Calibrating a vetically decentered work area

[Levi_Zurcher]

I’ve got a vertically oriented frame, but the work area is not vertically symmetrical. The bottom is about 1 foot above the lower anchor points and the top is about 2 feet below the top anchor points. When calibrating, the sled will pull off the top edge of the board and it’s not taking measurements in the lower part of the board. The calibration_grid_x and calibration_grid_y settings are within the size of the 4x8 sheet. I guess I could shrink the calibration_grid_y setting or increase the Maslow_calibration_offset_Y, but then I imagine it wouldn’t be even worse on getting measurements in the lower part of the board. Is there some way to tell it it’s de-centered?

[md8n]

@Levi_Zurcher

When calibrating, the sled will pull off the top edge of the board and it’s not taking measurements in the lower part of the board.

You can set the maximum size of the calibration grid before doing the calibration. Its default is 1000mm x 2000mm (1 metre by 2 metres). But it is always centered according to the anchors points.

I have the same situation, so for the top edge I just screw in a strip of material that’s the same thickness as the board so that the sled has something to slide along.

Also, for calibration I’d recommend measuring all the distances between the anchor points (top vs. bottom, left vs. right) and then use the maximum values for each as your starting values for your frame (actually anchor point) dimensions.

[Levi_Zurcher]

Thanks for the tips. It would be nice to be able to run calibration based on an arbitrary starting point, like just jogging it over to roughly the center of the work piece and hitting calibrate. It seems like this would be theoretically possible since the Maslow should know the length of the belts, but maybe there’s a good reason not to do that.

It wouldn’t be too hard for me to just lower the upper beam to make it symmetrical. With the old Maslow, I set the top beam higher, which improved accuracy at the expense of reducing the maximum working width to 6ft. With the Maslow 4, maybe that is no longer a concern?

[dlang]

Levi Zurcher wrote:

It wouldn’t be too hard for me to just lower the upper beam to make it symmetrical. With the old Maslow, I set the top beam higher, which improved accuracy at the expense of reducing the maximum working width to 6ft. With the Maslow 4, maybe that is no longer a concern?

The constraints are different, you can take a look at
http://lang.hm/maslow/maslow4_frame.html to play around and get a better feel
for the sweet spot of the maslow 4. It’s not that it doesn’t work outside the
green area, but that outside that area errors start to creep in that we haven’t
fully analyzed yet. you can go a bit out of the green area with no problems, but
going a long way out will cause strange things to happen.

Thanks for the tips. It would be nice to be able to run calibration based on an arbitrary starting point, like just jogging it over to roughly the center of the work piece and hitting calibrate. It seems like this would be theoretically possible since the Maslow should know the length of the belts, but maybe there’s a good reason not to do that.

I also wonder how a calibration routine would look that focuses on the extreme
edges (trying to keep in the green area of the above plot) rather than building
out from the center where you have the most (relatively speaking) errors in the
calculation.

The current approach is designed to be hands-off and not really require much in
the way of understanding about what it happening (plug-n-play)

I think a branch that let you jog around and then tell it to tighten up and take
measurements for a point would be useful for experimentation. The big problem
would be the handling of loose belts, making sure that they don’t tangle, get
caught in gears, or just spool back on incorrectly.

I also have a way for you to take manual measurements and get the anchor
positions

Maslow 4 manual configuration (onshape doc)

note that if the maslow thinks distances are X and your tape measure says they are Y, the maslow is correct (at least as far as getting good cuts are concerned) since the current calibration gives different coordinates on different runs of up to a couple of mm, that seems in the same ballpark as the errors of manually reading a tape measure, so I threw together this onshape doc that lets you enter measurements and it shows the frame and calculates the coordinates. I have you measure both diago…

David Lang

[bar]

Yeah, that makes total sense. There is a little bit of a “chicken and egg” type problem because

@Levi_Zurcher

just jogging it over to roughly the center of the work piece and hitting calibrate.

The problem is that we can’t jog safely until after calibration

It wouldn’t be too hard to make it so that the calibration process could be offset in X and Y by some value.

Generally I’d recommend just shrinking the size of the calibration grid to fit on the work area when centered. In testing we’ve found that the size of the grid doesn’t really matter that much. There is a VERY slight advantage to doing the grid larger, but really not much.