The real structural element is the beam connecting the motors directly, however having more structure aids the 2x4 so you could remove the plywood and the motors wouldnât move relative to each other, but itâs stronger with the plywood in place.
but the arms are not connected to the legs, they are connected to the plywood
and the legs are not connected to each other, only to the plywood
and the two new arms are connected to the plywood, not the legs
correct?
If so, you could not remove the plywood without ending up with 5 separate pieces
It would be easy to run a middle stiffener because there is a gap in the legs,
however running them right at the top or bottom could be an issue because the
legs attach at the top and bottom of the sheet. Basically the legs have not
changed from the original design so if you did it before it should still
arrive.
the stiffener doesnât have to go across the full width to be valuable.
Iâve never moved my motor mounts to cut different thickness of wood and I
havenât seen a chain skip in 6 months. I think the real issue is getting
everything aligned and making sure there is no flex which could make the
motors move out of alignment.
The issue doesnât come up if you are only cutting plywood, but if you are
sometimes cutting plywood, and sometimes cut on 3" thick slices of trees, you
want the ability to adjust the alignment.
That makes a lot of sense. I like having the strongest points be right where
the motors attach so bringing all the beams together there. Maybe we could
compromise by moving the A frames out to the edges so they are closer to the
motors but still not attach to the plywood?
they put the 2x4âs in their weakest position as far as forward/back forces go
(if the plywood is attached to them, they donât need side strength)
this angle makes it hardest to fasten to them on the side as you have the
thickest part to drill/bolt through.
This then requires the gussets at the corners to make things work. These need to
be CNC cut due to the angles, and still end up as relatively weak joints because
you have the 2x4âs only touching each other on corners, with all the weight
being transferred through the fasteners (especially, as was pointed out earlier,
since we donât even glue them together)
Also, the long flat surfaces of the legs means that if there is any uneveness in
the floor this is on, you loose stability (instead of the front leg and back leg
touching the ground, you may have the front and middle of the bottom board
touching the ground with the back of it up in the air)
The back of the legs arenât connected to each other, so you canât just put
casters under the thing.
I have thoughts on addressing some of these, but I wonât be able to do serious
CAD work until after business hours
We need to make sure that people understand that there are no critical angles in
this build, you can be off by a LOT and still have a working machine.
The only really critical things in the frame build (the sled has more critical
things) are that the motors are very rigid in relation to each other (why we are
moving to a solid piece of wood between them) and that the motors are spaced out
so that the chain angle is reasonable
Is that good though? I imagine angle variance would play a role in chain issues, sprocket wear, motor wear, skipping, and so on. Are these things not really a problem made better or worse by angles?
Understood that it doesnât play that much into accuracy to have the angles off but it does effect how the top beam gets mounted (per Barâs latest design). You get your angles too far off and your 10â beam wonât be long enough or too long. This will wreak havoc on the corner braces that are slated to be CNCed.
Also, depending on how things land, it might be hard to get your top beam level when your playing with angles. This effects more the position and orientation on your work piece being loaded on the machine. If my top piece is off level, I need to load my work piece off level to get things lined up.
@dlang is right that angle really doesnât matter. I was hoping that the fact that both the 2x4 and plywood are factory cut would help, but I see no need to have angles at all. That is valuable feedback.
I just meant that moving the A frames closer to the edge of the sheet would make the the point that they attach to the top bar closer to the motors like this:
We would probably need to add bracing to the middle of the sheet, but that is a good idea regardless
I was thinking of the C clamps like I use to anchor the work piece to the back. For those, big square holes would work, separated by about 12 inches. I wasnât thinking of those nifty peg based ones.
I think a discussion perhaps in a different thread. Clamps VS Fasteners VS Adhesives, When to use each. Really this would be a tutorial. Anyone want to tackle this?
FYI - I would suggest a 2x4 across the diagonal of the back of the plywood. Iâve been thought to use reclaimed lumber since before I cold ride a bike. It can be multiple parts.
Is that good though? I imagine angle variance would play a role in chain issues, sprocket wear, motor wear, skipping, and so on. Are these things not really a problem made better or worse by angles?
If you:
have the motors mounted solidly (relative to each other)
attach the chains fairly near the center of gravity on the sled (height above
the workpiece)
have the motors positioned so that the chain is fairly parallel to the
workpiece (± 1/2" is easily good enough, ± 1" is probably good enough)
then the rest of the machine can vary quite a bit and get good results, the tilt
back can be 5-20 degress (it seems 10-15 is probably ideal), the height of the
motors above the workpiece can vary a lot, the spacing between the motors can
vary a lot (both of which can affect accuracy in some areas, but itâs not clear
that the âstockâ measurements are actually ideal)
so worrying about angles being exact is just not something you should be
worrying about.
Understood that it doesnât play that much into accuracy to have the angles off
but it does effect how the top beam gets mounted (per Barâs latest design).
You get your angles too far off and your 10â beam wonât be long enough or too
long. This will wreak havoc on the corner braces that are slated to be CNCed.
well, that includes a lot of things about the new design, and why it may change
to not include the wings
Also, depending on how things land, it might be hard to get your top beam
level when your playing with angles. This effects more the position and
orientation on your work piece being loaded on the machine. If my top piece is
off level, I need to load my work piece off level to get things lined up.
This is one of the reasons I like the 10â 2x4 on the bottom, I can setup a quick
job to make a âhorizontalâ cut that just hits it. If it wasnât level to the
motors before, it will be after the cut.
May I suggest that we try posting the frame designs without the plywood so that
we can see all the frame pieces? In Onshape, just supress the sheet of plywood
and everything else should show up properly.
If the frame wonât stand without the plywood, then I consider the plywood
structural, even if itâs not in the middle of the chain tension diagram the way
it was stock
The one exception is that I am willing to account for the plywood to fight
racking forces (the tendancy for the machine to fold up if pushed from the
side). The machine would be able to stand alone without the plywood long enough
to change it out and even a pretty chewed up and drilled through sheet of
plywood still provides a lot of racking resistance
So, throwing my hat into the frame design⊠This is what Iâve been thinking of. It does not use the plywood backboard for structural support but will use it for extra rigidity. Iâd like a beefier attachment between the top beam and the legs but am just showing some screws at the moment. Iâd also like to find a good way to terminate the slack end of the chain so that it is also in-line/at same height as everything so as to avoid it getting torsioned when its length is short (i.e., during the âmeasure distance between motorsâ in calibration)
I very well may have some flaws in the design, I only ask that you be gentle.
Here is an alternate 3 (only roughly to scale, I didnât define all the exact lengths), following Barâs example for spacing out the top beam (I show this with two layers of 2x4, which is ~5" from the frame to the chain) as Iâm typing this, Iâm not sure 5" is enough. isnât it about 3" up from the sled, which is another 1.5" just from the sled and plywood, not counting your backer board and workpiece. So that would mean 3 2x4 spacers, or 6 3/4" plywood spacers. Thatâs a lot of spacers to stack (especially if they arenât glued so that you can add more to work on thicker material)
Iâm not sure that is going to be any better than the arms of my alternate 2 version
the length of the back legs can be made longer if you want to have space for storing plywood inside
Both of these use the 2x4s in their strongest direction, and require no CNC cut parts for the frame (so no temporary frame is needed)
alternate 3 is 25 pieces of 2x4 + the top beam (if you donât want the 10â 2x4 on the bottom, you can add a couple of blocks to mount them to the legs, flush with the frame and then add tabs) I havenât gone through to figure the lengths of all of these and how to pack them into standard lengths
speaking of standard lengths, in metric countries, what are the standard lengths of boards? here in the US they start at about 6â and are available in 2â sizes beyond that
@madgrizzle the only think I donât like about your design is the frames have the wood in the weak direction and require plywood gussets (and therefor a temporary frame)
and itâs better if you donât have horizontal things the full length of the frame in any direction. This is why furniture always has some sort of relief/detail up from the ground between the legs in both directions, it may not be much, but you donât want a long flat surface to discover the hump in your floor.
