Minimum output in Volt for NI 9238

[Lc90]

Hello

I am studying ISO standard 9869, focused on measuring the heat transferred through the walls of a building.

I am recording the heat-flux via a transducer and the NI 9238 is the compatible datalogger. Flexlogger is the data acquisition software.

this standard gives you a methodology on how to quantify the uncertainty budget. From the NI 9238 datasheet I am not sure which is the correct value that corresponds to ISO requirement as minimum output provided by DL expressed in mV. The datasheet instead provides you the minimum voltage input +/- 0.496 V. This is not the value I am searching for. 

Can you help in retrieving this minimum output of the DL? I am guessing the minimum is 0 from the interval given above. I don't have competencies in electrical field.

Regards

LC

[rolfk]

It sounds cryptic. I guess DL means data logger. One value I could think of being important would be the analog resolution of the measured signal. But I can’t think of how that could possibly be worded as “minimum output in mV”.

[Lc90]

Hi Rolfk. 

Thanks for your support.

yes DL means Datalogger.

is the analog resolution included in the datasheet?

Regards

LC

[ZYOng]

NI 9238 is a 24-bit AI module with an input range of +/- 500 mV.

The resolution or the smallest amount of input signal change that a device or sensor can detect is (0.5+0.5)/(2^24) = 59.6 nV

Reference: Specifications Explained: NI Multifunction I/O (MIO) DAQ

[Lc90]

Thanks a lot. Could you kindly share the page of the document you mentioned? I cannot find the equation.

Regards

LC

[rolfk · Accepted solution]

From the datasheet that ZYOng linked to:

Input voltage range (AI+ to AI-)
Nominal ±0.5 V
Minimum ±0.496 V
Typical scaling coefficient 74.506 nV/LSB

So you have a range of 1 V (±0.5 V) and a resolution of 2^24 bits. This gives as formula:

1V/(2^24) = 59.6 nV/bit

This comes close to the mentioned typical scaling coefficient in the datasheet. How this typical scaling coefficient is exactly calculate I could not find out in a cursory Google search.

Now this is the theoretical minimal voltage change that can be measured. The device has also several accuracies that are in the uV range, including gain and offset error as well as noise level.

Also the offset drift of 1.3 μV/°C will influence your signal much more than the device can theoretically detect as voltage change, so hoping to get to the 60nV accuracy is an idle hope.