The collets are inline.
You need the rails to bring the tool up as high as possible, the only way to go up higher would be a second set of lower bearings on the center…thought about that…no.
Bringing the lower holes up make it stronger for router mounting…and weaker for other low mounts (pen, drag knife, laser, extruder) but those are all such low force it should not matter.
I’m thinking about the function of the section of the Z rail below the bottom hole, and it’s not clear to me that it does much. It’s not nothing, but conceptually (simplified) I might think of it as four solid anchor points and the rest of the contact with the rails is relatively soft. Then the extra length of the rails below the lower hole is of not much value.
I recognize we want to minimize the length of overhang from the cutting bit to the end of the Z axis but is it the end of the rails that matters or the location of the lower hole? Extending the rails past the holes makes the overhang look shorter (compared to the rails) but is it really helping?
The rails dictate the height to which you can raise the Z axis. So end of the rails = the end of the collet. There are rare cases (carving a large bowl) where you would possibly want the collet to stick out further.
So the rails have a set distance, but the mounting holes do not really, they just need to be close to the clamp bolts or lowest and highest mount points for the tool (router).
So if we all used the same router it is a no brainer. But to keep it as universal as possible the holes matter a bit more. I think choosing the 4 routers in that pic I have encompasses all the most torque you can apply and all routers should fit into those categories/sizes. Everything else is less force on the axis.
So I think there could possibly be a tiny rigidity gain moving the hole up 10-15mm for the 660, but it is in the right spot for the 611. Will it matter…probably not a noticeable gain. Keeping it as is makes it an easy upgrade for all the current users.
Just get people to drill 3 holes. I do for my mount. They can use two only if they want but I like the 3 personally.
Late to the party, but this sounds fun.
Looking forward to new revisions.
A question about the belt attachment. I really see no good point of why the belts on current version is attached with a zip tie loop in both ends…
Wouldn’t it be possible to have it slide in place in the leg at 0/0 instead and then tightened at the other end of your frame?
This way the teeth of the belts will be aligned, so will pulleys and motors so it should be a tiny bit more precise?
As is we ziptie both ends and hope it aligns(unless you have dual endstops)
I use a belt tensioner block. No zip ties.
Still the same right? loose in both ends.?
I don’t see a major disadvantage to this, although there is a minor disadvantage to having a larger variety of parts and having to pay attention to the asymmetry.
But aligned belt teeth might not be as helpful as you think, because the cogging of the motors is at 1/50th of a revolution which is not a whole number of belt teeth. So you don’t automatically get each end to be at the same position from the end just by aligning the belt teeth. Each cog of the motor is 16 teeth / 50 = 32 mm / 50 = 0.64 mm. The fact that it’s not a multiple of tooth size means that if you loosen the belt and manually cog the motor, you can adjust the relative distance by much less than a belt tooth. This means aligning the belt teeth to the same distance is not necessary for getting a square machine. And even if the belt teeth were perfectly aligned, you would have to align the cogs of the motors to take advantage of it, so essentially the same process, meaning there is not much advantage to aligning the belt teeth.
Not sure if that made sense…
Well, that’s not exactly right, since a single step on a 1.8 degree motor is 1/200 of a revolution. so it’s 32mm/200 instead. 0.16mm. So it’s 12.5 steps per belt tooth. If you go to 20T gears though, then it’s an even multiple 40/200 = 0.2mm, or 10 steps per tooth.
It’s true one “step” is 1/200th of a revolution but one “step” is a 90 degree phase shift in the magnetic field (or current of the two coils). So if you “skip a step” it’s a full 360 degree phase shift or 4 whole steps which is 1/50th of a revolution.
Edit: This was discussed here Squaring with lasers and Peter confirmed this with an encoder and long pliers where he could feel the force.
So when the motor loses power, does it relax to a full step or a full phase shift rotation?
If you de-energize and then re-energize (via enable pin without shutting down the stepper driver), it will return to the same place or a multiple of 4 steps away, a multiple of one full 360 degree phase shift, which is a multiple of 1/50ths of a revolution of the motor.
I don’t know if the drivers work that way. When it re-energizes, it will pick one of the 4 full steps, but there’s no guarantee the driver powered off when it was at that same phase.
Thats right, I think if the driver loses power completely it will come back at “zero” phase (or something like that) which might not have any relation to where it was.
But if just the enable pin is toggled, e.g. via M84 then it should come back to the same motor current and magnetic field as before, which could still differ physically from where it was, but in whole 4-step increments.
That only happens at exactly 180 degrees. Anything less and it’ll cycle back to the same place. Motors on 3d printers can get out of sync using a series wiring (and the mpcnc/lowrider etc) but in printing on a printer using that system for a year and a half, I never noticed it happening.
I’m not quite following, but I would like to know more about this. Yet this is straying from the original topic… perhaps we can split this to its own topic?
I can help.
The Primo version no longer uses zip ties for the belts.
