Jenny

In configuration.h

If commented by default, uncomment the thermal runaway protection to activate it.

#define INVERT_X_DIR false

#define INVERT_Z_DIR true

  #ifdef Z_SAFE_HOMING
 
    #define Z_SAFE_HOMING_X_POINT (100)    // X point for Z homing when homing all axis (G28)
    #define Z_SAFE_HOMING_Y_POINT (100)    // Y point for Z homing when homing all axis (G28)
 
  #endif

  #ifdef AUTO_BED_LEVELING_GRID
 
    // set the rectangle in which to probe
    #define LEFT_PROBE_BED_POSITION 70
    #define RIGHT_PROBE_BED_POSITION 170
    #define BACK_PROBE_BED_POSITION 170
    #define FRONT_PROBE_BED_POSITION 80
 
     // set the number of grid points per dimension
     // I wouldn't see a reason to go above 3 (=9 probing points on the bed)
    #define AUTO_BED_LEVELING_GRID_POINTS 2
 
 
  #else  // not AUTO_BED_LEVELING_GRID
    // with no grid, just probe 3 arbitrary points.  A simple cross-product
    // is used to esimate the plane of the print bed
 
      #define ABL_PROBE_PT_1_X 70
      #define ABL_PROBE_PT_1_Y 80
      #define ABL_PROBE_PT_2_X 120
      #define ABL_PROBE_PT_2_Y 170
      #define ABL_PROBE_PT_3_X 170
      #define ABL_PROBE_PT_3_Y 80
 
  #endif // AUTO_BED_LEVELING_GRID

#define DEFAULT_AXIS_STEPS_PER_UNIT   {80,80,400,110}  // default steps per unit for Ultimaker
#define DEFAULT_MAX_FEEDRATE          {100, 100, 12, 120}    // (mm/sec)
#define DEFAULT_MAX_ACCELERATION      {8000,8000,200,10000}    // X, Y, Z, E maximum start speed for accelerated moves. E default values are good for Skeinforge 40+, for older versions raise them a lot.

Using the Ziflex print bed I switched the preheat PLA values (80°C preheating gives approximately 55°C on the surface of the Ziflex)

#define PLA_PREHEAT_HPB_TEMP 80

In ultralcd.cpp

if the knob encoder is moving backwards in the LCD when turned clockwise change this value :

#define ENCODER_PULSES_PER_STEP -1 //instead of 1

---

I set this one up mid June 2019 and it works OK

I was following this thread on Thingiverse, copied someone's code and tweaked it to Jenny's needs.

M140 S55 ; set bed temp don't wait
M104 S200 ; set nozzle temp don't wait
G28 X ; home x
M190 S55 ; wait for bed temp
M109 S200 ; wait for nozzle temp
G28 ; home all axes
G29 ; auto-leveling
G92 E0 ; Reset extruder
G1 X0.0 Y0.0 Z10.0 F4000 ; Go to Prime position
M109 S200 ; wait for nozzle temp
G1 E20.0 F1000.0 ; prime
G1 X100.0 Y100.0 F4000 ; Go to print area
G92 E0; Reset extruder

Possible upgrades : Prime is not necessary when using a skirt. It wastes time and requires user assist to remove primed plastic lest it clutters the print area.

12/03

Firmware flash as given by eMotion Tech. Flashed without issue.

Parameters on Repetier Host as indicated by eMotion Tech. Could not save them to custom printer. Issue should be adressed asap.

Y motor homes in the right direction. X motor does not home in the correct direction. Firmware change in configuration.h :

#define INVERT_X_DIR false

While examining firmware, thermal runaway protection was commented. Uncommented for next flash. Thermal runaway should be tested live when printer is fully functionnal.

14/03

The two Z spin in two different directions and need to be inverted on the board.

The custom motor holder v1 has several major defaults :

• the Z sensor hole should have a diameter of 12
• the M4 nut holes are not large enough to hold M4 nuts and need a little more margin. 0.25mm margin on each side should help ease them in.

The stepper drivers were set at a VRef of 680mV for each X Y Z drivers which is equivalent to a 1.7A current draw on the motor, per coil. If Marlin operates in full-step mode however, the motor will actually draw only 70% of that value approximately, and some more will be lost in heat.

The extruder stepper motor was set at 780mV as a wild guess.

The stepper motors still buzz and high pitch whistles.

Extrusion works fine. Need to check calibration.

18/03

Problem on the Z fixed by inverting one of the motor wiring. The proper wiring is : on the outer Z motor, from bottom to top, Blue/Red then Green/Black, and on the inner Z motor, the one closest to the stepper driver, Green/Black then Blue/Red.

The Z probe might not be functionning.

The length on the X Motor and X end-stop should be set when the carriage is at a high position. At the moment they will pull heavily on the system if the cables go too high.

Inverting Z value and reflashing the firmware. In configuration.h :

#define INVERT_Z_DIR true

Electronic boards were fully dismounted to solder two male pins on the RAMPS board to power the HotEnd cooler fan through the Z sensor board.

When the electronics were remounted, the Z motors were stepping in different directions again. The wiring was aligned as it was before, that is the same bottom to top color alignment on both Z motors, effectively cancelling the previous fix mentioned earlier.

Z alignment was recalibrated.

19/03

I discovered that the G92 command actually works but repetierHost does not dynamically update the information on location. You can query the data using M114. Here is a proper sequence to use G92 and then manual control (in RepetierHost):

• The axis must have been homed prior (i.e. display in black, not in red)
• Set G92 Z10 (or other)
• Use G0 Z8.9 to move only 0.1, the display will correct
• Use manual control

Debugged the problem of G28 all axis homing, it homed X, then Y, then tried to home Z using probe outside of bed area.

Reflashed firmware after changing this value in configuration.h :

  #ifdef Z_SAFE_HOMING
 
    #define Z_SAFE_HOMING_X_POINT (100)    // X point for Z homing when homing all axis (G28)
    #define Z_SAFE_HOMING_Y_POINT (100)    // Y point for Z homing when homing all axis (G28)
 
  #endif

One thing good to know : the arduino keeps the Z offset in memory when you reflash the firmware

20/03

I cut a piece of aluminium to serve as a print bed (going to provide polyimide tape to put over it).

I reset the Z probe position for aluminium sensing (<1mm offset from hot end)

I reset correct movement settings using M203 (and M503) and set them in slic3r PE : X/Y : 100mm/s max Z : 1mm/s max

acceleration : X/Y : 4800 mm/min Z : 100 mm/min

I'm almost printing but I have a problem with first layer height which is way too high.

These values have to be set in slic3r as well because they are re-set at the beginning of each Gcode.

21/03

First cube printed. Extruder is not calibrated so it's over extruded.

I had a lot of difficulty getting the correct Z offset in the system so that the first layer would actually print at the correct height.

I recalibrated the Z steps per unit from 1600 to 410.25. The steps per unit for the extruder have to be set the same way.

This change led me to change back to 12 mm/s the feedrate on Z but it's going very slowly, it could go faster.

I need a cooling system for the print and polyimide tape for the print bed

26/03

I recalibrated the extrusion at a step per unit value of 105. Seems good but it under-extruded a cube.

I still seemed to have a problem with the Z at each new print so I recalculated the Z motor steps per unit and got closer to 396. I set it to 400 which as a round value seems to be a good idea.

I found out that to calibrate the Z offset it's useful to set a G92 Z## value that is “less” than what RepetierHost seems to think it is so that you can use manual control to lower Z to 0 instead of having to input G0 Z## commands repeatedly. Caution using this method a misclick on a Z order can send the hot end into the heatbed though.

27/03

I changed default accelerations back to 8000 for X/Y and 200 for Z

I printed a cube and a low poly dino and the prints went to the end but with warping on both, due to bad adhesion to the build plate. I will change the build plate material soon which might correct this issue.

I modified the firmware values for auto-probing and set them to (in configuration.h) :

  #ifdef AUTO_BED_LEVELING_GRID
 
    // set the rectangle in which to probe
    #define LEFT_PROBE_BED_POSITION 70
    #define RIGHT_PROBE_BED_POSITION 170
    #define BACK_PROBE_BED_POSITION 170
    #define FRONT_PROBE_BED_POSITION 80
 
     // set the number of grid points per dimension
     // I wouldn't see a reason to go above 3 (=9 probing points on the bed)
    #define AUTO_BED_LEVELING_GRID_POINTS 2
 
 
  #else  // not AUTO_BED_LEVELING_GRID
    // with no grid, just probe 3 arbitrary points.  A simple cross-product
    // is used to esimate the plane of the print bed
 
      #define ABL_PROBE_PT_1_X 70
      #define ABL_PROBE_PT_1_Y 80
      #define ABL_PROBE_PT_2_X 120
      #define ABL_PROBE_PT_2_Y 170
      #define ABL_PROBE_PT_3_X 170
      #define ABL_PROBE_PT_3_Y 80
 
  #endif // AUTO_BED_LEVELING_GRID

I also changed :

  #define Z_RAISE_BEFORE_HOMING 5       // (in mm) Raise Z before homing (G28) for Probe Clearance.

and :

#define DEFAULT_AXIS_STEPS_PER_UNIT   {80,80,400,110}  // default steps per unit for Ultimaker
#define DEFAULT_MAX_FEEDRATE          {100, 100, 12, 120}    // (mm/sec)
#define DEFAULT_MAX_ACCELERATION      {8000,8000,200,10000}    // X, Y, Z, E maximum start speed for accelerated moves. E default values are good for Skeinforge 40+, for older versions raise them a lot.

I need to find a way to quiet the extruder fan on command… M106/M107 controls the print fan

If I still see uneven extrusion I can try to modify the maximum feedrate of E and the acceleration of E. And I can adjust the steps/unit if needed

03/04

I launched another low-poly print to test the printer and still had some difficulties getting it at the proper Z height for the first layer. After fiddling a while, resetting the Z-offset in the firmware and adding a Z-offset value inside the slicer the print launched rather successfully it's first layer.

All other layers after that seemed at a proper Z difference from the previous layers so the Z steps/unit value seems to be OK.

The result of the print is mitigated. Some parts are really proper and smooth, but some other parts, particularly in the overhangs are completely messed up. Also, there was a lot of warping during the print. This is most probably due to improper cooling since the print cooling is still done with the fan just zip-tied @ approx. 40mm away from the print zone with nothing to focus the airflow directly on the print zone.

I'm modelling another X-carriage and print fan blower to fix that problem. It will be based on a radial fan instead of a 40mm classic one.

The print bed is still laden with white masking tape which isn't ideal as a print surface. Kapton tape should be provided.

16/04

I changed the print bed and finished modeling a new vent and carriage system based on a 50mm radial fan. This was installed and a new calibration is necessary.

17/04

I recalibrated the Z-probe and wrote a procedure of Z calibration.

I encountered some major difficulties printing. The first layer was never on the bed. I tried many different values of Z offset going all the way to -1.25 offset in Slicer but to no success.

Also, I reset many times the offset in the firmware value and found out that if the Z offset value is set to 0 in Slic3r the G-code will being at an effective height of 0.2 from a 0 point. The thing is, when it runs the G28 homing procedure the head is left at a distance of Z something from the print bed which is not compensated during print.

18/04

I finally got the first decent print out of Jenny.

I recalibrated her Z sensor and accidentaly put a dint in the new buildtak print bed in the process.

The inductive sensor needs to be very close to the nozzle bottom to detect properly. I could get same results with less hassle by swapping for a capacitive sensor.

After resetting the Z sensor probe properly and lowering it to an appropriate spot and recalibrating the Z offset I sent a successful Gengar print at full scale that turned out quite nice.

19/04

I updated repetierHost, Marlin and this wiki page with the proper bed shape settings.

I also added a -0.15mm Z offset in slic3r to get a proper first layer.

I sent a 4 point auto-level G29 command and saved it to memory.

25/04

I discovered the reason why the first layers were too thin and pressed hard against the buildtak build plate.

The Z offset set in memory is set using the probe value of the inductive probe. That value is a distance of detection of the aluminium plate underneath the buildtak. The buildtak's thickness is approximately 0.35mm thick as measured with a caliper.

I reset the Z offset value from -0.40 to -0.15 to compensate. I'm still leaving a 0.1mm delta so the print will be well pressed against the buildtak.

I haven't printed anything yet to test these new values but that will be done first thing next morning.

I need to change that inductive sensor for a capacitive one, and if I can couple it with a new buildtak glued on a heat-resistant flexible material resting on the aluminium buildplate I should be fine.

Buildtak adhesion seems to work so well I could even test printing on a cold bed.

20/05

I swapped the inductive Z-probe for a capacitive one. What a change ! It is so much better, detects any type of material at a greater distance.

I also reprinted in PolyMax PLA my custom carriage in a v2 version along with the fan duct. Mounting it is a little tricky but with a little experience you get the correct gestures in the correct order.

I need to recalibrate on the regular buildtak to be sure I have the correct procedure but I should be good from now on with Z endstop.

25/07

I swapped the previous LCD screen for a full graphic LCD screen.

I had to reflash firmware with the following changes :

In configuration.h : comment

//#define REPRAP_DISCOUNT_SMART_CONTROLLER

and uncomment

#define REPRAP_DISCOUNT_FULL_GRAPHIC_SMART_CONTROLLER

You have to install the ug8lib library from the library manager and if you get the following message at compile :

utility.com/u8g.h: No such file or directory

You need to replace the following line

#include <utility/u8g.h>

for this one :

#include <clib/u8g.h>

• M851 Z0 : Cancel any previous Z offset.
• Heat up the hot end and the bed to nominal values (to take in account any thermal deformation of the hot end and build plate)
• Manually bring the hot end to a paper's thickness from the bed. This is the tricky part. It can be done using the following commands :
  • Home X, Y, Z (G28)
  • Raise Z 10mm
  • G92 Z20 to make the printer think it's higher than it is
  • G0 Z10 to bring it back down 10mm (it will be a few mm from the bed)
  • Lower 0.1mm at a time until you get the paper-grab (G0 Z9.9 - check - G0 Z9.8 - check - etc.)
• G92 Z0 at paper-grab point
• Raise Z 10mm (G0 Z10)
• G30 to probe and get the Z offset value of the distance between your probe and the 0 point of the nozzle.
• M851 Z-X.X (X.X is the offset value from previous step)
• M500 to save current configuration
• G29 can now be run to get an auto-leveling grid
• M500 again to save current configuration

Concerning M851/M114 in RepetierHost

When homing X/Y/Z in RepetierHost the position indicator might indicate 0 on all positions when the reality is probably X/Y at Z home position (100/100 on Jenny for example) and Z at last registered M851 position.

Using this, one could set M851 to a larger (negative) value than the previous one to use manual jog on Z to set the new one, instead of setting M851 to 0 as indicated above and then having to manually input G0 codes to get Z to paper-grab height.

Warning ! If using this method be very careful on where you click on the Z jog for you are tricking the printer's knowledge of where the print-bed is and you can send it straight into it.