Author Topic: 1 mm accuracy IC sensor  (Read 3188 times)

0 Members and 1 Guest are viewing this topic.

Offline KlemkenTopic starter

  • Contributor
  • Posts: 10
  • Country: si
1 mm accuracy IC sensor
« on: February 19, 2020, 08:34:58 am »
Hello,



anyone knows of any distance IC sensor (max size similar to a TQFP48 package), touchless (laser, tof ,reflective,... ), with accuracy of +/- 1 mm. Distance i need to sense is 1 mm to 100 mm.

I checked ton of site but best i found in this size had accuracy of +/- 5mm. Thanks for any help in advance.



Kind regards.
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15799
  • Country: fr
Re: 1 mm accuracy IC sensor
« Reply #1 on: February 19, 2020, 05:56:34 pm »
Given your size requirement, I kind of doubt you'll find anything like that.

Example of sensors that could match your accuracy/range requirements: https://www.sensopart.com/en/products/distance-sensors/distance-sensors/ft-25-ra-170-pnsul-m4m.html
Definitely not small and likely rather expensive.

Many of the usual sensors are closer to +/-5mm, and I'm personally not aware of any that would entirely fit within a TQFP48 envelope.

If you can disclose your application, we may be able to find some alternative ideas...
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: 1 mm accuracy IC sensor
« Reply #2 on: February 19, 2020, 06:15:02 pm »
 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7044
  • Country: nl
Re: 1 mm accuracy IC sensor
« Reply #3 on: February 19, 2020, 06:20:57 pm »
Would attaching a pattern on the other surface and using a camera be an option?
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15799
  • Country: fr
Re: 1 mm accuracy IC sensor
« Reply #4 on: February 19, 2020, 06:26:31 pm »
Would attaching a pattern on the other surface and using a camera be an option?

For sensing distances starting from 1mm, using a camera seems a bit tough.
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: 1 mm accuracy IC sensor
« Reply #5 on: February 19, 2020, 06:42:17 pm »
1mm is tough in any case, move the sensor far back and work with offset.
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20361
  • Country: gb
  • 0999
Re: 1 mm accuracy IC sensor
« Reply #6 on: February 19, 2020, 06:52:53 pm »
Are you trying to measure the proximity of an object or just the position of something, such as a print head? If it's the latter, a linear encoder will do and 1mm in 100mm accuracy is pretty common and cheap. If it's an actual proximity sensor, then it's much more complicated and I don't know of anything which will fit your space constraints.
 

Offline KlemkenTopic starter

  • Contributor
  • Posts: 10
  • Country: si
Re: 1 mm accuracy IC sensor
« Reply #7 on: February 20, 2020, 08:55:28 am »
Thank all for info, i ordered VL6180X-SATEL board (and also VL53L1X)from ST and will see how much i can get out of the sensor.
 

Online Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 2077
  • Country: us
    • The Messy Basement
Re: 1 mm accuracy IC sensor
« Reply #8 on: February 20, 2020, 12:56:56 pm »
I'm pretty sure it can be done with ultrasonics, but don't know of any commercial product. You'd probably have to DIY the whole thing.
 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 7044
  • Country: nl
Re: 1 mm accuracy IC sensor
« Reply #9 on: February 20, 2020, 01:11:07 pm »
Without going to MHz range it's hard to keep a tight beam with ultrasonics. Also you'll need temperature compensation.

Though the ST solutions hardly seems temperature insensitive either.
« Last Edit: February 20, 2020, 01:13:15 pm by Marco »
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15799
  • Country: fr
Re: 1 mm accuracy IC sensor
« Reply #10 on: February 20, 2020, 01:35:53 pm »
Without going to MHz range it's hard to keep a tight beam with ultrasonics. Also you'll need temperature compensation.

Agreed. And even so... keeping 1mm accuracy over 100mm with ultrasonics will not be that easy, and will never fit within the volume the OP was aiming for.

Though the ST solutions hardly seems temperature insensitive either.

The ST proximity sensors will not give you nearly the accuracy the OP is looking for. They are typically used as proximity sensors for mobile phones/tablets just to estimate how close you are to them... Just take a look at the graphs in the datasheets. You won't get any better than +/-5mm or worse even on the shortest range model. You may be able to partially overcome this with frequent calibration, but I wouldn't count on that. Or that may make the solution very cumbersome to use in the field.

Knowing more about the application may help think of an alternative. What is the typical environment? Is the distance going to be always measured in open air? What material is going to be the object to measure the distance from? What's the typical operating temperature range?

If the object is conductive, one possible approach could be to combine an inductive or capacitive proximity sensor (for the short distances, such as < 20mm) and ultrasonic or ToF sensor for the rest of the range. The 1mm accuracy over the full range would still be hard to get, and the overall size... much larger than expected.

 

Online AndyC_772

  • Super Contributor
  • ***
  • Posts: 4315
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: 1 mm accuracy IC sensor
« Reply #11 on: February 20, 2020, 01:47:33 pm »
What is it that you're trying to detect, and what's the end application?

Offline KlemkenTopic starter

  • Contributor
  • Posts: 10
  • Country: si
Re: 1 mm accuracy IC sensor
« Reply #12 on: February 20, 2020, 02:45:20 pm »
Basically i will measure distance from sensor to a wall covered by different color and materials of fabric. That is why i am sticking with ToF technology. There will be several sensors at 1cm space beetwen each to wor kas a simplified lidar system. They will be all mounted on a rotating pcb, that will rotate with 20RPM (smooth bldc motor rotating in, no jerky step motor).
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15799
  • Country: fr
Re: 1 mm accuracy IC sensor
« Reply #13 on: February 20, 2020, 03:35:52 pm »
Ouch, so to make things even easier, the sensors are going to move! :-BROKE

Why do you need it to work at such a short min distance (1mm)? That would mean that your sensors are rotating at 1mm from the wall, sounds pretty risky on top of that, unless I missed something.
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 10174
  • Country: gb
Re: 1 mm accuracy IC sensor
« Reply #14 on: February 20, 2020, 03:51:04 pm »
Basically i will measure distance from sensor to a wall covered by different color and materials of fabric. That is why i am sticking with ToF technology. There will be several sensors at 1cm space beetwen each to wor kas a simplified lidar system. They will be all mounted on a rotating pcb, that will rotate with 20RPM (smooth bldc motor rotating in, no jerky step motor).

It's important to put major requirements like that in your OP, so you don't waste peoples' time.

If you want ToF then Ultrasonic will have more resolution (not accuracy) at very short distances but the sensors will be bigger than your footprint and spacing requirement (fabric might be an issue too).

Maybe you can do something with small reflective opto detectors rather than ToF, but again varying surface reflectivity is an issue. 1cm spacing sounds unachievable when you take into account motors and slip rings/couplers though. [EDIT: Ah you maybe mean 1cm vertically stacked on a a single PCB] I can't see a Laser ToF system being that compact either.
« Last Edit: February 20, 2020, 03:57:13 pm by Gyro »
Best Regards, Chris
 
The following users thanked this post: tooki

Offline KlemkenTopic starter

  • Contributor
  • Posts: 10
  • Country: si
Re: 1 mm accuracy IC sensor
« Reply #15 on: February 21, 2020, 07:28:22 am »
Ye so veeeerrryyy basicaly, what i need is to make a miniature slow moving 3D scanner based on lidar ToF principle. But yea, after a lot of research, there isnt any technology capable of meeting my requirements.  Due to varying surfaces reflective principle is not usable. I change my accuracy from 1mm to 3mm so now i am testing light based ToF VL6180X and sound based ToF CH-201.
 

Offline DaJMasta

  • Super Contributor
  • ***
  • Posts: 2393
  • Country: us
    • medpants.com
Re: 1 mm accuracy IC sensor
« Reply #16 on: February 21, 2020, 07:41:14 am »
There are definitely technologies that can do it, but probably not in priceranges you're looking at.  Commercial/industrial fine detail 3d scanners often use structured light in conjunction with photogrammetry, which uses cameras and computing to build a model from pictures at many angles and this can produce very high resolution and detail when done correctly.  Lidar sensors can be made to be this sensitive, but not that inexpensively, so for very accurate displacement measurements, laser interferometry is used instead (and actually, used laser displacement sensors may be cheap enough to start moving towards low budget applications).

For particularly slow moving scans, you can also try just averaging your measurements, even to the point of introducing some kind of external controlled noise source to try to reduce error, but you still do run into issues with reflectiveness of materials in some cases - though a light spray of a matte color paint can often overcome this, if it's an option.
 

Offline KlemkenTopic starter

  • Contributor
  • Posts: 10
  • Country: si
Re: 1 mm accuracy IC sensor
« Reply #17 on: February 21, 2020, 01:20:29 pm »
Thanks for your input. Price is actually not the biggest problem (manufacturing costs can be up to 500 or 800€). Main problem is that whole thing must be as compact, because it will be used to scan inside of a object, that size of a average 0,7 l bottle.
 

Offline Prehistoricman

  • Regular Contributor
  • *
  • Posts: 216
  • Country: gb
Re: 1 mm accuracy IC sensor
« Reply #18 on: February 21, 2020, 01:32:13 pm »
If you wanted to go custom, you could make your own ultrasonic range sensor. The poor minimum range of most US solutions is due to this dead zone created by the transducer itself. The transducer is designed to resonate at the chosen US frequency, and it will continue oscillating at that frequency for a short while after you stop driving it. So the non-cost effective solution is just to have two US transducers: one for transmit and the other for receive. I think most commercial designs don't do this because the transducer is the most expensive part.

The company I work for has been working on eliminating this dead zone with a single transducer for the past 6 months to a year. Still we haven't been able to see objects closer than 15cm.

Thanks for your input. Price is actually not the biggest problem (manufacturing costs can be up to 500 or 800€). Main problem is that whole thing must be as compact, because it will be used to scan inside of a object, that size of a average 0,7 l bottle.
Ah. So US is probably not the way. You will have so many reflections off the sides of the object that you won't be able to see the end.

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15799
  • Country: fr
Re: 1 mm accuracy IC sensor
« Reply #19 on: February 21, 2020, 02:07:25 pm »
Ye so veeeerrryyy basicaly, what i need is to make a miniature slow moving 3D scanner based on lidar ToF principle. But yea, after a lot of research, there isnt any technology capable of meeting my requirements.  Due to varying surfaces reflective principle is not usable. I change my accuracy from 1mm to 3mm so now i am testing light based ToF VL6180X and sound based ToF CH-201.

I was suspecting it was something like a 3D scanner. I can understand the requirements about accuracy/resolution (could even understand if you were after something better than 1mm), but I'm not sure I understand your requirements in terms of distance range. Classic 3D scanners, even the small ones, are more working in the 100mm-200mm range or something, and often based purely on image processing, as Marco suggested. But again not as such short distances, why would you want to be so close to the target object? And why do you  need such a tiny sensor? There's probably still a bit of your project that we don't understand (but you might not want to disclose it).


 

Offline DaJMasta

  • Super Contributor
  • ***
  • Posts: 2393
  • Country: us
    • medpants.com
Re: 1 mm accuracy IC sensor
« Reply #20 on: February 21, 2020, 04:35:04 pm »
Main problem is that whole thing must be as compact, because it will be used to scan inside of a object, that size of a average 0,7 l bottle.

Perhaps less of a problem than you think.  Laser interferometers sometimes come with fiber optic heads (I've seen this more on vibrometers than displacement sensors, but they can usually do displacement as well), and photogrammetry could potentially be used on images from a boroscope - the lens distortion would probably be similar to what 360 camera units already have to stitch together.


Depending on the void you're measuring and the stuff on the outside, I wonder if you could use something like ground penetrating radar or even x-rays to get your data entirely externally.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf