Help needed: Just one axis will work!?

Good evening,

many thanks for your support here. Today I found some time to solve the problem with the mirror-inverted LinuxCNC text.
But regarding the Z-axis problem I'm still not moved any mm.
What I observed is, that the tip of the inserted pencil (my tool ) is getting lower with every Z0 move compared to the Z0 position it had before.
Is that make any sense?

Alexander

Alex_HH wrote:

Good evening,

many thanks for your support here. Today I found some time to solve the problem with the mirror-inverted LinuxCNC text.
But regarding the Z-axis problem I'm still not moved any mm.
What I observed is, that the tip of the inserted pencil (my tool ) is getting lower with every Z0 move compared to the Z0 position it had before.
Is that make any sense?

Alexander

this is a very common problem , it is mostly down to having your settings too high , or microstepping to high , or an under current rated power supply
reduce your speeds by 50% and try again . then move up in small steps untill the problem comes back , then reduce by 10% or so .

one point to remember , the higher the microstepping , the lower the torque the motor can handle , many people use microstepping to gain a higher resolution , this is not what microstepping is for
and end up with the problems you mention .

Hello,

all your hints are very helpfull. Today I changed my following settings:
Step time: 5000ns
Step space: 5000ns
Dir hold: 20000
Dir setup: 20000
75% current
25% decay
1/2 step mode (Z axis), 1/16 (x,y axis)
max vel: 48.25 m/s
max acc: 762 mm/s2

into this:

Step time: 50000ns
Step space: 1000
Dir hold: 1000
Dir setup: 50000
75% current
25% deccay
1/2 step mode (z axis), 1/16 (x,y)
max vel: 25,4mm/s
max acc: 381mm/s2

With the new settings I plotted the LinuxCNC test lettering - I looks very good -z-axis is doing fine!

BTW: My system is made of: A HP Thin PC HSTNC-001L-TC, LinuxCNC 2.5. (Ubuntu an LinuxCNC are mounted on a USB stick), a 3-Axis blue board (TB6560), eShapeOko

Many thanks for your help.

Best regards
Alexander

your using a Chinese board , as Andy has said go back to
Step time: 5000ns
Step space: 5000ns
Dir hold: 20000
Dir setup: 20000

they are fine for that board ,

Hello gents,

unfortunately, I had been glad already too early. I tested the new configuration (the one I posted here) with my Dremel by milling into Roofmate material some lines in a defined depth.
I used two written G-Codes to mill six lines 5mm/2mm deep into the material.

Here are both codes for inspection:,
1. G-Code
G21 G90 G64 G40
G0 Z10.0
G0 X50 Y50
G1 F400.0 Z-5
G1 F400.0 Y100
G0 Z10.0
G0 X60 Y50
G1 F400.0 Z-5
G1 F400 Y100
G0 Z10.0
G0 X70 Y50
G1 F400.0 Z-5
G1 F400.0 Y100
G0 Z10.0
G0 X80 Y50
G1 F400.0 Z-5
G1 F400.0 Y100
G0 Z10.0
G0 X50 Y50
G1 F400.0 Z-5
G1 F400.0 X80
G0 Z10.0
G0 X50 Y100
G1 F400.0 Z-5
G1 F400.0 X80
G0 Z10.0
G0 X50 Y50
M30

2. G-Code
G21 G90 G64 G40
G0 Z10.0
G0 X100 Y50
G1 F100.0 Z-2
G1 F400.0 Y100
G0 Z10.0
G0 X110 Y50
G1 F100.0 Z-2
G1 F400 Y100
G0 Z10.0
G0 X120 Y50
G1 F100.0 Z-2
G1 F400.0 Y100
G0 Z10.0
G0 X130 Y50
G1 F100.0 Z-2
G1 F400.0 Y100
G0 Z10.0
G0 X100 Y50
G1 F100.0 Z-2
G1 F400.0 X130
G0 Z10.0
G0 X100 Y100
G1 F100.0 Z-2
G1 F400.0 X130
G0 Z10.0
G0 X100 Y50
M30

The results of milling the first G-Code is a change in the depth of lines.
The first line has a depth of 5,2mm
2. Line: 5,4mm
3. Line: 8mm
4. Line: 9mm
5. Line: 10,5mm
6. Line: 10,5mm

The results of the milling the 2. G-Code showed the following depth:
1. Line: 2,8mm
2. Line: 6mm
3. Line: 6,2mm
4. Line 6,3mm
5. Line: 6,3mm
6. Line: 6,2mm

What could be the reason for such change in Z axis during the processing of the G-Code?

I'm looking forward to your ideas and hints.

Many thanks.
Alexander