Long-distance proximity sensor
-
@afllak what accuracy do you need?
Ultrasonic is not accurate at all because it is affected by temperature.
There are some low-cost laser distance sensors available now, but they are not reproducible enough to use as a Z probe.
-
@fcwiltyou are absolutely right, I didn't notice that detail until you brought it to my attention. Thanks a lot.
-
@dc42 It needs to be accurate enough for Homing the Printer and for Bed Leveling. The printer size is 2200X1250X1250mm and I'm using a Pulsar Pellet extruder so I need a sensor with a long sensing distance. Is there a Duet sensor that can do this job, or if you can recommend something that will work well with my Duet 3 6HC Board? Thank you.
-
@afllak said in Long-distance proximity sensor:
I'm using a Pulsar Pellet extruder so I need a sensor with a long sensing distance.
What is it about the setup that requires a long sensing distance?
Frederick
-
@fcwilt The Pulsar is over 300mm long and I don't want to mount the sensor to the tool, instead I would like to use the Pulsar mounting bracket to hold the sensor.
-
@afllak said in Long-distance proximity sensor:
@fcwilt The Pulsar is over 300mm long and I don't want to mount the sensor to the tool, instead I would like to use the Pulsar mounting bracket to hold the sensor.
Could you use one of the servo controlled probes with an extended actuating pin run down past the Pulsar?
It would take some design on your part.
I have a basic design using an optical beam break sensor but the actuating pin is only 5mm long.
A longer pin would need additional support to keep it in position.
Frederick
-
@fcwilt No, the plan was to use ultrasonic or laser but the problem is I don't have any experience with both of them and because I'm using an ABS sheet for bed surface I can't use an inductive sensor.
What you suggested might work but I don't know how good it will be with high temperature, it can get to 80C inside the printer. -
@afllak said in Long-distance proximity sensor:
What you suggested might work but I don't know how good it will be with high temperature, it can get to 80C inside the printer.
80c - wow - that's warm - it might not be compatible with the servo and IR beam break detector.
You can easily replace the beam break detector with a simple micro-switch but the servo would have to be spec'd to work at 80c.
Frederick
-
@fcwilt I tried that but the long-distance make it hard to mount and to keep straight, also I'm looking for something off the shelf so it will be easy to replace in the future.
-
@afllak said in Long-distance proximity sensor:
@fcwilt I tried that but the long-distance make it hard to mount and to keep straight, also I'm looking for something off the shelf so it will be easy to replace in the future.
Perhaps you could use a laser pointer type device in the much the same way that @dc42 IR probe works - by shinning the beam at an angle to a mirror on the moving part and having it bounce off to a sensor.
Frederick
-
what about piezo sensors under-bed? might be more useful ...
coincidentally I did look at ultrasonic sensors over the last few days. If you are willing to spend a couple hundred bucks, you can get some with 400mm sensing distance and 30hz reporting time ... which will get you a probing speed of about 2mm/sec. They are temp stabilized (assuming stable ambients around the probe to the measured distance of course), and come in NPN configurations which might just work like an inductive probe. I don't have the funds for an experiment and my printers are all smaller than the large sensing distance.
Dead zone of these probes is something between 20 and 50mm, which might be more useful, but then their precision is not super great either (something on the order of a magnitude worse than the usual z probes on 3d printers).
-
@afllak there are laser sensors that are expensive, but not crazy expensive that work over longer ranges. From what I have seen, the longer the range, the wider the trigger window, and the lower the repeatability. One example device range, the version with a 30mm range has a window of ±5mm and repeatability of 0.1um, where at at 400 ± 200 mm version the repeatability is 300um. Now repeatability of 0.3mm for a printer with a 0.3mm layer height would be terrible and completely unworkable. With a pellet extruder like the pulsar you have a 1.0mm nozzle to a 5.0mm nozzle so assuming you are working with a large layer height, 0.3mm repeatability of the sensor might be ok.
-
@fcwilt Sounds complicated
-
@oliof I don't know if the piezo sensor would work with an ABS sheet laminated to an MDF Board they are very heavy.
I don't want to spend this much on an Ultrasonic sensor It will be better to go with a laser one. -
@t3p3tony Yes, I think a laser sensor would be the best choice, and Yes 0.3mm is nothing, I'm using a 5mm nozzle and 4 mm layer height.
-
Thank you everyone I really appreciate your help.
-
@afllak said in Long-distance proximity sensor:
@t3p3tony Yes, I think a laser sensor would be the best choice, and Yes 0.3mm is nothing, I'm using a 5mm nozzle and 4 mm layer height.
Well glad to hear you may have found a solution.
Frederick
-
@fcwilt Thanks, appreciate your help.
-
@afllak Take a look at laser interferometers. The commercial/industrial units are expensive, but it might be an opportunity to figure out how to make it cheaply. The resolution is very high- like 5 nm, which you don't need, so cheaper optics might give the needed performance without the high cost.
Example: https://www.keysight.com/us/en/assets/7018-06711/technical-overviews/5964-6190.pdf
-
We experimented with a sensor based on two Adafruit TOF (time of flight) laser chips two years ago. Specifically, they were the VL6180X chip.
The lasers were mounted angled on a smooth and level surface parallel to the bed and the average of the two measurements obtained was calculated with and Arduino.
We achieved very good repeatability, but from a distance that was too close for what we needed. Corrections or changes of focus would have to be made in how to apply triangulation, optical elements or other methods, but it was a promising attempt.Perhaps someone will dare to develop it to apply into a reliable "solid state" autolevel, which has a greater distance than current systems.
Currently we make a long probe with a flag with an optoendstop that measures the activation point, and a solenoid to deploy and pick up the probe. But this system is "only" reliable up to 20-25mm of free probe travel.