Had an idea on how to possibly fix some of the accuracy issues at the lower corners, thought I’d share it and see what you think.
The right and left motor chains attach to the bottom of the pivoting rails. A third motor moves the sled on the rails. Maybe an encoder on the pivot could measure the angle to improve accuracy.
that could work, but one problem you would have is the length of the pivoting
piece, how long does it need to be to reach the bottom corners and what does
that mean about the total height of the machine when it’s in the center.
the problem would be how to model the system (accounting for chain sag, which
you can never completely eliminate)
But if you could make this work, we could pull the motors in on the sides,
making the machine narrower.
thinking about this a bit more. It has the potential to do interesting things.
flip this so it’s fixed on the sled and has the sliding part in the top center
(just high enough above the workpiece to clear the sled, possibly notching the
sled.
This would let you pull the corner motors in close to the edges of the workpiece
Fix it to the sled and we know know the tilt of the sled, so we don’t have to
use a ring or linkages, we could go back to just brackets. Pivot points are
still important, so I would do different brackets than we did before.
if we have some sort of encoder, we can measure the length of the middle bar,
which could give us a significant math improvement. We can use a motor-gearbox
combination that can be backdriven so if the various lengths don’t exactly
match, it will slip without breaking (and the encoder will tell us where it
really is)
The drawbacks:
The center bar needs to be able to push (i.e. not a chain) and needs to be ~6 ft
fully extended
It may be possible to make the center thing able to be shorter.
There are straightforward ways that you could have it telescope to 1/2 the length
@bar @madgrizzle you did research into tape measure type material instead of chain. is there a wider/heavier duty version available that could push with several pounds of force? could a tape measure be backed by a telescoping tube so that it doesn’t have the ability to bend far enough to collapse?
by the way, measuring the angle isn’t going to work well, when you get out 4-6 ft, the precision that you would need to measure the angle at is too expensive to do.
But it may be possible to use an angle measurement as a homing tool, if you don’t know the chain lengths, but roughly know the angle of the center (and know the length of the center), you can figure out how to move the chains to raise and center the sled until you hit a microswitch for the center length and rotate it to a known center angle (I can even think of some microswitch based approches to get a precise position)
I’m going to be driving about 8 hours today, but I need to spend some time trying to draft out this mechanism.
And when I saw the title, I thought this was another “put a motor below to pull down” suggestion and was going to give my standard “this is why this frequent suggestion won’t work” answer, congratulations on a new twist on a few ideas. We’ve had people suggest rigid rods instead of the chains, but they end up FAR too long when pushed from the edges, but something near the center has potential.
one other possibility from this approach, if you have the rails pivot on one
end, can you have the other end ride on rails itself and eliminate the sled
(getting true 3d capabilities)
I think to answer that, I need to understand better what you are asking. But I looked at tape measure tapes, but opted for webbing to use with external encoders. Now trying out HTD5 belts.
As drawn the orange area is 4’x8’ and the rods are 9’
Maybe could make them a little shorter by moving the pivot closer to the work area.
Rather than use two motors, what about one motor and a counterweight/spring on the other side?
you need two motors in the top corners to pull, and you need the center motor to
push. you can’t just add weight to the center, and you can’t pull down from one
place (you would end up pulling away from the corners)
pushing from the center helps in the corners, but you still need to two top
motors to do the main part of the positioning, the center motor is just
providing a supplement to gravity to prevent chain sag counter the problem of
too steep of an angle in bottom corners (and I think it does it so well that we
can narrow the machine and not need a 12’ top beam)
what I am thinking is that with a extra wide tape measure type material, we could have the spool of material at the top pivot point, with a motor pushing it out against the sled/chains.
if it’s stiff enough, that could be the entire linkage (fixed attachment to the sled so that the sled rotates based on the angle). If that can’t handle the side loading, we could have a telescoping something backing the tape that handles the side loads and provides something to back the tape and make it less likely that the tap will fold in it’s narrow direction
the pivot point needs to clear the sled, so at least 9" above the workpiece (figure 12" to allow for real world dimensions)
If the rods are a fixed length, then you have two options (each with their own problems)
Pivot point at the top
This requires more clearance around the machines
9’ rods would mean the machine is 10’ tall and needs 18’ of clearance out the sides
** rods fixed to the sled**
*needs vertical clearance of rod height + workpiece height + a bit
*when the sled is high on the machine, the leverage of the rods above the pivot point could be higher than the leverage of the sled, causing the sled to pivot up
6" bar would let the machine be 8-9’ wide, but would require 10-11’ of height
telescoping ‘bar’ fixed to the sled and the pivot
*what is the telescope ratio? even 2:1 could work, but 3:1 or better would be needed to fit in a normal size room comfortably
*how do you apply a pushing force
this avoids any extension of the pivoting mechanism beyond the workpiece (allowing for sled overhang that is), so it seems like it’s worth trying to figure this out.
options I see that are worth looking into
one interesting thing about the zig-zag approach is that there is the potential to take up the line with a weight not a motor, providing a constant force outwards from the top center. It would probably need to be a very heavy weight on a pulley system on the back of the machine to work, but this could potentially reduce the weights needed on the sled
hmm, I wonder if you could use drawer sliders to manage the telescoping with just enough between them to attach them together and mount pulleys
If you have one motor and chain to control the swing, with the rigid swing mount you could replace the second motor with a tension device. I’d consider putting a motor in one bottom corner, the counter pressure device in the other. In that case the motor could position the rigid frame as appropriate, then the other motor would pull the carriage up (counterweight or spring to move down), gravity/spring to move it down. Z as normal.
Springs would be nice, other than worrying about chain sag you could change the table angle from horizontal to vertical
You could use a motor at the pivot to swing the carriage, no chains to sag
I think I need a diagram to understand it.
please diagram what you are proposing.
it’s good to digram what it looks like in the corners and top/bottom center. One
advantageof doing this in onshape is that you can make an assembly that moves
and you can grab the router and just pull it around to see what things look like
in the various places.
look at the angles of any lines that you have and what direction they are
pulling in these locations, as well as how far things have to move. If you have
any springs or weights, think about how hard they are pulling in the different
locations.
David Lang
Looks like the guide rods need to be 7’ to cut over a 4’x8’ area. So you would need just over 14’ width and a little under 8’ in height.
