Deploying probe

gnydick

I want a bed probe that works like the bltouch does in terms of being able to be deployed and retracted. But I don't really like the bltouch I want something different. Are there typical parts that you can buy to create such a thing? I would like to fix a metrol limit switch that can be auto deployed. I also don't want to Jerry rig servos and printed parts.

Veti

this is what i use
https://www.thingiverse.com/thing:3303618
it works very well and behaves like a bltouch to the system

deckingman

@gnydick said in Deploying probe:

I want a bed probe that works like the bltouch does in terms of being able to be deployed and retracted. But I don't really like the bltouch I want something different. Are there typical parts that you can buy to create such a thing? I would like to fix a metrol limit switch that can be auto deployed. I also don't want to Jerry rig servos and printed parts.

A bit off topic but I can highly recommend Metrol precision switches. Mine has been working like a charm for years. Highly accurate, highly repeatable, no signal conditioning - just two wires to an end stop input.

Though I really don't like the idea of having a probe that has to be deployed, because that mechanism in itself is likely to introduce positioning errors which are far greater than the precision of the switch. My solution was to use steel dowels and sintered Bronze oil filled bushes to mount the hot end such that it is rock solid in X and Y but free to move slightly in Z. The Metrol switch is fixed to the hot end mount so that when the bed touches the nozzle, it lifts it slightly (about 0.2mm) which triggers the switch. I use a negative Z offset (roughly -0.2) to determine Z=0 (essentially the nozzle travel before the switch triggers).

It's mechanically a bit tricky to design but in operation it is simplicity itself. The nozzle is the probe - XY no offsets required, no deployment needed, no signal conditioning required, no probing speed limitations.

JoergS5

@deckingman said in Deploying probe:
such that it is rock solid in X and Y but free to move slightly in Z. The Metrol switch is fixed to the hot end mount so that when the bed touches the nozzle, it lifts it slightly (about 0.2mm) which triggers the switch. I use a negative Z offset (roughly -0.2) to determine Z=0 (essentially the nozzle travel before the switch triggers).

You could combine your method of Z lifting by the nozzle with the CNC 3D measuring sensor where the endstop are three electric contacts which is interrupted when the nozzle is lifted. (in CNC three contacts as triangle) like IMG_1408 in https://www.cnc-aus-holz.at/index.php?thread/30-projekt-3d-taster/
This should result in a very precise measurement.

jens55

@deckingman said in Deploying probe:

@gnydick said in Deploying probe:

I want a bed probe that works like the bltouch does in terms of being able to be deployed and retracted. But I don't really like the bltouch I want something different. Are there typical parts that you can buy to create such a thing? I would like to fix a metrol limit switch that can be auto deployed. I also don't want to Jerry rig servos and printed parts.

A bit off topic but I can highly recommend Metrol precision switches. Mine has been working like a charm for years. Highly accurate, highly repeatable, no signal conditioning - just two wires to an end stop input.

Though I really don't like the idea of having a probe that has to be deployed, because that mechanism in itself is likely to introduce positioning errors which are far greater than the precision of the switch. My solution was to use steel dowels and sintered Bronze oil filled bushes to mount the hot end such that it is rock solid in X and Y but free to move slightly in Z. The Metrol switch is fixed to the hot end mount so that when the bed touches the nozzle, it lifts it slightly (about 0.2mm) which triggers the switch. I use a negative Z offset (roughly -0.2) to determine Z=0 (essentially the nozzle travel before the switch triggers).

It's mechanically a bit tricky to design but in operation it is simplicity itself. The nozzle is the probe - XY no offsets required, no deployment needed, no signal conditioning required, no probing speed limitations.

This is the cat's meow that I would have liked to combine with a piezo switch. The compliance needed for the piezo switch to work introduced slop in the x/y axis but I could not think of a way to have rigid x/y with a compliant z axis. I was thinking about stamp/die mechanics but figured the weight would be way too high.
Do you happen to have a write-up or pictures of your assembly ?

deckingman

@jens55 said in Deploying probe:

..........................Do you happen to have a write-up or pictures of your assembly ?

The best I can come up with is a bit of a description about my printer that I wrote a couple of years or so ago. The printer has changed a fair bit since then but the "nozzle as probe" mechanism is fundamentally the same. About half way down this page https://somei3deas.wordpress.com/my-corexy-printer-build/

You can see it in action in this video - at about time = 2 minutes, although the 0.2mm or so of nozzle movement is almost undetectable by eye. https://somei3deas.wordpress.com/my-corexy-printer-build/

jens55

Thanks ... interesting though that the bushings are in a 3d printed seat. I would have thought that alignment would be too critical to use 3d printing directly.

dc42

@deckingman said in Deploying probe:

Though I really don't like the idea of having a probe that has to be deployed, because that mechanism in itself is likely to introduce positioning errors which are far greater than the precision of the switch.

There are ways to avoid this. I remember reading a post on the E3D Bigbox forum, where someone had moved the IR sensor so that instead of looking directly at the bed, it looked at the top of a pushrod that was pushed up by the bed. A small RC servo with a pawl on it held the rod up during normal use, and released the rod during bed probing. So the exact position of the servo did not affect the trigger height.

gnydick

@deckingman that sounds really cool

gnydick

very cool ideas, everyone

gnydick

This post is deleted!

gnydick

@dc42 I get the idea, but i'm having a hard time picturing it. Once it's released, how did it pull it back up?