Real-Time Measurement and Control

cancel
Showing results for 
Search instead for 
Did you mean: 

Incorrect reading with Ni 9205 in differential mode. Role of COM not clear.

Hi,

I am trying to use 9205 AI module in differential mode. I am not getting the correct measurement. I notice in the discussions that COM pin has a role to play.

The wiring diagram in th emanual is not clear to me.

e.g I connect one signal V1 to channel 1 (AI0 and AI8) ie pins 1 and 20 on the module
and another signal V2 to channel 16 (AI16 and AI24) ie pins 11 and 30.

What is worng with this connection? Do I have to use COM? If yes how?

How do I get correct value in FPGA? Is there some manipulation of channel readings to be done to get the correct 16 bit number?

Thank you.

Anmol.


Message 1 of 14
(14,427 Views)

Hi Anmol,

The cRIO FPGA gets the uncalibrated data from the 9205. You would need to pass the uncalibrated data along with the calibration values to the cRIO controller to convert the data into calibrated values. There are examples in the example finder on how to do this.

With the release of cRIO 2.4 driver, you would be able to get calibrated data in the FPGA as a fixed point value from the 9205. If you do not wish to go through the trouble of converting the uncalibrated data, then I would suggest you to try upgrading to cRIO 2.4 driver. 

Regards,
Stanley Hu
National Instruments
Applications Engineering

 
0 Kudos
Message 2 of 14
(14,245 Views)

Hi,

 

I read the messages on the forum, then the document " Field Wiring and Noise Considerations for Analog Signals" (https://www.ni.com/en/shop/data-acquisition/measurement-fundamentals/field-wiring-and-noise-consider...) and also the manual of the NI 9205 but I'm still confused. I don't know what to do with the COM pin of the 9205 when using the differential measuring method. Can somebody help me by telling me exactly if the COM must be "not connected" or "connected". And if connected, to what ? the "chassis ground" or "earth ground". This question from Anmol was not clearly answered.

 

Thank you.

 

 


Anmol a écrit:
Hi,

I am trying to use 9205 AI module in differential mode. I am not getting the correct measurement. I notice in the discussions that COM pin has a role to play.

The wiring diagram in th emanual is not clear to me.

e.g I connect one signal V1 to channel 1 (AI0 and AI8) ie pins 1 and 20 on the module
and another signal V2 to channel 16 (AI16 and AI24) ie pins 11 and 30.

What is worng with this connection? Do I have to use COM? If yes how?

How do I get correct value in FPGA? Is there some manipulation of channel readings to be done to get the correct 16 bit number?

Thank you.

Anmol.



 

 

0 Kudos
Message 3 of 14
(12,803 Views)

when measuring differentially, both legs of the signal must be within the common mode voltage specs of the device.  if they are, the COM pin can be ignored.  if your signal is truly floating like a thermocouple in air, there is no provision that will insure the common mode voltage.  in that case you may connect a pair of resistors (100k for example), one on each leg to COM to bleed the common voltage so that the signal remains in range.

 

In the original post, the complaint was incorrect readings.  No description of the problem.  Kind of incorrect, totally incorrect, sometimes incorrect. this mayhave been due to differential method or something else entirely.

Stu
0 Kudos
Message 4 of 14
(12,769 Views)

Hi Stu,

 

Thank you very much for your answers. I understand the point to connect the legs of the signal to the com pin through bias resistors when it is necessary. I was confused about the connection of the pin com itself to the ground. If you look at this document for example (https://www.ni.com/en/shop/data-acquisition/sensor-fundamentals/measuring-direct-current-dc-voltage....), in the figure 4, the com pin is connected to nothing but in figure 5, it is connected to the earth ground. So I just want to know if it must be connected to the earth ground, or to the chassis ground or to nothing. Thank you in advance.

 

PS: I'm using the NI 9205 with the cRIO 9014. I guess the connections are the same as for the compactDAQ.

0 Kudos
Message 5 of 14
(12,735 Views)

figure 4 is a simplified version of figure 5.  if the unit under test includes a reference to chassis/earth ground such that it will not float, no extra connections are required resulting in the schematic represented by figure 4.  This is the most common connection for typical signals.  if your signal is coming from a battery driven device, isolated output device or thermocouple, a reference must be supplied similar to my description.  I prefer to put resistors on both legs to keep the impedance matched.

 

The reason to connect via differential is two-fold.  to eliminate ground loops and to eliminate common mode noise. 

Stu
0 Kudos
Message 6 of 14
(12,729 Views)

Referring to figure 5 DIFF connection(ground-Referenced). If the source has a different reference than the module's common what keeps the source within the common mode range of the module ?
Does the AI GND need to be connected to the source reference to keep it within common mode range ?

 

0 Kudos
Message 7 of 14
(11,655 Views)

Hello,

 

The figure below was taken from the 9205 User Manual: 

 

Diff Pt 1.jpg

 

 Could you please clarify, using illustrations, your question regarding 'source reference,' the 'module's common,' and 'AI GND?' On the 9205, the ADC is isolated form the system chassis ground. In a differential configuration, to prevent the ADC from floating altogether the CH- is also connected to the COM terminal, as reference for the ADC.

 

Hope this helps. Please post back any further questions.

 

Thank You, 

Patrick Corcoran
Application Engineering Specialist | Control
National Instruments

0 Kudos
Message 8 of 14
(11,622 Views)

If you look at this document  https://www.ni.com/en/shop/data-acquisition/measurement-fundamentals/field-wiring-and-noise-consider...  Table 1..

 


Testpic.PNG

 

The source common and the AIGND are isolated..

Also look at same document fig.17 

 

GroundLoop.gif

They are implying not to connect the source ground to the AIGND or you will get ground loop currents.

 

NI9205_pic.PNG

 

If source ground is connected to NI9205 com(AIGND) would that cause ground loops as illustrated in Fig 17 above.

Also the illustration from table 1 shows the source ground is isolated from AIGND.

Should source ground be tied to COM as illustrated above or isolated ?

0 Kudos
Message 9 of 14
(11,548 Views)

I believe we are pulling from different areas of the article, that are discussing different topics. The Earth ground reference provided to a differential configuration is entirely dependent on the source being measured, see Table 1 below:

 

Differential.jpg

 

A grounded signal source provides a earth ground reference with respect to AI-, internal to the instrument providing the signal. For pure floating sources, you can see that bias resistors must be employed from the AI input configuration to earth ground to reduce the effects of bias currents on the measurement system. What type of source are you working to measure with the 9205? 

 

Thanks,  

Patrick Corcoran
Application Engineering Specialist | Control
National Instruments

0 Kudos
Message 10 of 14
(11,540 Views)