I have read through the frame design topic a number of times, but one thing I keep missing is what are the critical dimensions.
For example:
Or maybe this is taken care of in calibration…?
Let me know your thoughts.
The default settings in GC should be good to get you started with calibration. The rest is measured there.
It’s good to measure both as good as you can by hand, to have them as a reference, if the results of the calibration seem off. Exact targeted distance is not required.
The calibration measures the distance between motors and calculates the distance from the motors down to the workarea’s top edge.
This value varies, but it needs to include
On my setup, that’s in the neighborhood of 3.5 inches. My spoilboard is cheap 5mm ply. Some use 1/2" foam board or OSB, so would need more depth for that. I use 4 half-bricks for weight which make the hanging balance point (hanging at about 15 degrees) about 3.25 inches. I’m most often cutting 3mm or 5mm stock, so the 3.5 inch depth works for me. If I were going to do thicker stock, I’d need to add some depth by either moving the motor mounts or putting a thinner workarea sheet on the frame.
Edit - not sure I made it clear that my workarea sheet is a sheet of 3/4" ply, not part of the sum as it’s (in theory) part of the machine on which the items above stack. It’s that stack that determines the depth…
Yes, exactly! By ‘cutting bed’ I think you’re describing the pieces I call ‘spoil sheet’ plus ‘material to be cut’ . On my frame, I have additional plywood as part of the ‘structure’, and the ‘?’ Measurement correctly measures from the top surface of that. A very good diagram.
Yes, that’s it. . …
I was actually going to ask the same question, but instead of this dimension, I was wondering if there are optimal locations for the motors, not depth from the cutting surface, but width and height above cutting area. (Assuming a 4x8 cutting area). And if it’s been calculated how high/wide the two motor mounts can be with the stock chain length.
If I understand correctly, it doesn’t really matter were the motors are. As long as the motors are NOT too far out from the cutting surface, as to not exceed the length of the chains, or too close in that the sled cant cover the entire cutting surface. The variations in spacing is taken care of when you do your calibration. I think that the general design is something along these lines:
With a quick calculation, that would use up most of the 11’ chains. The accuracy drops towards to the top and corners of the the cutting surface from what I understand. Creating a higher and wider mount for the motors should increase accuracy, however only up to a point where it would start becoming less accurate again.
I wonder if tests have been done on where that point is? How accurate is it currently at the worst areas and the best with stock dimensions?
The default value for the motor height above the workarea is 463mm, a little more than 18".
@Steve_Hall Wonderful diagrams! Generally speaking the motors should be about 1 foot out from the side of the cutting area. The dimension is correct in your drawing, but could be misleading if someone wanted a shorter bed. The default setting is 463mm (~18.25in) for the motor height, so you’re a little low.
As far as I know, no one has run bed accuracy tests on a stock size machine. I have tested as many of the linkage kits as I’ve been able to acquire. Full thread here, I’ve set the link to load at my post when I actually started running tests. That is, in case you haven’t already seen it 
You’d be surprised with the level of accuracy you can get with the 45 degree kit. I have a short top beam, which is physically 112" long, and I have reliable accuracy across much of the bottom of the bed. I know people have also dialed their machines in better than mine. I’m sure with the 120" top beam the new stock design is going to have, that would be pretty tight tolerances across the bottom of the bed.
I’m still very curious to see how @bar’s ring will compare to the 45 degree linkage 
But I digress, this is a thread about frame dimensions, not machine accuracy.
I have read through the frame design topic a number of times, but one thing I keep missing is what are the critical dimensions.
For example:
- Distance between the two motors.
at least 10’, 12’ would probably be better (if they are too close, you can’t cut
well in the bottom corners, if they are too far apart, you have trouble in the
upper center)
- Depth of offset between the motors and the cutting bed.
the chain should be about 6" from the frame (allowing for 3/4 backer board, 1/2
wasteboard, and a 3/4 workpiece.
This is not taken care of in calibration
Yes, you want the chains to be parallel to workpiece, so the motor sprocket
and the chain attachments should be in the smae plane
but you don’t fix this by adjusting the plane at the sled, you want the chains
to attach at, or slightly above the centerof gravity on the sled (which would
make the sled top move sligtly froward when hanging it in air by the chains) and
then adjust your motor mounts to match.
as you work on thicker or thiner material, you ideally want to move your motor
mounts in and out to keep things this way, but you can go ± 1/2" without a
problem
the default width between the motors is just under 10’, the default height
above the work area is 18"
we know a little wider would be better (but you would need a little more chain)
height has not had as much attention, but if you go too much higher, you need to
go wider to maintain the same chain angles in the bottom corners.
It’s all about the chain angles, when the andles hit their extremees, the
machine is closer to the edge. We had someoenthis week have problems cutting at
the top of the area because the sled was too heavy, if the motors were higher,
he would have had less trouble.
The same person is having throuble cutting in the bottom corners because his
motors are too close together. if the motors were higher, he would have more
problems here (at the same chain angle, the motors being higher causes more area
you can’t reach)
Thank you for all of the information. I have the space so I have been thinking about building my frame to have the motor mounts higher and wider but need to figure out what those dimensions should be.
How repeatable is the inaccuracy? Are the inaccuracies on each pass consistent or do they vary? If so, is it ultimately chain sag and the lack of pull to the sides to hold the sled in place at the extreme angles the main cause of this?
The Maslow is accurate on repeating inaccuracies. Each pass will on top of the other
I tried to find the holy grail of ratio here Is it true that there is an optimum ratio between width and hight of the frame? - #49 by Gero
I’m still searching 
New motor-mounts with the chance to move them will give more insight.
Interesting. Now that may be because there is now a slot for the bit to follow which corrects further inaccuracies. Would the same be true if a pen was being used I wonder.
However if it’s repeatable accuretly then it should be fixable in software in theory.
Is there a controller for an additional motor on the Maslow board by any chance?
With the second path you would notice the offset. The repeatability is extreme high from what I’ve seen and cut. If you don’t have a chain-jump, flex on the motor-mounts or a hanging chain-mount, the positioning is accurate. Back in April 2017 I cut 15 squares all over the sheet and after each I drilled in the centre hole. Drilling 15 times, the centre hole was 6.5 mm, with a 6mm bit. That was with the fixed chain-mounts.
You underestimate how easy the bit cuts a slot, you’re suggesting that a running router, put on a vertical board would just stay put where it is if the bit is in. Even if the bit just barely touches it will cut.
There is room for an extra motor on the ic, but there isn’t a port on the latest shields.
