Calibration of Ormerod

  5 min 12 s  

The aim is to calibrate the Ormerod so that I can accurately print meshing gears for my BluRay Robot. My current version doesn’t work well enough.

I also want to explore using the stepper motor limit currents to allow probing of the extruder. I am assuming that the limiting current is proportional to the force.

Good Prints

Requires both Slic3r settings and RepRap firmware settings to agree as well as be calibration to actual hardware and to the filament.

Motor Currents and stall detection

The new duet hardware and firmware is able to detect and log motor stall currents. It also allows the motor current to be reduced by a proportional amount from the maximum that is set in the sys/config.g file as M906

Stage 1 the Extruder

1a Empty extruder

Taking off the bowden tube and zeroing the filament allows you to test the extruder length and required currents.

First try to find lower limit load

First remove the cold extrusion limit. Currently the extruder is set to a limit of 1.2A. I have found that althougth the driver can do up to 1.8 to 2A the stepper motor gets very hot. I had to use a high current with a worn gears and a bad filament path.

  1. Starting condition make sure extruder works. Bowden tube disconnnected. No load filament path.
    • Remove cold extrusion limit by M302 P1, (M302 reports current situation) I think cold extrusion limit is hardcoded to something like 160.
    • tick Test can extrude and retract 10mm. Which I can (although response not instanteous, maybe WIFI problem)
  2. Try out reducing feed current and logging stall
    • Log stalls on Extruder. Hit first snag, the extruder doesn’t appear to be a valid motor M915 E S10 F0 R1 This seems to be ok but not sure.
    • Reduce current, set extruder 0 to 5% of normal current M913 E5
    • Tried at high current and just go humming noise tick
  3. Try and read the log
    • See logging page.
    • Set logging on M929 P"extruderLog.txt" S1 and rerun extrude but only 10mm this time. Get short hum
    • Extruder log says logging has started but there is no log. cross . So the logging hasn’t worked
    • Turn loggin off and just use the hum. Also planning to change the macro files for quick testing.
    • cross gave up macro as need to take sd card out.
    • Turn logging off M929 S0 4 Test no load at 15mm/sec
    • Use binary chop of extruder current - Extruder set to 1.2A
Load Direction Result
5% Extrude cross
30% Extrude tick
17% Extrude Stuttering
23% Extrude tick
20% Extrude Stuttering
22% Extrude tick
21% Extrude Stuttering
22% Extrude tick
22% Retract tick
21% Retract tick
17% Retract Stutter
18% Retract tick

Result Extrude 22%, Retract 18%

5 Retest at 5mm/sec, 24C

Load Direction Result
17% Extrude tick
13% Extrude tick
8% Extrude Stutter barely audible
11% Extrude Stutter barely audible
12% Extrude Stutter barely audible
13% Extrude tick
11% Retract cross
12% Retract cross
13% Retract tick

Result Extrude 13%, Retract 13%

6 Retest at 60mm/sec, 24C

Load Direction Result
13% Extrude cross
28% Extrude Stutter
32% Extrude tick
30% Extrude tick
29% Extrude Stutter but not evenly
29% Retract tick
25% Retract tick
21% Retract cross
23% Retract tick
22% Retract tick

Result Extrude 29%, Retract 22%

Test effect of unwinding and path

Need longer sections of filament so will test 100mm. Can also use to measure length.
Only need to measure extrusion.
100mm at 60mm/sec at 24C

Load Direction Result
29% Extrude cross fails when needs to unwrap from PrintDry
32% Extrude Stutter
33% Extrude Stutter
35% Extrude Occasional Stutter
36% Extrude Occasional Stutter
38% Extrude tick

Length 98.7, 98.68 Diameter 1.73,1.75,1.73

Now taking path on top

Load Direction Result
38% Extrude tick

Length 98.9

Going to tweak gearing calibration

Extrusion at 15mm/sec 100mm

X axis stall

Trying lots of variations but cannot get stallguard to work.

Just checking to convert steps/second to mm/min so in the X axis
F1000 = 1000mm/min = 16.7 mm/sec
M92 for X is set at 87 steps/mm so 16.7 mm/sec = 1450 steps / sec.
For stall guard it needs to be above 200.

Trying higher sensitiviy

Sensitivity ranges from -64 to + 63, not recommended to go below -10 but going to try most sensitive to see if it will trigger at all.

cross Nice idea but I can’t get stallguard to work cross
Ooops Axes on Duet Wifi are not numbered in squence so I was using setting Stallguard on P2 that I thought was X

New features

2018-06-29 Update

Have turned off pausing on stall (M915) and have turned on logging. Also boosted X axis current from 800 to 900 mA Y axis still at 1200mA.

Nozzle is 0.4mm (I thought I had installed 0.25, however extrusion settings were set for a 0.4mm nozzle)

Going to create calibration first layer

Tests both extrusion of first layer and also correct bed height.

First sample

eg square single layer thickness and then a a quaer double layer, and 1/8th triply layer and then 5mm lever to take off surface easily.

Looking to see if get even fill

Then print at different temperatures to test adherence and quality.

First layer height is 0.4mm and subsequent are 0.2mm.

Z calibration displayed 1.4 measured 1mm +-0.3mm using cards.

2016-06-29 Too high - as bottom not all shiny (some dull patches) Under extrusion as when bend it, breaks into harp like threads ( in parts)

Can half height of square tower grip from 5mm to 2.5mm, nice to have something to hold onto. Also reduce from 3 layers to 2 layers thick.

Calibration 2

The first time I really focused on the extruder, I have since changed gears and think the problem was with the gears so probably worth revisiting. The current plan is to:

Temperature calibration

Going to use a tower block from Eibwen.

(testing python GCODE modifier)

Nozzle / filament calibration

Here is a quote that nozzle diameter may not be that important:

https://reprap.org/forum/read.php?1,113374,113953

Calibration is by single wall vase and measeuing wall thickenss.

© 2018 - 2025 · Bytestone Blog · Theme Simpleness Powered by Hugo ·