Autoleveling bed

Hey RepRep!!

I have been printing for a while and skulking the forums, but this is my first post... Let's go!

So I have been toying with an idea and I figured I would put it to the community, it goes like this:

I am designing an autoleveling bed that uses the same 3 point mounting scheme that everyone uses. The difference is that the bolts do not have nuts holding them down they are held by a servo or geared servo. The nozzle, being commanded by Marlin script or Pi, goes to the top of each bolt (covered by glass) and it determines the height. This is done by either a laser attached to the x-gantry or solenoid switch. After height data is collected, it is sent to the micro controller and it calculates by how much it has to turn the servo so that the bed is leveled. This can be done iteratively until an acceptable tolerance is met.

What do you guys think?

too complicated, just put probe on hotend...


dont get me wong, i didnt say its bad idea, just it cost too much.

a lot of people drive heated bed on one axis, thats a lot of additional weight

The only thing is that putting a probe and doing something like the Kossel won't work on a printer with two heads. I guess this idea is mainly for ultimaker style printers.

That would seem to add a degree of complication that's probably not needed, but I was wondering in another thread only yesterday if it would be possible and or desirable to have an auto calibration routine for a Delta by using a metallic plate connected back to the controller, with the other side of the "switch" being the hot end nozzle, which could be slowly microstepped until it contacts the calibration plate. The other thought was a piece of glass with silver foil stuck on it.

In theory, (I think) it could be used to ensure correct centre position ( I was thinking Delta for this) after checking the upper stop limits, and then also used to calibrate bed level, then check again to make sure that everything is right, then it could be used to make the needed adjustments to deal with mechanical variances. the aim being to get a good overall performance across the entire print bed. The only adjustment then is to change the max Z to take out the calibration plate depth.

What I don't yet know, as this is all very new, is if the firmware is capable of using the adjustment information in an appropriate manner, or if the calibration of a delta is still at the moment a bit of a fudge, and I would be the first to admit that my mathematical geometry capabilites are not good. I am confident I will be able to build a Delta Pi to the accuracy needed, but working out the finer points of coordinate geometry would not be my strong point, hence my interest in a self calibrating printer.

In theory, this would be a viable alternative to a probe on the hot end carrier, and be repeatable and checkable to make sure no wear or belt stretch has happened over time.

Thoughts, comments, whatever welcomed, this is part of how I learn, and learning the theory is cheaper than printing rubbish,

Steve

---

Shore, if twas easy, we'd all be doin it

Irish Steve

So Delta printers have their bed leveling in the software so I was thinking of how we can do it for cartesian printers. I saw this and it looks promising but it is still not completely automatic and I am not a big fan of smooshing the nozzling into the printbed.

[www.youtube.com]

thoughts?

That looked very agricutural to me, there was significant movement in the bed when contact was made, and it was movement that would seem to put accuracy at risk.

Not sure I'd be too comfortable driving the head into the bed with that degree of force, it only has to touch as far as I can see.but that is a different design, so maybe it's OK.

---

Shore, if twas easy, we'd all be doin it

Irish Steve

[hydraraptor.blogspot.co.uk]

[www.youtube.com]

[www.youtube.com]

---

[www.hydraraptor.blogspot.com]

Thanks for the links, especially the one that has the code on it, clearly, the information is out there, but not perhaps as clearly documented as it might be, but that's not a complete surprise, these sorts of fixes tend to get made when up to one's backside in alligators, and documenting how the swamp was being drained becomes a distraction that then gets forgotten in the heat of the moment. The more I am reading, and seeing (and I have to admit an aversion to Youtube tutorials, depending on their quality) the more I am at the same time encouraged and slightly frightened by the task I'm taking on,

It will be interesting to see how things develop over the next few weeks and months.

---

Shore, if twas easy, we'd all be doin it

Irish Steve