Calibrating X,Y,Z axes

Hi all,

I used Pronterface to run the X carriage 100 mm to calibrate
and I measured a distance of about 104mm,

I am still a bit unsure of how to calculate the number to be entered into
the Sprinter file, I have read up on it but I'm still not clear

should I change this part?

"// X, Y, Z, E steps per unit - Metric Prusa Mendel with Wade extruder:
float axis_steps_per_unit[] = {80, 80, 3200/1.25,700};"

or am I looking for " XAxisScale measurement" in Sprinter?

All the Best
Simon

Check the belts are tight enough then measure it again. You need to be exact.

You shouldn't really need to change the X,Y or Z settings without your using non standard belts/pulleys, different microstepping or different thread on the rods.

If its covering less distance then it should first check the stepper motor drivers.

As for calibrating E this blog post has a brilliant explanation.

---

My Reprap blog

jds-reprap.blogspot.com

FInd the current X carriage setting in Sprinter, lets say it is set at 505 which is a complete guess as I don't know what it is off the top of my head.

so you have told it to move 100mm but it moves 104mm so do the following calculation 505*100/104 which would give you 485.576.

You would set this and upload it then test again. the only number that stays the same in the equation is the no 100 as that is what you are telling it to move

---

__________________________________________________________________________
Experimenting in 3D in New Zealand

I realize this is an old topic, but would the same work for the Marlin firmware? My X and Y values are both at 80, but my circles turn out slightly oval every time. The only thing I haven't tried is changing that value.

Is this formula correct for Marlin?
New setting = current setting*distance moved in host/actual distance moved


NelsonRap Wrote:
-------------------------------------------------------
> FInd the current X carriage setting in Sprinter,
> lets say it is set at 505 which is a complete
> guess as I don't know what it is off the top of my
> head.
>
> so you have told it to move 100mm but it moves
> 104mm so do the following calculation 505*100/104
> which would give you 485.576.
>
> You would set this and upload it then test again.
> the only number that stays the same in the
> equation is the no 100 as that is what you are
> telling it to move

---

--

Charles S.
Software Engineer
Prusa Mendel I2, RAMPS 1.4, Marlin 1.0 R2, Pronterface, Slic3r

Lodorenos Wrote:
-------------------------------------------------------
> I realize this is an old topic, but would the same
> work for the Marlin firmware? My X and Y values
> are both at 80, but my circles turn out slightly
> oval every time. The only thing I haven't tried is
> changing that value.
>

Is you oval shape orthagonal to the X/Y axis? If so, you need to calibrate those values. 80 should be close, but belt tension and tooth spacing may change it slightly in the method you outline.

If the oval shape is on the diagonal then you have backlash and need to resolve that before changing the firmware multipliers.

-Rob A>

Lodorenos Wrote:
-------------------------------------------------------
> I realize this is an old topic, but would the same
> work for the Marlin firmware? My X and Y values
> are both at 80, but my circles turn out slightly
> oval every time. The only thing I haven't tried is
> changing that value.
>
> Is this formula correct for Marlin?
> New setting = current setting*distance moved in
> host/actual distance moved
>
>

The X, Y and Z firmware values should be formulaic - based on the belt pitch, pulley teeth, etc. You should be able to compute the EXACT value you need. The extruder is the only "axis" that should require the iterative approach you describe, and only because of the variability of the hobbed bolt diameter. If the computed values for the X and Y axes result in ovals, then I would look into backlash as suggested by Rob A.