First of all, I wanted to thank you Maslow to making this to reality. I always wanted a CNC machine but the cost and garage space were the biggest issues. So thank you for making my dream come true. !!
That being said, this is my very first CNC machine, so please don’t judge if i ask some stupid questions
I have just put this linkage kit together and looking for some tip on calibration. One of the road block i had was setting the “Rotation Radius” ( it was somehow set to -1100mm, GC wasn’t happy with this value ). After setting the “Rotation Radius” to 260mm, I have follow through the 10 steps calibration. The last step was a little confusing, The description says “This process will repeat until the spacing is accurate to within .5mm”. However it only does one horizontal cut and after entering the value the first time, the calibration procedure is completed. (also the picture need to be updated, its still showing the Quadrillateral style sled.)
When im doing some test cut, the measurement is off is little bit, Im wondering how i can fine tune this to get a more accurate results.
first test cut: 500mm x 500mm square
Actual result: 491mm (X axis) x 498 (Y axis)mm
second test cut: 28mm diameter circle
Actual result : 24.5mm diameter ( this is way off, maybe it has problem with how i create the Gcode? )
Here is how i create the G-code:
Sketch-up → export to SVG → Makercam.com → G-code
Can anyone tell me how i can fine tune this without going through the 10 step calibration process?
Which version of linkage kit do you have? Somewhere around 260mm is appropriate for @pillageTHENburn’s 45-degree kit, somewhere around 130mm for @dlang’s top-mounted kit.
You can tweak the number yourself in the Advanced Settings panel, I think that’s what the calibration step does.
There is a fix in the next release for an error in the triangular calibration step which affects users with the top-mounted linkage. If that’s the style you’re using, either try the development branches of firmware and GC (must update both) or wait for the releases on Wednesday.
What did you measure the distance between the two cuts to be? I believe using the linkage kit (triangular kinematics) the process will repeat until you get to 1905mm +/- 0.5mm (i.e., somewhere between 1904.5 mm and 1905.5mm). If your first measurement is in this range, then it will finish.
I dont remember what i entered for the distance between the two cuts, Is there a way to look it up in the setting? but the process did not repeat. Maybe it got it right the first time?
I will reprint it when i get home and see what value i get tonight.
No way to find out after the fact… The software takes the difference between your measurement and 1905 mm and then changes the rotation radius by 90% of that value. So if you came in at 1900 mm first measurement, the delta is 5mm and it would subtract 4.5mm (90% of 5mm) from rotation radius and then do another set of cuts in a slighting different spot. Make sure you are measuring it per the directions (iirc, left side of one cut to left side of other cut).
Im looking at my rotation radius, and its still at 260mm, The same value i entered before i do the 10 steps calibration procedure. Maybe it didn’t take my measurements at all.
i will definitely reprint the marks when i get home and see if i get 1905mm.
cut a square about 6" per side and see how close to square it is. If it’s within
0.5mm, then the calibration should not have needed to do a second pass.
with triangular kinematics, you only have three variables and you should not
have to change this after the initial setup.
space between the motors
rotation radius
distance from the motors to the top of the workpiece
if these are wrong, then your shapes will be distorted (and distorted more as
you move away from the center) you can play around with the simulation mode to
see how differences between the real system dimensions and what the software
thinks the dimensions are translate into errors on the sheet.
So i have reprinted the marks yesterday and measured the distance, its pretty off. After printing it three times, it finally got 1905mm (the description in this page should mention something about the target distance is 1905mm).
After that, i have crated a 6" square. The X axis is pretty accurate now. however the Y direction is still off, about a quarter inch off.
I have couple questions here.
When i enter the distance between marks during the last step of the calibration procedure, what variable is CG changing in the setting?
Which calibration step i need to redo to get a more accurate Y axis cut?
How do i get to the simulation mode? and is there anything i can read up about this?
When i enter the distance between marks during the last step of the calibration procedure, what variable is CG changing in the setting?
Rotation Radius for Triangular Kinematics
Which calibration step i need to redo to get a more accurate Y axis cut?
Wish I knew… sorry. Is your Y direction off in measuremed length of the cut or where the cut is placed on the board you are cutting? If latter, then that’s a simple adjustment of “Motor Offset Height in MM”. If the length of the cut is wrong, I have to leave it to someone else to answer.
How do i get to the simulation mode? and is there anything i can read up about this?
WOW! this simulator is AMAZING! it really helps me understand how error will affect the actual print!
After playing with the simulator, I think i need to remeasure the motor spacing (I think this means the distance between the two motors, correct?) to get a more accurate Y direction cuts.
After playing with the simulator, I think i need to remeasure the motor spacing (I think this means the distance between the two motors, correct?) to get a more accurate Y direction cuts.
Correct. I believe, if you hand measure, it is from when both sprockets have a tooth at 12 o’clock position and its the distance between those two teeth.
You can measure between any two same points on the sprockets/chains. I don’t have a second pair of hands handy, so I hook the tape on the 9 o’clock side of the left sprocket with the chain on, and measure to a bit of chain in the same position on the right sprocket.
What the calibration is doing is tweaking the rotation radius until you get it
close enough to give you an accurate square. The exact dimensions of the square
will vary based on the size bit you use, but it should be the same vertical and
horizontal size
WOW! this simulator is AMAZING! it really helps me understand how error will affect the actual print!
yep
After playing with the simulator, I think i need to remeasure the motor
spacing (I think this means the distance between the two motors, correct?)
to get a more accurate Y direction cuts.
you do need both the motor distance and the rotation radius to be accurate.
for the calibration steps, we aren’t running through any CAM software. We just
have a g-code path to cut. As a result, there is no bit size compensation, but
that doesn’t matter since all we are looking for is if the result is square or
not.
). After setting the “Rotation Radius” to 260mm, I have follow through the 10 steps calibration. The last step was a little confusing, The description says “This process will repeat until the spacing is accurate to within .5mm”. However it only does one horizontal cut and after entering the value the first time, the calibration procedure is completed. (also the picture need to be updated, its still showing the Quadrillateral style sled.)