Full calibration

[lordbinky]

Hey guys, I just wanted to post I've got a 90% complete implementation of the marlin style calibration. I don't agree with all the aspects of the methodology and plan to test other variations to improve it, but for now I have it changing each tower's radius (distance to center) and angular coordinate (the angle:90,210,330). I have seen the routine improve movement between the tower and it's opposite edge, and I plan to clear up some oddities in the acceptance criteria so it stops when it's good. Hope to share this experimental routine with you smoothie-ware users soon (I'll make a huge effort to do this by the end of the weekend if I don't get it tonight) so I can get some feedback and improve it to where I feel it's pull worthy to the main repository.

[lordbinky]

....and then I read it's already implimented..... http://forum.seemecnc.com/viewtopic.php ... =90#p43631

.... Well, now I'm going to go cry in a corner for being a month behind the rest of the world.

[McSlappy]

These 3d printers have to be the fastest moving machines on earth right now....

I feel for you.

[lordbinky]

I'm going to finish mine out and share it, the other one doesn't work for me.

[GeraldO]

Curious, how did you mount your z-probe? Are you using it with a MAX or Orion (or other?)?

I've wanted to try this but I haven't found a suitable way to mount a z-probe and FSRs are out of the question if you use your HBP.

[lordbinky]

I already had a microswitch like this http://www.amazon.com/gp/product/B00BGC ... UTF8&psc=1
laying around. It fits perfectly into the top piece of the E3D printable mount.

[626Pilot]

I've been working on a custom version of ZProbe.cpp/ZProbe.h that allows using my Z probe (or any other) to do an automatic calibration. Unlike the official firmware, it supports probe offsets and implements a better version of my laborious tower rotation algorithm. (How laborious? It takes the printer an hour or longer, and the printer doesn't take breaks.)

If you want to use my probe, get it from Thingiverse (link in sig) and just mount it inside the effector. You want the poker to be as close to centered as possible, so the upper mount (closest to the circuit board) is best. If you already have a microswitch I'd use that instead, since you don't have to order a hall-theta board from Europe.

I have a github fork of the main Smoothie code that's supposed to house my Z probe stuff but I haven't figured out how to get it to play with branches. When I figure that out I'll upload the code.

[teoman]

This type of device may be beneficial if you really want to go into perfect calibration
http://m.aliexpress.com/item-desc/1227993426.html


does anyone have auto calibration for the rambo?

[626Pilot]

We already have something like that called a Z-probe. There is one in my sig. The Marlin autocalibration is supposed to work on RAMBo but I got bad results when I tried it last year.

[teoman]

But that one has a serial port. So you can have it closed loop.

[626Pilot]

The Z-probe operates in a closed loop system. There would be no way for it to work otherwise. The controller moves the effector to a given height and then descends slowly. When the probe gets close enough to the surface it triggers. It then takes a pin high (or low? I forget) on the controller, and then the controller knows how many steps the effector had to be moved downward to trigger it. Repeatability on mine was tested to ~15 microns.