After Calibration, everything looks nice and tight.
I now made a grid to controll the accuracy which messures 1300mm x 500mm, centered on the frame. When I route this on the Maslow, it has a size of 1293x500mm.
So there is a difference in the x-axis and no difference in the y-axis.
Is there a possibility to scale the x-Axis to compensate this error? It seems to be a linear issue because when i messure other grid points its allways the same factor in x direction.
Is there a possibility to scale the x-Axis to compensate this error? It seems to be a linear issue because when i messure other grid points its allways the same factor in x direction.
There is no ability to do this in the Maslow, but the maslow is very repeatable,
so you should be able to correct for it in your CAD.
I just want to make sure I understand correctly what is happening here. Do you mean that the error scales by a consistent factor as you deviate from the midpoint (e.g., halfway to the midpoint the error is 3.5mm) or that it maintains a 7mm error? If it is a consistent 7mm error, then that should be easy to compensate for in CAD, but compensating for a scaled error is likely to be much more difficult.
We’ve noticed that the X seems to be worse than the Y too. We’re trying to investigate why that might be the case, but nothing that we’ve found has given us the reason yet. We’ve found that just scaling everything by 1% in the X direction makes a big difference, but it’s a hack so I didn’t want to make it official. Until we can track down the cause of the error do you want a setting to scale everything by _%?
We’ve noticed that the X seems to be worse than the Y too. We’re trying to
investigate why that might be the case, but nothing that we’ve found has given
us the reason yet.
I would suggest doing some testing with square and portrait mode frames, see if
the problem changes.
I think it would also be good to test with small and extremely large frames
(including square) to see if this tracks with belt length
I’m remembering the problem we has with the original maslow until we opened the
gearboxes and discovered that the gear ratio the manufacturer specified was
incorrect.
If we track this to follow with belt lengths, then we can correct for it (my
suspicion is belt stretch under tension, which will require a correction factor
based on belt tension and length)
We’ve found that just scaling everything by 1% in the X
direction makes a big difference, but it’s a hack so I didn’t want to make it
official. Until we can track down the cause of the error do you want a setting
to scale everything by _%?
not a bad workaround for now. allow it to be set separately for X and Y
So I did scale the CAD Drawing and now i can get ±3mm accuracy. In between 2m in X direction in the center of the frame everything further out is not really accurate…
I had in mind to cut out boat planks, and with the frame calculator there should be accuracy for 2500mm in x direction… i hope we can improve that…
Think I am seeing this too among other issues. Tried to cut a desk with some leg pieces oriented vertically and others horizontally. The legs are different lengths.
Know I made my calibration again. I drawed some holes with a distance to each other of 850mm and messured the distance on the maslow and I get 847mm in X and Y direction which results in a scale factor of 1.0035419126. I will redo it again with the scale factor stored in on the maslow and we will see if this solves the problem. I guess there is a small accuracy failor when the maslow messures the belth lenght.
I’ve now testet the grid again, and scaled the x and y axis on the maslow by the factor 1.0035419126.
Now the x axis ist accurate and the y axis has a deviaton of 0.5mm on 400mm so pretty accurate. I now adjust the y axis and try again.
I will report soon.
It would be great, that the configuration value for /Maslow_Scale_X and Maslow_Scale_Y would be more accurate. At the moment it is only possible the set 1.001 that means, that on 1000mm that is 1 mm difference. it would be great to have 0.0001 or more, so it is possible to set 1/10mm.