, then you will need to work on passive noise suppression right at the analog input to the valve.

Best regards,

Gary D*ckinson

« Last Edit: March 26, 2018, 09:44:11 PM by garysdickinson »

congo

  • Newbie
  • Posts: 16
  • I'm a llama!
    • View Profile
Re:DAC stability
« Reply #2 on: March 30, 2018, 12:50:51 PM »
Thanks for the detailed response Gary, I'm working through your ideas.