05-08-2025 08:50 AM
Hi All,
We are using NI USBDaq-6216's analog input channel with differential connection to collect voltage signal from our SDI2210 accelerometer.
In normal condition, the SDI2210 accelerometer will give 0.8V differential voltage at +1G(gravitiy) and -0.8V at -1G condition, which we can it the standard output value.
We have confirmed that the accelerometer can give the above standard output value, which means the accelerometer are functioning normally. However, when we acquire its voltage signal from our NI USBDaq-6216. We saw floating signal (1.0V at 1G condition, and -0.6V at -1G condition), the signal has drifted 0.2V at normal condition, and when we acquire the signal at dynamic condition in a car (Accelerometer powered by car's cigarette plug with 12V DC), the drifted value then also becomes dynamic. (sometime higher than 0.2V, and then back to 0.2V, totally unpredictable)
To solve this problem, we are tried and test may things:
We first tested the AI channel by providing standard sinewave voltage signal from AO to AI, and the results are good, indicating the AI channel also functions normally.
We have also followed the recommendation by top-left of TABLE 1 in this article: https://www.ni.com/en/shop/data-acquisition/measurement-fundamentals/field-wiring-and-noise-consider...
But the signal voltage is pulled down to a even unnormal values (R1=R2=60K ohms, the voltage is 0.6V/0.0V at 1G/-1G), TABLE 1 cannot solve our problems.
We feel it is about grounding. We also tried to connect AI- to AI GND in 6216 but problems remains.
Can anyone can give us advises to solve our floating signal issues?
Thank you very much.
05-09-2025 06:57 AM
Since you have a DC MEMS sensor, is it for shure always mounted horizontal ... in a car?
~11° would result in a 0.2 V or g shift due to gravity 😉
Thats why you usually use a inertial sensor system including accelerometers and gyros 😉
05-09-2025 08:12 AM
If you have two of these accelerometers, you can mount them in 90° and calculate the vector. It should always show up with gravity (1g) in addition to your external aceleration.
(Ok in a first step, but if you do the dynamic mechanics, rotation around an arbitary point is still another source)
05-09-2025 09:36 PM
Hi We have done all the tests in a static environment.
We put the accelerometer on a flat plate (checked with level bubble).
We will get correct accelerometer readings (0.8V at 1G gratity )from another independent DAQ and the accelerometer is powered by the DAQ.
But we cannot get a correct reading from my usb-6216, and the accelerometer is powered by a external power inverter(because 6216 cannot supply 12V power to accelerometer). The reading becomes 1V at 1G and -0.6V at -1G. (We are vert sure the accelerometer is leveled. )
We feel it is a grounding issue but we cannot a way to correct this drifting voltage. (We tried to connect AI GND to power supply's V-, but the problem is not solved and the reading even becomes 0V......)
05-11-2025 10:34 AM
Please share a simplified electrical connection diagram of the connected components and possibly a photo of the setup.
05-11-2025 06:36 PM
Hi
Below is our simplified electrical connection diagram. Thank you.
05-11-2025 11:36 PM
Your DC/DC converter may be an isolated version and you need to connect the output GND of the 24V to the GND of 12V supply. And as well use resistors to AIGND since you use it in differential mode.
Side note, looks like your ACC can work from +8V, why not directly connect to 12V source instead of 24V? is it due to wiring restrictions?
05-11-2025 11:43 PM
THANK YOU.
We have put resistors between AI- and AI GND (but 24V GND and 12V GND is not connected) then the voltages measured is pulled-down.....
Do we also need to connect 24V GND and 12V GND?
And do wen need to connect 12V GND to NI's AI GND?
Yes, using 24V for our 8-32V accelerometer is our old design, we will change to 12V power.
Yi-Hsien
05-12-2025 01:25 AM
Since your Accelerometer can output in differential mode, you are not required to connect 24V GND and 12V GND as long as you have bias resistors from AI+ and AI- to AIGND.
Do you have any other single-ended signals from sensors powered by 24V?
05-12-2025 01:36 AM
Hi
We have put bias resistors between AI+ / AI GND and AI- / AI GND. However, it doesn't work.
We put 100K Ohms and the voltage is pulled down and biased 0.4V (1V --> 0.6V)
And if the we put smaller 10K Ohms then the voltage becomes 0v....
BTW, we also have laser displacement sensors which is singled-ended but we also connect it in a differential way.