How do we set the Z offsets to get better calibration?

Troubleshooting
,
1

@bar

I’m struggling to understand the Z axis values in the maslow.yaml file are calculated / derived and want to confirm I have the right idea in my head now.

Previously the yaml file included a comment with the default values:

# Z axis values 
# These define the height of the anchor points in relation to each of the arms. You do not need to change these typically
Maslow_tlZ: 100
Maslow_trZ: 56
Maslow_blZ: 34
Maslow_brZ: 78

Does this mean that, for example, the belt attached to the bottom left anchor (blZ) is assumed to be 34mm below where the belt exits the corresponding arm in the Maslow device?

It seems that these numbers are calculated as the height in millimetres of the bottom edge of each belt above the base of the maslow sled (eg 39mm for blZ) minus a 5mm allowance as the belt end fitting holds the belt 5mm above the floor in the anchor.

39 - 5 = 34

So if I have anchors that hold the belt end fittings 16mm higher than the floor then my blZ value should be

39 - 5 - 16 = 18

Of course this is only the starting value, and when I add a spoil board (20mm) and some plywood (9mm) then their thicknesses get added to this value (in firmware pre v1.09)

39 - 5 - 16 + 20 + 9 = 47

and this 47 is what should be in the yaml (for blZ) before I run any calibration.

@mgandi gave us a good diagram of the components of the z-axis in their post Trouble during Z travel - #3 by mgandi

image
image723×231 7.44 KB

In this diagram the belt shown would be the Top Right belt (second highest arm is trZ)

With Firmware 1.09+ there are now values for Maslow_spoilboardThickness and Maslow_workThickness in the maslow.yaml file, and the default values for the Z axis offsets are now called tlZ, trZ, blZ, brZ and found in the new MaslowKinematics section in the yaml

I’m wondering how these new thickness settings interact with with the z axis offsets settings.

Should we be calculating the z axis offsets without including the workpiece or spoil board now?

Is it ok to calibrate / find anchor point locations with the spoil board and workpiece in place, so long as we enter their thicknesses in the yaml file before calibration?

Thanks
Andrew

2

This is exactly correct :+1: one addition is that all of these measurements are done with the z-axis lowered all the way down.

This sounds right to me

These new settings function exactly the same way and replace the old settings. I think that it is possible to update and end up with BOTH settings in your .yaml file, but only the new ones are used. It might be worth starting fresh with a new .yaml file

The big difference is that now that spoil board and work thickness are broken out you don’t have to include them in your calculations because that is done automatically by the math.

I haven’t had a chance to play with them much to figure out how much of an impact changing those has so I would love feedback.

Exactly :smiley:

Yes, and in fact I haven’t even tried entering values for those to see how much of a difference they make. I’ve been leaving them as 0 which I know is wrong…I just haven’t had the time to do a side by side comparison

3

Andrew Matthews wrote:

your understanding of how the Z offsets is correct. the default offsets are
assuming that the anchors are at the same height as the bottom of the sled

I¢m wondering how these new thickness settings interact with with the z axis offsets settings.

they get added to the Z offsets

so if the Z offset is 34 and the spilboard is 1" foam (25mm) and the workpiece
is 18mm (3/4 plywood) they add together to 77mm and would be the same as
entering 77 as the z offsets and 0 for the thicknesses.

Should we be calculating the z axis offsets without including the workpiece or spoil board now?

yes, the idea is to make it easier to adjust for different spoilboards and
workpiece sizes

Is it ok to calibrate / find anchor point locations with the spoil board and
workpiece in place, so long as we enter their thicknesses in the yaml file
before calibration?

yes.

David Lang