Reflective sensor

Hi there

I humble ask (as I could not find any comments on it) is there any experiance with reflective opto sensores. Especially with regadring z-sensing of the glass (or what you have on it). My ideas (acturlly my son) are that sensing the distances to the glass plate and not to the glass plate via the frame, clamp, bolt, etc will give the most accurate distances measure. Further it will help in leveling the glass bed.

Are these ideas centuries to late and well testet/ proven not to work/ works but other consiterations eliminated the approch?

Please let me/us know - if it is untried... I have the sensor and the weekend is approching...
Moose

... there is some experience with optical sensing - when searching through posts older than 30 days you have to set the option "All Dates"

Some people use optical sensors others inductive ... or any type they actually have at hand.

From an old robotic project I have short distance reflective sensors for detecting black/white borders (or the absence of the ground) with accuracies of maybe 0.1mm ... but inductive sensors achieve 0.01mm or even better (have some form Omron with 0.6 microns resolution over 2mm measuring distance)

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

If we are talking about same thing, (something like [www.vishay.com]), then they are not used to measure distance, they only measure a change in light level. They typically have a working range of 0.1 to 0.5 inches.

If you are talking about a laser range finder (aka laser tape measure), the home improvement store versions have around a 25 foot range and are accurate to about one sixteenth of an inch. Cost is about $90 (USD). You can build your own, but I think it actually costs more.

Ultrasonic range finders are cheaper and I'm pretty sure there is a discussion thread about them.

overkill

Hi again

Sorry for not beeing acurate. Yes there are a lot of opto work but it all uses the "light beam brake" method and are again reletive to the frame, clamps, etc. I have been reading that (and I have made one for a fintration equipment mayby 10 years ago - it is wonderful in that application as we would like to know where the filtration piston was relative to the top and base of the filter cake). but back to this application...

I read that alot is trying to go away from the micro swich - not pricise enougth. Then Opto, then Hall, and then happy. But they are all relative to you frame, etc. Thus your distances "lock" are never more precise than your build. Further what you need is to know where the glass surface is relative to you tip. That made me think about the reflective opto sensor (e.g. [www.engr.ucsb.edu]) it do not tuch the surface, it is consistent in signal, it is small, it detects the surface.

In the ideal situation it would be possibl to compensate for sligthly tilted glass-beds by tackking off the hight in the 4 conors, and propperly other ideas that I have not come across.

Let me hear any experiances.

Yours
Mogens

For a good sensor, you need a sharp on/off transition, or a fairly linear characteristic that you can convert to a distance. In both cases it needs to be consistent and repeatable. I haven't checked, but a microswitch probably provides a pretty good on/off point, there is some noise around the switching point. For the price and simplicity, it is not bad.

I would love to have a precise way of measuring distance to the bed, even nicer if it was non contact. Ideally the sensor would have a resolution of 0.01 mm and a range of 0-10mm, about 10 bits. Failing, a sensor with a 0.01 resolution and digital output.

The problem is if you replace the cheap microswitch with a cheap sensor, it is unlikely to be better without additional conditioning. The one you linked to doesn't have a sharp switch point nor a linear characteristic. That is not surprising, it is not designed to be a precision distance sensor. You could add a threshold detector to produce a digital output, I'm not sure how repeatable that would be. Alternatively, feed the output to an ADC input, but you would need to carefully calibrate the ADC to get accurate results. To get 0.01mm accuracy you would need 1% resolution.

There are some opto-sensors which provide a TTL output. Problems with optos include the need for a good reflector, and sensitivity to ambient lighting.

Nophead has a design for a bed probe which is cheap and simple using a microswitch, it is probably good enough.