Rostock not printing to scale

My-almost-original-design Rostock is about to be finished. After first tests I remarked that the scale of the objects is far off.
The 20 x 20 x 20 mm calibration box only is 11.5mm x 11.5 mm and the layers are very crumbled.
Where should I start to correct this ?

I have 2.5 belts, 20 teeth pulleys, 1/16 microstepping which should give me a 64 steps per mm. (Correct?)

I have read about the diagonal_rod calibration on http://minow.blogspot.de :

Quote

If the length of the object is wrong, adjust DELTA_DIAGONAL_ROD:

new DELTA_DIAGONAL_ROD =
100 / measured_length * original DELTA_DIAGONAL_ROD

then re-upload the firmware, and print and measure again. When the printed object measures 100 mm, the printer is printing accurately to size and is fully calibrated.

But in my case that would make a change from 285 to roughly 450mm which is way off reality.

Also I tried to scale up the objects in Repetier by changing the scale factor for each axis from 1 to 1.8, but it would not make any changes in the result.

Now where should I start to fix this issue, so the scale is correct?

---

Blogs:
Meine 3D Druck Abenteuer
[3dptb.blogspot.de]
FLSUN Delta Drucker für Deutschland
[flsun-deutschland.blogspot.com]
Books on 3D patents:
[goo.gl] (english)
[www.amazon.de] (deutsch)

Check again that your belt is GT2.5 (is pitch of the teeth on the belt 2.5 mm?), your pulley has 20 teeth, and 1/16 microstepping is really selected. If this is correct then yes, your steps/mm should be 64.

Not sure about Repetier, but steps/mm can be overwriten from EEPROM. Check the output of M503 (print EEPROM parameters) and specifically the parameters for M92 (set axis_steps_per_unit). Do you have there your number (64).

Home your printer and ask it to move platform down by e.g. 10 cm (no X/Y movement only movement along Z axe). Did carriages move down by 10 cm? If not then, you still have incorrect steps/mm for your tower motors.

Check the gcode file from slicer whether the lines have the correct length. This should rule out that scaling in slicer is not active.

You can easily measure your printer geometry with about ±2 mm precision. If you do and have it specified with this precision in firmware then it cannot lead to 42.5% error in the printed object. The difference is too huge.

The equation:

new DELTA_DIAGONAL_ROD = 100 / measured_length * original DELTA_DIAGONAL_ROD

can be used only in an iterative proces when trying to find out the right diagonal rod length provided the other firmware parameters are correct. The actual dependence between object size and the diagonal rod length is non-linear.

Quote
hercek
Check again that your belt is GT2.5 (is pitch of the teeth on the belt 2.5 mm?), your pulley has 20 teeth, and 1/16 microstepping is really selected. If this is correct then yes, your steps/mm should be 64.

The pitch seems to be 2 mm, so I changed that. Then I drove all axis home and measured the distance to the printing bed.
I send all axis to the bedlevel (g1 z0 ) and measured the distance to the bedlevel. It was around 100 mm above it. So I tweaked the axis to 81 and now everthing works perfekt.

---

Blogs:
Meine 3D Druck Abenteuer
[3dptb.blogspot.de]
FLSUN Delta Drucker für Deutschland
[flsun-deutschland.blogspot.com]
Books on 3D patents:
[goo.gl] (english)
[www.amazon.de] (deutsch)