9-Point Bed Leveling and the Accuracy of Z-Probes

Code hacking talk

9-Point Bed Leveling and the Accuracy of Z-Probes

Postby thnikk » 2017-Jan-Fri-20-Jan

Hey guys. I recently had to get a replacement for the z-probe on my Play and it's been giving me some trouble. After changing my z offset countless times now, I think I've finally figured out what the problem is. I bought the replacement directly from printrbot and noticed that it was a LJ12A3-4-Z/BY. The important thing here is that 4, meaning the sensing distance is 4mm.

The issue here is that these sensors are accurate to 10% of their sensing distance, meaning there could be a .4mm difference between probes. That is HUGE when printing at a .2mm layer height. This might be slightly mitigated due to the fact that the printer seems to check twice per point (though I'm not sure on that, it might just be course+fine so it can find the bed and then accurately probe,) but doesn't that mean that even 3 point leveling is going to be checking your offset more than it's actually leveling/tramming?

There seems to be a lot of hate towards 9 point leveling in this thread here: viewtopic.php?f=22&t=8974
However, a lot of posts that I see from people report that it did actually make a difference, despite the bottom line basically being "it's pointless, don't bother." Could it just be because of this accuracy problem that testing more points is able to give a better average for the offset and provide a more noticeable pattern for tilt in the bed?


I also have a quick side question. I tried enabling 9 point leveling in marlin by changing the 2 to a 3 in this line but it's still only checking 3 points.
Code: Select all
#define ACCURATE_BED_LEVELING_POINTS 2

Is there something else I have to do?
  • 0

thnikk
Waiting to connect to printer...
 
Posts: 6
Joined: 2016-Jul-Mon-12-Jul
Reputation: 1

9-Point Bed Leveling and the Accuracy of Z-Probes

Sponsor

Sponsor
 

Return to Hack talk

Who is online

Users browsing this forum: No registered users and 1 guest