M2 Auto-Leveling Kit (Beta)

[PcS]

Yes I am. I will post it up when I get home tonight .

[goopyplastic]

Does shimming the spider and removing the springs help with the deflection of the bed during printing? The mic6 bed has much more heft and I noticed that it moves up and down more than the boro at the edges while printing. I have just dialed down the print speed a bit to compensate... open to suggestions though!

[PcS]

It helps. I shimmed and bolted it solid. If you try to run too fast you will still get some movement from the z rods deflection . I was considering taking my heated bed to the local machine shop ....my dad !!! Lol and having him make it flat and try that . I will try tonight to upload firmware. Got in too late last night !

[goopyplastic]

Not sure that the normal heat spreader or regular / thinner aluminum will work out as it will tend to deform with heat etc but maybe the leveling will overcome that.

I am curious with trying to get 1.0.2 working with the normal 20x4 smartlcd (non glcd) to reduce the lcd overhead more, not having much luck in the firmware though rambo + lcd is a little more complex than I can see to handle on my own at the moment

Josh did post a new firmware on the lcd page for 1.0.2 and glcd and works great with auto leveling with a few quick changes, not sure I want to use glcd's though.

[goopyplastic]

Got it all going, v3b, probing, 20x4 lcd yay happy dance.

[PcS]

Awesome !!!

[Tim]

I want to know the full specifications of this kit and also about its applications.
Can anybody please post it here?
After that i ll discuss more about its electronic part.

The kit is basically just a 3-wire inductive proximity sensor LJ12A2-4-Z/BZ. The proximity sensor works by detecting a metal surface underneath, so it is necessary to have a metal surface to the print bed, like MIC6 tooling plate, or a Zebra plate which has a copper layer just under the print surface. The detector signals when within 4mm of the metal surface. It is mounted next to the extruder hot-end (but far enough away not to get too much radiant heat). In the "standard kit" that was (is?) being sold by MakerGear, the sensor is powered off of the 5V supply and is a direct replacement for the Z-min endstop switch. The 5V supply is out of spec for the sensor part, and appears to only affect the response time of the device, but not enough to make a significant difference for the use of the Z-probe (some of us, however, have configured the electronics differently to give the Z-probe a 24V supply, but it is necessary to make sure that the value never exceeds 5V at the Z-min switch connector).

In the standard configuration, the simplest application is that it simply replaces the Z-min switch and provides a more accurate measurement of the distance between the plate and extruder nozzle than the original Z-min switch, because it's mounted directly to the extruder. A more complicated application uses the Marlin firmware codes (I don't know them off the top of my head) to do a 3-point probe. The firmware then takes care of tracking the height of the bed, so that the bed no longer has to be perfectly level.

Before you re-invent the wheel, please look through the various forum threads on the Z-probe, as a number of us have modified the standard configuration for different purposes.

[goopyplastic]

I noticed some weird behavior on my machine. It will auto probe, then while printing the nozzle will go off randomly to different spots on the bed far away from the print somewhat slowly, then it drifts back to where it started and keep on printing. It will keep do this randomly it's pretty bizarre. I think it might be combination of bad sensor and running at 5v, going to order a replacement and try the voltage divider. Anyone else ever see anything like this?

[Tim]

I noticed some weird behavior on my machine. It will auto probe, then while printing the nozzle will go off randomly to different spots on the bed far away from the print somewhat slowly, then it drifts back to where it started and keep on printing. It will keep do this randomly it's pretty bizarre. I think it might be combination of bad sensor and running at 5v, going to order a replacement and try the voltage divider. Anyone else ever see anything like this?

It sounds more like deliberate behavior. The Z-probe, like any endstop switch, is only active at certain times, which doesn't include during printing. The endstops aren't even checked during printing; if they were to report false positives, nothing would happen. For probing, the Z-probe does its thing, saves the results, and then the rest is just doing calculations based on the Z-probe results and does not involve the Z-probe itself any more. What you describe sounds more like the slicer has inserted some sort of action for a nozzle wipe or something. If it's in the g-code, then the travel points should show up in a g-code previewer.

[innkeeper]

Hi Tim,

it has been a while since i was digging in the marlin firmware but...
there is a setting in the marlin firware to enable/ disable this functionality, which .. if i recall right is something called min_software_endstop and max_software_endstop or something similar to that.
if true, it will not let you go over the limits, of false it will let you, but homing will still function.

many set this ad false, to prevent false postivies from stopping movement. the risk of turning this on and off is dependent on your application. in the case of relatively low torque 3d printers, threes no harm in having it off and many do that.

when enabled, in my experience, it only prevents travel in the direction of the triggered end stop. So even if max_software_endstop is enabled and triggered it should only prevent movement in the direction of the trigger. in the case of the Z axsis, while printing, it would be a non issue as you are always moving away from that endstop.

note this is marlin only, not applicable to other firmware.