I would like to have a contactless distance sensor with 10um accuracy at a distance of about 200mm. The displacements I am trying to measure are <5mm, so the 200mm are only needed because I cannot get closer. The needed measurement frequency is low (10Hz is sufficient).
So far I used an ultrasonic sensor for this application, but the accuracy is limited to about ~100um. The displacements. So I was thinking about triangulation sensors. This one for instance would perfectly fit my application: https://sensorpartners.com/en/product/sensor-instruments-l-las-lt-130-sl/
It is only very expensive. My budget is about E100,-
@TomGeorge I would like to measure the deformations of a cylinder under aerodynamics loads (thats why I would like to take a distance of about 200mm). I know my deformations are in the range of 0.0-0.5mm (by means of the measurements I did previously with an ultrasonic sensor)
@robtillaart I know the datasheet is lacking info. That is why I am asking on this forum for anyone with experience with this sensor.
@jremington you could be right that 10um is unrealistic for E100,-. But usually hobby electronics are much cheaper then industrial ones
Here are the specs for that ultrasonic sensor. You seem to be overstating the accuracy (which is stated as 1%, usually interpreted as percent of full scale, so +/- 3.5 mm for 350 mm), so I'm wondering if you are confusing resolution with accuracy.
resolution 0.025 mm to 0.17 mm, depending on the analogue window reproducibility ± 0.15 % accuracy ± 1 % (temperature drift internally compensated)
For sensors, accuracy is absolute, referred to international standards and requires calibration. Expect to pay 10x for each factor of 10 increase in certified accuracy.
Why I can use the ultrasonic sensor is that I am only measuring deformations. So first I wait for half an hour to make sure my setup is thermally stable. Next I calibrate my sensor to zero when no load is applied (so that is why I am not concerned about accuracy). Finally, I average over a lot of samples to increase the precision to about 0.1mm.
Glad to have that cleared up. Manufacturers of professional equipment charge for the certified accuracy, so you can save some money by looking at the precision (sometimes referred to as repeatability) instead.
For a factor of 10X increase in precision with ultrasonic sensors, the frequency has to increase, so look in the 3-4 MHz range.
So you would prefer an ultrasonic sensor with MHz transducer frequency over a laser triangulation sensor?
I have done a quick google search, but I cannot find a ultrasonic sensor with MHz frequency. Do they have a specific name? Do you have any suggestions?
Thanks a lot by the way for your helpful responses!
The attenuation coefficient for ultrasound in air is 12 dB/cm/MHz, so it could be used in air for short (cm) distances. But again, probably not generally practical.
The uncertainty you are looking for is not unrealistic but I do not see it happening on your proposed budget. In the past I have used Micro Epsilon Lazer Measuring Systems but they are way, way beyond your budget. I don't know of anything at a low cost that will give you what you want.
Nice to see the issue of accuracy verse precision worked out. jremington did an excellent explanation on that subject.
This application note doesn't give any data on linearity or precision, but I'd expect it to be pretty linear over short distances.
The application is using a linear Hall effect sensor to measure displacement by measuring the flux density of a magnetic field. The distance between the sensor and magnet can thus be calculated.
It's an approach I've always wanted to experiment with but never had the time. That and I only have equipment that can measure down to 0.0001"
Arduino application: you can use a high resolution A/D converter to measure the Hall sensor output. It will require a bit of experimentation, but it might yield surprisingly accurate results.
@Ron_Blain I guess you are right. The only way forward is revisiting my budget.
@TomGeorge thank you for your suggestion. I could probably improve a little by using high frequency ultrasonic devices. laser time of flight sensors are definately not precise enough.
@cedarlakeinstruments I like your suggestion of the use of an hall effect sensor. I expect quite some issues with temperature drift, but it should be doable. But it will probably a project on its own.