Tolerance Z axe

I try to find out what the limits are , so I can take this in account when making designs.

set-up is

  • material plywood 9 mm
  • plate thickness 8,3 mm
  • set the plate thickness at 8,2 mm in Slingshot
  • drill depth 7,2 mm
  • standard bit supplied with Goliath, 6 mm

After cutting the plate, the plate thickness is 0,46 mm while I expected it would be approx 1 mm.

Is this within the tolerance of the Z axe ?

I have done some test with different depth sets 8,0 - 7,8 - 7,6 - 7,4 mm
which took quite some time because I needed to start everytime from scratch,
which all cut throug the plate. Depth setting 7,2 mm was the first depth which left a little bit of plate material.


This is probably something that you’re already aware of, but I just thought I’d put it out here in the forum.

I have found that the Z tolerance is dependent on a few things.

  1. The surface hardness of the material you’re cutting
  2. The bit diameter / sharpness

Harder materials seem to provide slightly more resistance when the bit is initially pushed down and into the material when being calibrated for the Z-Axis at the start of a job. Goliath seems to respond to this hardness/resistance quicker and pulls the bit back up effectively setting the 0 point for the Z-Axis.

Generally smaller bit diameters take less force to push into a material and so a greater time for Goliath to respond and set the 0 point for the Z-Axis.

Ex: My 1/8" (3.175mm) bit has a delta of 1mm extra depth of cut in maple veneer plywood vs. my springa supplied 6mm bit has a delta of .5mm extra depth of cut in maple veneer plywood.

My point is that both material and bit will likely cause some variance in the Z axis cut depth.

It’s tempting to take a small metal plate with a known thickness and let Goliath calibrate Z-Axis with that under it and build the metal plate thickness into your project cut depth. This idea assumes all bits would meet immediate resistance at point of contact with the metal and hopefully Goliath would calibrate their Z-Axis the same.

I have not tried this but it might get tolerances closer together, but admittedly probably not the best for sharp bits.


@Bmreddin tx for the info,

I gonna make a holder which fits around the plexi dust-cover,
mount in the middle a strip of "feeler gauge tape with a thikness of 0.1 mm.
The advantage of this kind of “tape” is that these are quit accurate of thickness.
During calibration the mill will be calibrated on the feeler gauge tape, so the mill will not be pushed in material and calibration is independable of material hardness.

It would be great if you could set via Slingshot a kind manual calibrate function.
If this function is activated, push the yellow button to execute the calibration, so could have perfect control of when the calibration needs to be done.
Or a preferences setting so you can choose if the Z axe calibration should be done automatically ( like it is now) or manually (and then you can use a calibration “help tool”)

FYI similar issue is also handled in post
Set Toollength manually or correct it



Hi dSt and Bmreddin,
what you are telling here is correct.
The Z axis accuracy depends on the first calibration, which,as you already said, is altered by the material hardness, the shape of the bit and the backlash of the Z axis slider.

As I already suggested in my reply in this topic Help with Tabs and Fill? - #11 by DavideC my suggestion is to do a test cut and measure how much is the difference between the cut depth cut and target depth, and the next time take this delta into consideration when setting the panel thickness and depth cut of each track. In the case you should have negative delta value, meaning that if you set the depth at 10 mm Goliath will cut 10.7. So the delta is -0.7mm if you need to cut at 10mm depth you should set the depth at 9.3mm

The opposite should be done with the solution proposed by dSt, put something hard under the spindle protection. This is usefull in order to eliminate variations in the zero calibration caused by material and bit shape, and have calibration of the zero be altered by tha Z axis backlash which is costant. This will lead to more costant result
But this time you will have a positive delta. meaning that if you set the depth at 10 mm ad you execute the calibration on a 1 mm thick metal plate Goliath will cut 9.5 mm. So you will have a delta of +0.5 mm and in order to cut 10 mm deep 10.5 mm should be set.

please note that the values given are for explanatory purposes only and everyone should run some tests in order to find the right values their case

One more thing that should be taken into consideration is that the position of the mill is altered by the postition wheels of Goliath. For example is Goliath is moving on an unven surface and the back wheel is on an higher spot the bit will be closer to panel surface and this will cuase to have less depth cut.

We are currently working on an improved calibration Z-axis calibration where delta values are already take into cosideration and zero calibration can be done manually.

Anyway dSt it would be helfull if you could write a topic inside Feature Request - Goliath CNC Community and describe how you would like it to be.

thanks in advance!


I am using this decorator scrape each time I zero z axis after tool change. Works fine it is about o.2mm thick.

1 Like