Calibration

[Delandtree]

Hello,

Does anyone have insight into calibration. I have followed the guide updated on the site here. I cannot get past the first M360 code. I have repeated it about 20 times and it never comes close to the same position.

Any thoughts?

Aaron

[Quentin Harley]

Right! Let’s start.

Please let me know what we are working with here:
1) Stepsize of motors: 1.8 or 0.9
2) uStepping selection: 16th or 32th
3) Post the output of M503 here as well

Things you should have done before calibration:
1) Homing should function correctly, and should end up in the same position every time
2) You should have measured the lengths of your four arms (that should be the same length) and entered into configuration.h
3) You should measure the home park position relative to the bed zero point and enter that value into the configuration.h as well.
4) check that you have the axis correct. The upper wheel is Theta (X) and the lower wheel Psi (Y)

Must work…

[Delandtree]

Motors are 0.9 and I have DRV8834 drivers that are both set on HIGH on both pins. Which I believe puts it at 16.

>>>m503
SENDING:M503
echo:Steps per unit:
echo: M92 X350.00 Y350.00 Z100.00 E450.00
echo:Scaling factors:
echo: M365 X1.00 Y1.00 Z1.00
echo:Maximum feedrates (mm/s):
echo: M203 X300.00 Y300.00 Z300.00 E45.00
echo:Maximum Acceleration (mm/s2):
echo: M201 X400 Y400 Z400 E8000
echo:Acceleration: S=acceleration, T=retract acceleration
echo: M204 S9000.00 T3000.00
echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum xY jerk (mm/s), Z=maximum Z jerk (mm/s)
echo: M205 S0.00 T0.00 B20000 X10.00 Z10.00 E5.00
echo:Home offset (mm):
echo: M206 X50.50 Y0.00 Z0.00
echo:PID settings:
echo: M301 P22.20 I1.08 D114.00

[Quentin Harley]

Which electronics are you using?

If you issue M360, does it go too far, or end up short?

[Delandtree]

Mega2560
RAMPS 1.4

A4988 running the 1.8 stepper motors on the extruder and z-axis

DRV8834 running the two 0.9 stepper motors on the X/Y pulleys.

It can home perfectly after I run the M360 code it moves the arms out and are short.

The nozzle sits out near the end of the bed with the Theta are about 30 degrees from being parallel with the stand.

[Quentin Harley]

Ah, ok.

Try setting M92 X430 Y430 as per the calibration page

Morgan Start-up and Calibration Guide

I did make some updates in the past week, so I would suggest you head over there and check it out again.

If 430 does not work, increase it to get the arms moving further back, and decrease to move it more to the front. When you can reach it using the jogging, you are close enough.

Hope it helps

[Quentin Harley]

Additionally, I see M206 X is 50.5 – That is very high. Mine is below 10 (5.34) and should be in that vicinity.

Try setting it to 0 (M206 X0) again, and then issue M360. Tell me where is ends up (picture would be nice)

[Delandtree]

Hello again,

So I reloaded all of the firmware. Started out fresh, got the homing to work right away.

If I send the command: M360 with the motors set at 430 it goes crazy, I included a video.

This is my current 503 output:

SENDING:M503
echo:Steps per unit:
echo: M92 X430.00 Y430.00 Z100.00 E450.00
echo:Scaling factors:
echo: M365 X1.00 Y1.00 Z1.00
echo:Maximum feedrates (mm/s):
echo: M203 X300.00 Y300.00 Z300.00 E45.00
echo:Maximum Acceleration (mm/s2):
echo: M201 X400 Y400 Z400 E8000
echo:Acceleration: S=acceleration, T=retract acceleration
echo: M204 S9000.00 T3000.00
echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum xY jerk (mm/s), Z=maximum Z jerk (mm/s)
echo: M205 S0.00 T0.00 B20000 X10.00 Z10.00 E5.00
echo:Home offset (mm):
echo: M206 X0.00 Y0.00 Z0.00
echo:PID settings:
echo: M301 P22.20 I1.08 D114.00

If I set the motors around 100 then they can get close but I cannot always tell the motors in which way to go.

[DragoslaV]

I have exactly the same problem and I was about to make a new thread when I just realized we are both using the low-current DRV8834. Could this be the source of this problem?

I also noticed that once in a while the arms would move away from the drive shaft rather than moving in.

[Quentin Harley]

Home, and then issue

G1 X0 Y0 F1000

Where does it go?

[Quentin Harley]

PS: Delandtree, how did miss your post…

[DragoslaV]

Straight into the drive shaft, just like in the video from 0:26 to 0:30

[Quentin Harley]

Ok. That means that the homing offsets are probably incorrect in the firmware.
Did you check out and follow the calibration guide?

Also, when homing, does the thin arm home first, and then the thick one (Theta first, Psi after)?

[Quentin Harley]

So that means your home position is far off the platform to the right, correct?

And your mesured tower offset?

Is it not possible to move the magnets to a position closer to X10 Y-55?

[DragoslaV]

Actually, the nozzle is 25mm from the right, its 80mm from the top. I think my homing position is +/- on the same position as Delandtree. Also the offset from the tower is -59.

I thought about moving the magnets but I can’t since I threaded the theta arm and drive wheel onto the drive shaft. I could ofcourse remove those and glue them into place.

[DragoslaV]

Yes, I got through calibration, I changed the home position a bit and used X and Y correctly this time.

I still got to do some final tweaks but this is great after a frustrating weekend 😀

[Quentin Harley]

Great news!

[DragoslaV]

I think I might have cheered to early.

I got the DRV8834 with an 0.9 degree motor but I need around 215 steps rather than the 430.. Calibrated I even have 185?

So if I ignore those low values and continue moving the nozzle around the bed after calibration I get a nice result:
-G1 Y0 X0: top right bed
-G1 Y0 X200: bottom right bed
-G1 Y200 X200: bottom left bed
-G1 Y200 X0: top left bed

Now if I run a test print with the provided slic3r settings the entire bed surface gets streched out: i.e. a circle becomes a long ellipse and G1 Y200 X200 becomes unreachable for the arms.

I first suspected the g-codes, but couldn’t find anything init, I am a bit baffled by this.

M503
Steps per unit:
  M92 X185.29 Y185.29 Z100.00 E450.00
Scaling factors:
  M365 X1.00 Y1.00 Z1.00
Maximum feedrates (mm/s):
  M203 X300.00 Y300.00 Z300.00 E45.00
Maximum Acceleration (mm/s2):
  M201 X400 Y400 Z400 E8000
Acceleration: S=acceleration, T=retract acceleration
  M204 S9000.00 T3000.00
Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum xY jerk (mm/s),  Z=maximum Z jerk (mm/s)
  M205 S0.00 T0.00 B20000 X10.00 Z10.00 E5.00
Home offset (mm):
  M206 X-4.29 Y-5.03 Z0.00
PID settings:
   M301 P22.20 I1.08 D114.00

[Quentin Harley]

To me it seems like the microstepping settings are not correct. Check your jumpers again.

Are your two arms moving at the same speed when homing starts? It should start off by just rotating the whole arm anti-clockwise until, the first home endstop is hit.

[Quentin Harley]

Strange indeed… Could be that you have a different version of slic3r than I used to create the settings.

[Quentin Harley]

That line is supposed to prime the nozzle… Works fine on my machine.

Interesting…

[DragoslaV]

I am running the latest firmware and I’m pretty sure I got the right offset on MANUAL_X_HOME_POS, MANUAL_Y_HOME_POS and SCARA_offset_y.

In my case its X-83 and Y-25 I measured it with a vernier caliper.

Yes, when homing Theta is first and then Psi follows.

[DragoslaV]

That’s the weird thing, all jumpers are on high, both arms move at the same speed when homing. (with Theta hitting home first)

I have checked the 16 teeths per 40mm and confirmed it. but I do have an 18 teeth pulley and couldn’t really find (both here and in the magazine) what it should be. That might explain the lower steps I need.

Could it be a bad driver or a bad mega clone?? or maybe OctoPrint?

[DragoslaV]

Ok, problems are totally caused by the STL to Gcode conversion and not the fault of the printer… When I loaded the STL straight into the Pronterface it loaded the correct dimensions and start printing correctly. When I convert STL manually to G-code (with Slic3r and your Slic3r sttings) than Pronterface will give some stretched out dimensions.

i.e.
The print goes:
– from 73.34 mm to 126.66 mm in X and is 53.32 mm wide
– from 73.34 mm to 126.66 mm in Y and is 53.32 mm deep
– from 0.00 mm to 39.95 mm in Z and is 39.95 mm high

and wrong:
– from 73.38 mm to 126.62 mm in X and is 53.25 mm wide
– from 0.00 mm to 126.62 mm in Y and is 126.62 mm deep
– from 0.00 mm to 39.70 mm in Z and is 39.70 mm high

Sometimes you can overlook and stare at the simplest facts when you can’t get it working. Glad I loaded into Pronterface and I got this feedback.

Well going back to trying to get a result this time 🙂

[DragoslaV]

Ok I have found out why it was doing this. I first off have tried 2 version of Slic3r (0.9.10b and 1.1.1-dev).

The reason why I was getting stretched out dimension was because of the custom start g-code.

I have removed all lines, and manually add each line, save and see if I could reproduce this issue.

And with success, I got strange dimensions after issuing ‘G1 E10’ when I disabled this line all went correctly….

[Author]

Posts