thermistor measurements using SCXI-1102 and SCXI-1303

[Kevin_Chen]

Hello, 

I have a SCXI 1303 front end module connected to a SCXI-1102

I have read the article detailing how to remove the pullup and bias resistors to enable thermistor/RTD measurements on the 1303. What the article didnt mention was where to plug in the excitation source, and what is the best voltage source to use (high precision, normal, etc)

I was wondering where I can find that information, also, could I do multiple measurement types on the same 1303? I have an analog voltage input from a flowmeter, and I would like to put that on the same module as the thermistors

Thanks!

Kevin

[johnsold · Accepted solution]

Kevin,

The pull up resistors connect to +5 V internally.  If the values are suitable for your application you can use those.  It does not give you much flexibility though.  The 1303 does not give you any other access to any power or voltage source through the screw terminals.

The pull up and pull down resistors are standard 9-pin SIP resistor networks. So you can change them in blocks of 8 channels for mixed measurements.

Lynn

[Kevin_Chen]

So if I pull up the first bank of pullup and bias resistors, then terminals 0-7 work as thermistor terminals without external excitation? I guess i would just need to use the VI to convert from voltage reading to a thermistor then!

thanks!

Kevin

[johnsold]

Kevin,

Each input channel has two resistors connected to it. One pulls up to +5 V and the other pulls down to ground. In the thermocouple application the resistors have two functions.  They provide a path for the amplifier input bias current and they provide open thermocouple detection capability.

If you remove both resistors then you have the channel input only.  You need to make sure that a DC path exists for the bias currents.

If you want to use the +5 V as excitation for the thermistors, you remove the pull down resistor network and install a pull up network with a value suitable for your thermistor.  If you used a 10000 ohm resistor network and the thermistors are 10000 ohms at 25 degrees, then the input voltage would be 2.5 V at 25 degrees and would vary according to the thermistor temperature.

If you are using some other excitation source and external resistors with the thermistors, then connect the ground of the excitation source to the 1303 ground and the thermistor output to the channel input with both the pull up and pull down networks removed.

Lynn 

[Kevin_Chen]

Hi Lynn

Thanks for the response! I guess I am not seeing why the excitation has to be 2.5V and not the +5V that comes with the input channel?

Thanks

Kevin

[johnsold]

Kevin,

A thermistor is a resistor which varies its resistance with temperature.  If you connect one end of the thermistor to 5 V and the other end to the input with no pull up or pull down resistor networks connected, no current will flow through the thermistor and the input will always measure 5 V regardless of temperature.  Since the SCXI-1102 measures voltage, not current, you need to create a voltage divider (or some othe more complicated circuit) to measure temperature with a thermistor.

With a 10000 ohm resistor to +5 V, one end of the thermistor to ground, and the other ends of the resistor and thermistor connected together and to the input you have a voltage divider which will produce 2.5 V at 25 degrees.  As the temperature increases the voltage will decrease.  The resistance of a thermistor is a function of the logarithm of the temperature, so you will need to do some math to convert the votlage to a temperature.

Lynn

[JimmyJin]

Hi Lynn,

I am using the exact same instruments (SCXI-1102 and SCXI-1303) and trying to measure temperature with thermistor. I found your answer very helpful. But I'm trying to analyze the uncertainty of mu measurement and I do have one more question: Do you know how constant the 5 V from the SCXI-1303 is? Do you know the variation/uncertainty?

Thank you,

Jimmy

[johnsold]

Jimmy,

I do not know that.

From what I have seen, I think they just use the +5 V power supply in the module for the pull up. So the accuracy is probably on the order or +/-0.5 V as that is fairly typical for regulators for logic power supplies. As for expressing it as an uncertainty, I cannot even venture a guess.

I suggest that you contact your NI Sales rep or an Applications Engineer if you have support and see if they can find someone who can help you.

Lynn

[johnsold]

Jimmy,

I had another thought: If you are not using all the channels, measure the +5 V with an unused channel. The uncertainty of the measurement is probably less than the uncertainty of the +5 V. It is likely smaller than the uncertainities of the pull up resistances.  So long as the measured noise and short term drift is smaller than the other uncertainties in the system, you may be able to calculate the effects without relying on knowledge of the behavior of the voltage source.

Lynn

[JimmyJin]

Hi Lynn,

Thank you for your prompt reply. 

I think I can try your way, but my concern is how to measure the voltage. Should I just short the channel or leave it unconneted (which I think is equivalent of open thermocouple detection and returning 5 V reading)?

Besides, the terminal block only comes with two types of resistor, 10 Ohm or 10MOhm. Do you know where can I get other types of resistors (precise and hopefully temperature insensitive) ? I'm using 10K thermistor so hopefully the fixed resistor would have the values in the same order of magnitude.

I went through the NI manual for SCXI 1303, is the "pull downs resistor" you mentioned interchangeable with "bias resistor" ? In the manual, there is only a very simple diagram of the pull up and bias resistor. Do you know where I can find a more detailed description (or electric diagram) for the chip?

Sorry for running the letter so long. I'm really looking forward to hearing from you.

Thanks again.

Jimmy