M2 Auto-Leveling Kit (Beta)

[Tim]

Tim, does Smoothie have a probe "repeatability" test similar to Marlin's M48?

No, but looking at the description of the command in Marlin, it looks pretty easy to implement the same in Smoothieware.

[askernas]

Just FYI:

I have verified that the probe does not recognize the bare heated plate with BuildTak on top (no glass in between).
Will need to look around for something of the equivalent of MIC-6 here in Sweden.. and someone who can cut it for me

Also mean I have plenty of time to actually get off my butt and into the electronics store south of town to get me one of them fancy NPN 3904 transistors in the mean time.

Oh, am I right in guessing that it is really only the value between the resistors, the ratio which is important here?
I have plenty of resistors laying about, but not 27k and 6.8k. Got others that sort of match 1:4 ... 10k and 47k .. for instance.. 22k and 100k .. might be a closer match..

//Micke

[Tim]

Oh, am I right in guessing that it is really only the value between the resistors, the ratio which is important here?
I have plenty of resistors laying about, but not 27k and 6.8k. Got others that sort of match 1:4 ... 10k and 47k .. for instance.. 22k and 100k .. might be a closer match..

The ratio does not even have to be close. The NPN bipolar will need a base voltage of about 1V to switch, and it can withstand up to 6V. The higher the total resistance, the less current you're using; in some applications, a higher current is needed to keep down noise, but this is just a switch, so the sum of the two resistances can be anything from, say, 10k ohms to 500k ohms. In fact, my choice of 27k and 6.8k was more based on an earlier attempt to drive the input without using the NPN transistor.

If "Ra" is the resistor connected between 24V and the NPN transistor base, and "Rb" is the resistor connected between the transistor base and ground, then the base voltage "Vb" is:
Vb = 24 * Rb / (Ra + Rb)

And Vb ideally should be somewhere around 2 to 3 V. If you use Rb = 10k and Ra = 100k, then Vb = 2.18V, which is perfect.

[askernas]

Thank you Tim!

What would we do without you

[goopyplastic]

Any chance you'd be willing to share your bracket model? I'm also mounting the sensor to a V3b.

https://www.dropbox.com/s/kcftjnlnjes9e ... e.STL?dl=0

[goopyplastic]

Couple things I noticed getting this going tonight.

- Josh's firmware didn't work for me, PCS's did.
- v3b folks are going to want to add 9mm of probe offset in the firmware
- with the extra weight of the mic6 bed, have a g1 z190 in your end gcode is probably a good idea ... that heavy bed sure does drop fast!
- 5v seemed to work fine
- one of my probes was bad (light always on)

[PcS]

5v worked for me as well. 24v seems to be more accurate though. Glad the firmware worked !!!

[goopyplastic]

5v worked for me as well. 24v seems to be more accurate though. Glad the firmware worked !!!

Are you using an lcd? do you have a version of the firmware cooked for that? Thanks!

[PcS]

I use Octoprint or my reprap full graphic display. Both are nice options to have. I do have a firmware for this setup. I had a Viki 1.0 didn't like it. Version 2.0 didn't interest me either. I find for 10 dollars the reprap one is better at least for me .

[goopyplastic]

Are you using rambolcd adaptor?
I have plenty of those and lcd's so your firmware would still interest me