Z homing Gcode bloody for beginners

Ted

@Phaedrux Thanks for the detailed explaination! I will try it out soon! It really cleared things up for me.

I also found a missing hint in the documentation that got me hard: In one page it says for the Z trigger high to insert the measured value negated to the G31 command and on the other page there is no hint on any negation via G31.

Phaedrux

@bearer I think it's most commonly and originally used on deltas, but there doesn't seem to be any limitation on its use to only deltas.

@dc42 The table in the gcode wiki does say on delta (only), is that accurate?
https://duet3d.dozuki.com/Wiki/Gcode#Section_G0_G1_Move

Phaedrux

@ted That's a good catch on the negated value thing. I don't think it should say that. The value will be negative in cases where the nozzle makes contact with the bed (like with a piezo, or maybe microswitch) in which case the nozzle tip is below the bed plane slightly. Otherwise it should be positive, indicating the height above the bed that the nozzle is when the probe triggers.

I'll change that wording in the mesh compensation article. I think it may say negative there because it's borrowed from the Smart Effector documentation, which uses a strain gauge as a nozzle contact probe, which would have a negative value.

Ted

I got sidetracked by the mesh bed compensation while trying to find out where on my bed I should probe for finding z=0. I could get mesh bed compensation working but now I wonder how I could combine it with probing for z=0. Could not find any information about this.

MBC finds out the least square error of all probe points on z axis and uses it to correct the actual z=0 position transparently (if I'm not mistaken). I could not find any more information.

Should MBC run after z=0 probing?
Should it run before z=0 probing?
Does bed mesh compensation include setting absolute z position so z=0 probing is not necessary?

Imaging a bed mesh compensation made using 9 measurement points while the center measurement point is used for probing z=0 (actually: current-z-height=z-probe-trigger-height):

If you first probe for z=0 and then apply MBC the transparent correction would lead to very different height values per measure point depending on whether your z=0 probing upfront happens on a measure point with negative error or positive error: The error of the measurement point used for finding z=0 gets applied to all points. In that situation I would have to introduce another static correction offset that compensates for the error of the measurement point used for probing z=0. This static offset breaks when having a fluctuating bed surface, e.g. when printing with cold, warm and hot bed due to different bed deformation depending on temperature. That would be very bad as I want to use the probe to eliminate ANY manual z-correction step per print.

First doing the MBC instead could eliminate the problem when the correction is taken into account when probing for z=0. Is it?
If it is not taken into account we have the same situation as when first probing, than doing the MBC: The found relative error of the probing point is introduced to all points uniformly.

If the z=0 probing includes the MBC z-correction for the measurement point or when both is done in the same step, the algorithm could remove the error of the point.

Did I understood something wrong? Hope you can shed any light on this!

Phaedrux

You should probe for Z0 at bed center, or rather the center point between your mounting screws.

Are you use mesh compensation in a few ways. 1 it can show you what your bed surface looks like which can be useful to get a visual representation of how flat it is. Second, it can be used to compensate for a flat bed that is out of level slightly. It can also compensate for a bed that is convex or concave, or a gantry that is arched or sagging.

So at the very least it's a good diagnostic to see what your bed is like, then you can decide if and how you should apply correction.

If it's simply flat but a little tilted, you can use a sparse grid pattern with 9 points to define a plane.

If it's curved in anyway, you can do a more detailed grid. If the bed is curved, you can tapper it off after a few mm. If the gantry is sagging, you should not tapper it off since it will persist the entire print.

If your bed is very stable and doesn't change much you can save the heightmap and load it before the print to apply the correction. Or if your bed is less stable and changes shape you can run G29 before each print. Not ideal if you're using many points.

In your case, I think the Cetus bed is pretty stable and it's not very big and it's probably pretty flat. So you could either run a detailed G29 and save it, or run a smaller grid before each print.

Before you run the G29 you should try and make the bed as mechanically level as possible and you should have established Z0 in some way (ie, homed the printer). This is so that the heightmap has a reference for where Z0 is and the topography is either above or below that point. That way when the mesh compensation is active, it should move the print head up or down in relation to the bed in real time so that Z0 keeps the nozzle touching the bed at all X and Y points along the surface of the bed.

Does that alleviate your confusion at all, or make it worse?

Ted

@phaedrux Thanks a lot for the detailed explanation. It really cleared things up I was confused about.

What I want is:

Automatic z=0 probing: So I don't have to worry about meanwhile changed bed height due to mechanical problems (lokbuild thickness changed due to mechanic impacts) or slight changes to the hotend mount (I often replace it with a laser cutter toolhead)
Automatic bed mesh compensation: To also compensate for a meanwhile changed bed level (e.g. my bed corners come up when heated and I don't always print with heatbed on).

What I read from your text is that I first need to find z=0 and then run the MBC if I want an up to date z=0 and flatness compensation. Since needed time for MBC is nothing compared to the actual print time I want to do this at start of every print.

What I still don't understand is how the MBC is working in detail: Where will it position the previously measured z=0 when applying error compensation on z axis?
Somwhere I read about a least square error based bed mesh compensation but can't find it anymore. I got confused by that as it brings up the question if it adjusts the z=0 somehow for when the point we used for finding z=0 has an square error > 0. If not it would then just add an error to all points depending on the scale of the error of that measurement point. This would make the compensation obsolete as every point on the mesh map will inflict in this error.

So did I understand it correctly that MBC just finds a relative built plate z-offset map that is applied to all points respectively without any square error thing? It does NOT contain any additional z=0 setting, just an offset is applied?

Then I don't understand how the MBC knows that it does NOT need to add the offset to the centered point. Imaging that we have the following process:

Use the center point A to find z=0 with G30
Use G29 to build and store a correction mesh map
All correction offset values are applied respectively to the corresponding point x/y point on the build plate when printing
For the next print you find z=0 again with G30 using a different measurement point B (which is lower on z than center A)
You load the existing correction mesh map again with G29 S1 (bed did not change, so mesh is still correct)
All correction offset values are applied respectively to the corresponding point x/y point on the build plate when printing

The result: If the two measurement points have different correction offset values due to different real world heights of the built plate (we assumed that B is lower than A) the actually corrected z position for measurement points is different for when printing leveled at point A compared to when leveled at point B (using the same map).
This is because when probing z=0, using either point A or B, we do not store the information where on the build plate it was measured. In fact the first print (leveled with point A) is higher than the second (leveled with the lower point B) uniformly over the whole bed.

Of course using two different points is just the extreme situation constructed to maximise error for illustration. The problem obviously occurs for a single point, too, as the algorithm that applies the correction offsets to z axis while printing does not know on which point (and what error that point had) we used to find z=0.

My conclusion is that the MBC algorithm can only work if it actually sets where Z=0 is using its own measured z-values.

What am I thinking wrong?

Ted

Another linked question:

G30 does probe for z=0 (actually correctly applying the z-trigger-height of the probe). It does that by driving down about 10mm (lets call it the probing window). If the sensor (I'm using BLTouch) does not trigger it brings up an error and does not change z=0.

My problem is that I can't ensure mechanically that my hardcoded starting z height is not further away than the 10mm probing window. That already lead to the situation that the probing stopped with the error 1mm above the build plate as it didn't reach the probing distance. In other words: The probing window was not big enough.

I could also thing of the opposite where the build plate is higher than expected so that it does not start probing with the necessary minimum distance to the plate.

Is it somehow possible to tell G30 command that it has a probing window of 180 mm (which is my z build volume height)?
An alternative would be to somehow use the BLTouch as endstop and move down with a relative G1 Z-200 S1. Is that possible?

aidar

@ted said in Z homing Gcode bloody for beginners:

Is it somehow possible to tell G30 command that it has a probing window of 180 mm (which is my z build volume height)?

M558 parameter H does this. Another question is, do you really need this.
https://duet3d.dozuki.com/Wiki/Gcode#Section_M558_Set_Z_probe_type

Phaedrux

@ted said in Z homing Gcode bloody for beginners:

lets call it the probing window

Let's call it the dive height instead.

Usually, the way this is handled is that the dive height is set at some minimum amount so that it doesn't take forever to run a G29 mesh compensation routine because it has to raise the probe to the dive height for each point, possibly multiple times. For the BLTouch 5mm is a good value. But really it only needs to be slightly more than the measured trigger height.

But the dive height doesn't set a minimum travel distance for G30. If your bed is far from the nozzle and the Z axis isn't homed, and you home the printer, it will move the z axis until the probe triggers, not just the dive height and then stop. Just like for a homing move using an endstop. After that first trigger, if it needs to probe again it will then travel to the dive height and probe again.

To ensure that the probe is a safe distance away from bed after you home the printer it can be a good idea to raise the print head after homing is finished. This could be the same as the dive height, but really it just needs to be enough distance to allow the BLTouch pin to drop freely without hitting the bed.

Can you better describe the situation where you're getting the error of the probe not triggering? How far away is it from the bed? During what operation is this happening? What are your actual settings?

Ted

@aidar Thanks, that solves the problem for me!

@phaedrux I set the dive height to 50 mm. It gives me a lot of failure tolerance. It is working fine for me. So far no problems and much safer than just the 5 mm default. The mentioned error while probing (which I now avoid with a higher dive height) was due to mechanical tolerances in my extruder / laser mount.

It is working fine for single probes using G30. Wasn't aware that this would affect the dive height of G29, too. Haven't tried G29 in a while now as I still don't know if it resets z=0 when called AFTER G30.
I got the problem that whenever G29 after G30 the printer prints with different z=0 heights. It is always a way too high and it is different every sequence! Trigger height of BLTouch probe is wonderful stable with < 0.01 mm tolerance, which is amazing! Also checked if I have missing steps and lowered max z speed to 1 mm/s. It's fine. Still looking for the error and still thinking about my question above: When G29 is not setting z=0 than this could cause the changing z=0 height. Need to investigate here.

BTW: When having dive height at 50 mm and then G30 down and get the probe triggered already after 1 mm it is not driving back 1 mm but driving back whole 50 mm. So it ends at 50 mm over the trigger point while started at 1 mm over the trigger point. Not sure if this is not a bug. When first setting the dive height to 180 mm (my build volume) I got the problem that it rises even over the endpoint switch. Since G29 seems to use that dive height also it is not a good idea to do so anyway (unless I have too much time).