Auto-leveling command sequence

[Lofton Harris]

I could use some clarification regarding the proper order of steps/commands for auto-leveling using G32. Should I do this:
G28
G32
M500
G28
G0 Z5
(Probe for correct Z height using Pronterface controls)
M306 Z0
M500

Or this:
G28
G32
G28
G0 Z5
(Probe for correct Z height
M306 Z0
M500

I'm uncertain whether M500 after G32 is over-written by M500 after M306 Z0. Or does each M500 save different parameters? If it mattere, I'm using an AX5 mini v3.
Thanks!
Lofton

Also, can I auto-level for more than 3 locations and if so, how is this done?

[wolfmanjm]

makes no difference when you do M500.

[wolfmanjm]

3 points define a plane so there are no more needed for three point levelling.

if your bed is not flat then you may need grid levelling instead. however it is better to use a flat bed rather than use a s/w hack to try to compensate, which will always be inferior to a flat bed.

On Sunday, April 2, 2017 at 5:53:11 PM UTC-7, Lofton Harris wrote:

[Lofton Harris]

Thanks, Jim. My bed is flat but one area seems to always print a little higher than others. This doesn't change when I rotate the bed. Maybe it's a gremlin.
Lofton

[Michael S]

I could use some clarification here too. On my deltas with fsr's under the bed, I follow the instructions and everything "just works". Home to max, probe min. 0 offsets. Leveling routine does the rest. Awesome.
However I can't seem to wrap my head around the correct sequence of events and configuration values for my Cartesian FT-5.
Everything here seems counterintuitive to me on this setup.
Let me be clear, I don't think smoothieware is doing something wrong, I think I am, despite having read everything i in the forums and watching tutorials.
I think the sequence of events and overrides might help.
On this machine I home to min, the nozzle is too low,
My config applies a z-min offset. That says after homing my nozzle is too low and a g1 z0 jumps the nozzle to the correct z-min.
So far so good. There is no override file as of yet.
Now I issue a g1 z10, clear probe error, if it exists, drop the probe and leveling occurs normally
The plane is measured and comes back within tolerance and loaded, it's not at the correct height any longer because the probe triggers before the nozzle would crash into the bed.
So I set z-0 again at probe point one with paper under the nozzle at probe point one. Adjusting for the fact that prove offsets are not allowed. Save everything with m500.
In theory, I should be done as I understand it.

However the plane never seems to wind up where I expect it to be and home seems to be moving around
I want to be able to use different thickness build plates, coatings, tape as required and not have to paper test all the time.
The axis movements are not losing steps and are very repeatable when measured with a machinist hiatus.

I know the leveling with smoothie works well on my deltas. so I am doing something wrong with my Cartesian . But I really can't figure out what it is. In this setup a probe offset makes sense to me.

I know the decision to not use a z-offset for the probe was a deliberate programming choice and I am not disputing it.
I just can't seem to make it work out for me in this config.

I know how to control the bltouch it's working. It's my understanding of when and where the values are applied that could use a little help from more detailed documentation and examples on what smoothie is doing when with these values and when they are applied.

Should I be setting up the machine differently?

Right now after homing to stops, my nozzle is below the bed but off to the side.
The probe tip would be even lower were it deployed
Any g1 >0 takes the nozzle to 0 but the probe tip would still be lower than the bed

[Arthur Wolf]

 How exactly are you setting your Z 0 ? There are several methods. Have you read http://smoothieware.org/gamma_max ? Maybe try another method and it'll work better for you ?

--
You received this message because you are subscribed to the Google Groups "Smoothieware Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smoothieware-support+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Courage et bonne humeur.

[wolfmanjm]

without knowing what levellign strategy you use or the config settings, I'd say yo need to increase probe height so the probe is above the bed when deployed.

[Michael S]

I am going to attempt again using only the smoothie web interface, its possible a stray relative positioning/absolute positioning command was messing me up. 

i have another 6 smoothie conversions to do after this one, so stay tuned (only 2 cartesians the FT-5 and BoXZY). lol, but the deltas I dont have any leveling issues with since i switched to the latest edge (Many thanks Arthur, for the swift memory fix there, and the entire smoothieware team for all their efforts)

i will post a full config and override file if there is no joy after that. 

[Michael S]

just as a clarification, if there is a x/y offset from the probe to the nozzle, the final paper test is done at the probe's coordinates for test point one with the nozzle correct? not where the nozzle is at when the probe is over test point one.