Ultrasonic distance sensor

[shantanu@india]

Hi,
I need to measure diameter of pvc pipes through non contact method.
I have procured a stereo vision camera and tried image analysis in Raspberry Pi but this method is yet to provide fruitful results.
Was toying with the idea of precision distance measurement from two diametrically opposite points and calculating the diameter.
I saw lot of ultrasonic AFE's from TI. Have anyone worked with them.
I need precision of 10 microns.

[Gamboa]

Hi,
I have worked with 40kHz ultrasound and the sensors have a 10-15cm blind zone that they cannot measure. From that distance you can obtain accuracies of cm. Higher frequencies and greater precision are used in medicine but I don't think it reaches a precision of um.

Regards,
Gamboa

[JonW]

Check out TOF sensors
https://www.st.com/en/imaging-and-photonics-solutions/time-of-flight-sensors.html

Not sure if they can get to 10um...

[trastikata]

I need to measure diameter of pvc pipes through non contact method.
I need precision of 10 microns.

Hi Shantanu,

Not really on the topic, but here's a thought ...
 
The thermal expansion coefficient of the PVC is significant. Depending on the pipe diameter, even 2-3 degrees could easily put you behind those 10um specs.

I work in the CNC programming/machining and from time to time come young engineers requiring ridiculous and absolutely unnecessary tolerances, for example few microns tolerances for bearing beads in plastic parts (or for parts that are being welded afterwards) which makes the parts unnecessarily expensive. That's why when we see tight tolerances, we usually ask what are the parts for and sometimes they are not really required.

My point is make sure that 10um accuracy in the PVC pipe is really needed.

Regards

[See_Mos]

I have had a play with VL53L0X short range laser module, but nothing serious.

They are cheap and easy to use.

Another forum member Craig has also used them.

[JonW]

You will also have to ensure you setup the measurement system to say 10x the minimum resolution you need or at least be able to calibrate it out with a high degree of precision in real time to ensuring R&R is maintained over temperature and vibration etc

Agree with Trastikata on trying to get PVC to within 10um or is 10um the resolution or accuracy you need to achieve? 

[shantanu@india]

Actually I was thinking about a digital readout which would indicate the diameter of the pipe in millimeter with two places after decimal point precision. Maybe one place after decimal point(0.1mm/100 microns) is good enough...
Plumbing pipes have to be fitted with sealing fluid to tees or elbows where snug and tight fitment is a necessity. Presently they measure regularly by vernier. There always remains a risk of batch rejection

[david]

Hi,
40kHz ultrasonic has a wavelength of nearly 1cm so even with a mix of TOF and phase offset I doubt you will resolve much less than half a wavelength.  You can certainly go higher in frequency but I wouldn't think you would get anything like 10um resolution.  Temperature and pressure variations would also impact the speed of sound and be much greater than the resolution you seek.
If it's for quality control of the extruded pipes I would suggest automating some data calipers to close on the pipe, take a reading and send it to the data acquisition software for real time analysis of the process.  OK it's not non-contact but it should be more easily deliverable.

Cheers,
David

[shantanu@india]

Yes David, I think you are right. I had thought about two linear encoders touching the pipe surface but nevar could figure out a way to ensure that they touch on diametrically opposite points.
Laser sensors have the required resolution but are far too expensive

[shantanu@india]

I have had a play with VL53L0X short range laser module, but nothing serious.

They are cheap and easy to use.

Another forum member Craig has also used them.

Trevor they have 1mm resolution

[See_Mos]

I did a search for 'non contact measurement'.  There are a lot of options such as the MicroTrak 3 but I suspect they will be too expensive if production quantities are required.

[JonW]

 Take a look at these, probably be low cost
http://www.industrial-measuringinstruments.com/quality-11774202-laser-system-non-contact-diameter-measurement-for-tube-diameter-monitoring.html

[shantanu@india]

Thanks John

[RGV250]

Hi,
How about something like this  https://www.keyence.co.uk/products/measure/micrometer/ls-9000/

Bob

[shantanu@india]

Thanks Bob.., the resolution of 2um is excellent!!
Have to check the price... quite likely to be beyond my budget.
OMRON has similar sensors which are priced around £1500