Jedicut for DIY cutting machine
Posté : lun. janv. 22, 2018 4:42 pm
Hallo Jerome !
hallo Boardmembers !
I studied now for a while your very interesting blog as well as the forum and learned
a lot regarding the form-cutting programme built by you. Congratulation, it's great !!
Now I started building my own cutting machine using your programme modified to my
homebrewed cutting circuit. It works but unfortunately not as satisfactory as I expected.
My basics:
- Windows-PC running Win98SE at 100 MHz (an ancient device - I know -but
suitable for doing some experiments)
- Jedicut Programme 2.3.2 (newer versions do not run properly on Win98SE)
- Communication-Modus: CncNet98.dll activated (others do not work properly)
- external timer ( I tried several cycles e.g. 2000, 3000, etc.)
- homebrewed cutting circuit with AVR-microcontroller doing step
preparation - bi-polar stepper motors
My ajdustments:
Options / Machine:
Nb mm/pas 0,026 (1,25mm/48 full steps)
Vitesse de Découpe: 50
ca. (mm/s) 1,00
Vitesse de Découpe: 20 (rapid, fast)
ca. (mm/s) 2,36
external timer 2000 cyc
My results:
"manual engine control":
slow: measured elapsed time 1,00 mm/s >> (corresponds exactly with Options/Machine)
fast: measured elapsed time 2,36 mm/s >> (corresponds exactly with Options/Machine)
But in a "circle of cut" (very simple example) there appear unexpected deviations in measured elapsed time !!
Step 1: 100 mm horizontal movement ("Horizontale") - SnapShot_3.jpg - fast
Step 2: "Extrado" of profile (depth 100 mm) - SnapShot_2.jpg - slow
Step 3: "Intrado" - " - - SnapShot12.jpg - slow
Step 4: 100 mm horizontal backward movement ("Horizontale" - RAZ Compteur) - SnapShot.jpg - fast
( For reference see enclosed sreenshots taken from running program).
Measured elapsed time:
Step 1: ca. 2,36 mm/s >> (corresponds exactly with Options/Machine)
Step 2: ca. 0,72 mm/s ???????? >> 1,0 mm/s expected ????????
Step 3: ca. 0,72 mm/s ???????? >> 1,0 mm/s expected ??????????
Step 4: ca. 1,70 mm/s ???????? >> 2,36 mm/s expected ??????????
Please advice how to overcome and solve this problem
regards / Hans-Peter
hallo Boardmembers !
I studied now for a while your very interesting blog as well as the forum and learned
a lot regarding the form-cutting programme built by you. Congratulation, it's great !!
Now I started building my own cutting machine using your programme modified to my
homebrewed cutting circuit. It works but unfortunately not as satisfactory as I expected.
My basics:
- Windows-PC running Win98SE at 100 MHz (an ancient device - I know -but
suitable for doing some experiments)
- Jedicut Programme 2.3.2 (newer versions do not run properly on Win98SE)
- Communication-Modus: CncNet98.dll activated (others do not work properly)
- external timer ( I tried several cycles e.g. 2000, 3000, etc.)
- homebrewed cutting circuit with AVR-microcontroller doing step
preparation - bi-polar stepper motors
My ajdustments:
Options / Machine:
Nb mm/pas 0,026 (1,25mm/48 full steps)
Vitesse de Découpe: 50
ca. (mm/s) 1,00
Vitesse de Découpe: 20 (rapid, fast)
ca. (mm/s) 2,36
external timer 2000 cyc
My results:
"manual engine control":
slow: measured elapsed time 1,00 mm/s >> (corresponds exactly with Options/Machine)
fast: measured elapsed time 2,36 mm/s >> (corresponds exactly with Options/Machine)
But in a "circle of cut" (very simple example) there appear unexpected deviations in measured elapsed time !!
Step 1: 100 mm horizontal movement ("Horizontale") - SnapShot_3.jpg - fast
Step 2: "Extrado" of profile (depth 100 mm) - SnapShot_2.jpg - slow
Step 3: "Intrado" - " - - SnapShot12.jpg - slow
Step 4: 100 mm horizontal backward movement ("Horizontale" - RAZ Compteur) - SnapShot.jpg - fast
( For reference see enclosed sreenshots taken from running program).
Measured elapsed time:
Step 1: ca. 2,36 mm/s >> (corresponds exactly with Options/Machine)
Step 2: ca. 0,72 mm/s ???????? >> 1,0 mm/s expected ????????
Step 3: ca. 0,72 mm/s ???????? >> 1,0 mm/s expected ??????????
Step 4: ca. 1,70 mm/s ???????? >> 2,36 mm/s expected ??????????
Please advice how to overcome and solve this problem
regards / Hans-Peter