Multifunction DAQ
NI 9237 Quarter Bridge Absolute Accuracy
Solved!
Given a NI 9237 Bridge Completion Module with NI 9944 Accessory Terminal and 120 ohm strain gauges with GF = 2.11, how do we calculate strain accuracy?
I was told that the absolute accuracy for quarter bridges is given by
Absolute Accuracy = (Gain Error * Reading) + (Offset Error * Range) + Noise + Half Bridge Tolerance + Quarter Bridge Tolerance.
Since I'm using the NI 9944, the Quarter bridge tolerance would be 500 uV/V (given by NI R&D). The Half Bridge tolerance is given in NI 9237 manual to be 1.2 mV/V.
1) There are no Input "Noise" values for quarter bridges in the NI 9237 manual; do we just use the half bridge noise?
2) If I am using a sampling rate of 1.613 kS/s (the lowest valid sampling rate for NI 9237) and my system is still in its 1st year of usage, is my gain error still 0.05%? It's noted that 0.05% applies to 50 kS/s; if the gain error doesn't apply at my low sampling rate, how do I find the gain error?
3) If my maximum/minimum strain measurements are +/-600 ue (microstrain), how do I change my "Reading" and "Range" values in the absolute accuracy equation above, if they need to be adjusted?
4) Why does the absolute accuracy for a quarter bridge set up include half bridge tolerance?
5) Is the equation for conversion of voltage accuracy to strain accuracy for quarter bridge, , where U is the voltage accuracy given by the absolute accuracy equation above.
Sample calculation using the following assumed values for quarter bridge:
Gain Error/Offset Error = 0.05%
Reading/Range = 25 mV/V
Half Bridge Noise = 1.6 mV/V * 3
Half Bridge Tolerance = 1.2 mV/V
Quarter Bridge Tolerance = 500 uV/V
Absolute Accuracy = (0.0005*.025 V/V) + (0.0005*.025 V/V) + ((1.6E-6)*3 V/V) + (1.2E-3 V/V) + (500E-6 V/V) = 25 mV/V +/- 1.73 mV/V
Strain Accuracy = -4(V/Vex) / GF(1+4(V/Vex)) = -4*(25 mV/V +/- 1.73 mV/V) / (2.11)*(1+4*(25 mV/V +/- 1.73 mV/V))
How would I simplify this strain accuracy to get a reading +/- range of accuracy?
Thank you for any help.
Hello,
I'll do my best to answer your questions.
1) There are no Input "Noise" values for quarter bridges in the NI 9237 manual; do we just use the half bridge noise?
If you look on page 8 of the device operating instructions and specifications document you'll find that the NI 9944 completes the half bridge. So yes, you will use the half bridge noise.
2) If I am using a sampling rate of 1.613 kS/s (the lowest valid sampling rate for NI 9237) and my system is still in its 1st year of usage, is my gain error still 0.05%? It's noted that 0.05% applies to 50 kS/s; if the gain error doesn't apply at my low sampling rate, how do I find the gain error?
This applies at a data rate of 50 kS/s. Lower data rates can have up to 0.20% of reading additional gain error. This can be found on page 24 of the device operating instructions and specifications document
3) If my maximum/minimum strain measurements are +/-600 ue (microstrain), how do I change my "Reading" and "Range" values in the absolute accuracy equation above, if they need to be adjusted?
The reading will be a particular value for the sensor and the range will be the range of voltage for the device.
4)Why does the absolute accuracy for a quarter bridge set up include half bridge tolerance?
I believe the response for number two will answer your question.
5) Is the equation for conversion of voltage accuracy to strain accuracy for quarter bridge, , where U is the voltage accuracy given by the absolute accuracy equation above.
Yes, I believe this is the correct equation. Please let me know if you have any further questions.
Hi Joe,
Thank you for responding and answering most of my questions!
2) If I am using a sampling rate of 1.613 kS/s (the lowest valid sampling rate for NI 9237) and my system is still in its 1st year of usage, is my gain error still 0.05%? It's noted that 0.05% applies to 50 kS/s; if the gain error doesn't apply at my low sampling rate, how do I find the gain error?
This applies at a data rate of 50 kS/s. Lower data rates can have up to 0.20% of reading additional gain error. This can be found on page 24 of the device operating instructions and specifications document
Do you happen to know how to calculate what Percent of Reading (Gain Error) I would get depending on what sampling rate I use? Otherwise, I guess I could use the 0.20% gain error as a worst case scenario.
5) Is the equation for conversion of voltage accuracy to strain accuracy for quarter bridge, , where U is the voltage accuracy given by the absolute accuracy equation above.
Yes, I believe this is the correct equation.
This is more of a math question - since U will be in the form of a reading +/- a range, (e.g. 25 mV/V +/- 1.73 mV/V), do you know how I would simplify or interpret the strain accuracy after substituting in the value for U?
Strain Accuracy = -4(V/Vex) / GF(1+4(V/Vex)) = -4*(25 mV/V +/- 1.73 mV/V) / (2.11)*(1+4*(25 mV/V +/- 1.73 mV/V))
2) If I am using a sampling rate of 1.613 kS/s (the lowest valid sampling rate for NI 9237) and my system is still in its 1st year of usage, is my gain error still 0.05%? It's noted that 0.05% applies to 50 kS/s; if the gain error doesn't apply at my low sampling rate, how do I find the gain error?
This applies at a data rate of 50 kS/s. Lower data rates can have up to 0.20% of reading additional gain error. This can be found on page 24 of the device operating instructions and specifications document
Do you happen to know how to calculate what Percent of Reading (Gain Error) I would get depending on what sampling rate I use? Otherwise, I guess I could use the 0.20% gain error as a worst case scenario.
Yes, I would calculate the Gain Error for a worst case scenario (0.2%)
5) Is the equation for conversion of voltage accuracy to strain accuracy for quarter bridge, , where U is the voltage accuracy given by the absolute accuracy equation above.
Yes, I believe this is the correct equation.
This is more of a math question - since U will be in the form of a reading +/- a range, (e.g. 25 mV/V +/- 1.73 mV/V), do you know how I would simplify or interpret the strain accuracy after substituting in the value for U?
Strain Accuracy = -4(V/Vex) / GF(1+4(V/Vex)) = -4*(25 mV/V +/- 1.73 mV/V) / (2.11)*(1+4*(25 mV/V +/- 1.73 mV/V))
Just use the worst case scenario again to calculate for both positive and negative values regarding 1.73mV.
