Signal Conditioning

cancel
Showing results for 
Search instead for 
Did you mean: 

saturated thermocouple readings with NI 9213

Hi all,

 

I have a system with a cRIO-9073 with a two NI-9213s in the first two module bays.  The first 9213 has 16 type K T/Cs connected and the second one has 10 type K T/Cs attached.  The overall system provides monitoring and control for a set of six heaters which maintain a gas stream at a specified temperature.  Occasionally, sometimes very frequently and sometimes not at all frequently, one or two of the temperatures will go full-scale positive or full-scale negative.  When this happens, I've pulled the T/C input wires for the "bad" channel from the 9213, plugged them into a handheld T/C reader, and get a proper reading.  Thus, it is not a lack of signal to the 9213.  I've reviewed the 9213 manual in detail, the message boards here, and the "Field Wiring and Noise Considerations for Analog Circuits" document at http://www.ni.com/white-paper/3344/en/ and it looks like I have an issue with the common-mode voltage of the signal getting out of range of that for the channel and I likely need to add bias resistors.  Page 5 of the referenced document states "A single bias resistor is sufficient for low-impedance DC-coupled sources such as thermocouples", and the resistor should go on the "signal low (-) input", and "Typically, values between 10Kohm and 100kohm work well low-imdenance sources such as thermocouples".

 

I am OK with everything so far, but have a few of questions on how to actually implement this on the 9213.  Also... I'm a mechanical engineer not an electrical... so be gentle, haha.

 

1.  If I connect a resistor to the low (-) input for one channel on the 9213, do I connect the other end to an earth ground in the control cabinet or to the COM terminal on the 9213? 

2.  If I connect the resistor to COM, do I then need to connect COM to a ground inside the cabinet?

 

Thanks for the help and input.  Best regards,

Randy Smith

 

0 Kudos
Message 1 of 3
(4,206 Views)

The white paper you referenced is for devices that are ground-referenced and can't handle large common-mode voltages. The 9213 is a bank isolated device (as you probably saw in the manual), so the device can handle common-mode voltages up to very high voltage levels with no problem and works very well for applications where the thermocouples are floating (not connected to any voltage source). But, since it is bank isolated, and not channel-to-channel isolation, what it can't do is handle a relative common-mode voltage between two channels on the same device (specifically +/- 1.2V difference is all that's supported). Since your issue is specific channels that are railing, then that implies to me the problem might be with the relative voltage of your channels.

 

What I would do is analyze the things you have your thermocouples connected to. Unfortunately, if the problem is relative common-mode voltage between thermocouples, then there isn't much you can do with this device except change the way you've connected the thermocouples to reduce the common mode voltage. Options include: 1) have better grounding of your heaters to ensure they are all at the same voltage level, or 2) electrically isolate your thermocouples from the heaters. Since you have two 9213s, they are isolated from each other, and you could, in theory, group your thermocouples that have similar common mode voltages (i.e. different heaters might be operating at different voltages from each other).

0 Kudos
Message 2 of 3
(4,184 Views)

Logan, thank you for the input and clarifications.  I really appreciate your help.  Per your recommendation, I will focus on potential grounding issues with the heaters. 

 

I'll keep you posted regarding what I find.  Thanks again and best regards,  Randy Smith

0 Kudos
Message 3 of 3
(4,178 Views)