Solved Auto leveling

I've been thinking a lot about this, trying to understand the logic of it all. Especially about the inequality between the Z axis lead screw nuts, which can be adjusted with the screws that secure the limit switch holder. I've now set them to be roughly equal, to get the right side limit switch to engage sooner. But it seems to me it doesn't much matter WHEN the limit switch engages. The lead screws will turn until the nozzle touches the bed no matter where the nuts are. As long as the X axis is parallel to the bed, the auto leveling function ought to work no matter where the Z axis nut is, within reason. It seems to me that, after this, the Macro code to set the Offset (M565) is what really matters. Too big and you're printing air. Too small and your'e scraping the bed. Am I wrong?

I've watched all the demos on bed leveling and read all the posts I can find, but I still don't understand how to deal with my current problem. The right side of the bed is high, the left side is lower. In the picture, you can see the results. I think scraping the bed is what clogged my nozzle, and I really want to avoid doing that again.

How do I trim the level I now have? By turning one of the Z axis lead screws a tiny bit?

Your Macros have steps of 0.1. But I could create finer steps, right? For example, if I need an offset of 1.15, that's legitimate isn't it?

 

Thanks for that. It's the answer that made sense to me.

But why does the auto level function not take care of this? Is it because the precise degree of level required is too fine for the software to resolve, and we need to make up the last bit with shims or factional turns of the lead screws?

 

I've been tweaking the lead screws, tweaking the M565, and printing one layer of a circle for five hours. I'm not convinced I'm any closer than when I started. One you get off in the weeds, it's HELL to get back. I've seen filament streaming into thin air and I can hear the nozzle scrape the glass. And I still haven't see a first layer like I should.

The default offset values that come up in MatterControl are 0.8 to 1.2 ... a long way from -1. And when the machine was new, +1.1 worked fine. But no more.

Are you CERTAIN that negative values raise the print head? Seems to me I had an offset of 30 and the head was way up in the air.

 

M565 Z1.0 - prints in air
0.5 - right side closer but too high, left side, not sure
0.3 - Whole circle looks like it’s too close and the nozzle is dragging the glass.

 

I gave up on paper shims because they were getting crushed and changing thickness over time. The lower magnet is in a pocket, ad the long it sits there, the more the paper gets pushed into the hole and the bed drops. I think I can manage level by hand turning the lead screws, but I can't find "the zone" for M565. Are you sure more negative values are a higher print head? Becuase that's not what I'm seeing.

 

For future reference, here is what I have arrived at:

Offset test
0.5 - Printed a small square. First layer seems too thin.
0.6 - Not much different. Very thin layer 1
0.9 - The skirt is spotty, but the layer is thick.
0.8 - The skirt is spotty, but the layer is thick.
0.7 - Has a skirt, very thin first layer.

Using too much hairspray raises the level of the bed surface, and the nozzle drags though it, leaving a track that looks like the print nozzle is too low. The hairspray makes it very difficult to judge exactly what’s happening. Small test prints with nothing on the bed gives much clearer results.

No setting seems to give a first layer that is very opaque, or nearly as thick as the second .3mm layer, even though MatterControl was told to use a .4mm first layer.

Also, when trimming with the lead screws for level, one has to consider that a .3mm layer of filament corresponds to just 1/13 of a turn of the lead screw. It's very easy to manually adjust them too far and overshoot the golden zone.

 

I don't understand that, but I seem to be back on track so it's okay. My values are not negative, they're positive. Though you say to never use an actual positive number, my default Offsets were positive numbers starting at 0.8, which I confirmed with the M565 query. It currently returns:

<- zprobe_offset: 0.70

Maybe it does not care about the sign. Anyway, the larger the positive number, the farther from the bed it is.

 

M565 Z-0.6
and
M565 Z 0.6

Give me totally different actions -- one is perfect for the printer it is on and the other is way off of the bed​

I wish to understand this. If I enter M565 in the gcode terminal, it returns my positive numbers. And, the preprogrammed default values on the MatterControl (2) Macro settings were all positive numbers when I downloaded it. Why would it be that your rule of thumb is negative values, and all of my experience is positive numbers? Does it just depend on where the limit switches are arranged to trigger relative to the nozzle tip?

 

This makes total sense to me. Auto leveling on every print seems unnecessary and is a tedious overhead of time, especially on small prints. I wish I could disable auto leveling and execute it manually when I needed it. Do people ever do that on the Robo R1?

And, it seems like a better sensor, like the BLTouch would do a more consistent job. Could you not rig a BLTouch-like sensor on the carriage and attach it to the current sensor wires so that the center-traveling probe triggered the current auto leveling procedure? This would probe the bed height much more accurately than two side mounted micro switches. But, it would require a manual leveling procedure instead of an automatic one on every print.

 

I swear to you that I'm not trying to be difficult. I just want to understand.

The image below shows the default settings for the Macros that came with the MatterControl I downloaded a month ago. I've only had the (new) Robo since January 11. The numbers are all positive, and when I issue the query "M565" manually, it returns positive values. When I edit one of the macros, it shows the M565 and M500 codes, and the values are positive.

Would it not be true that the value needed for M565 would depend on the relative positions of the nozzle tip and the level at which the limit switches opened?