This is good advice and someone who serially births new machines, I know exactly what you mean Please donāt be offended if I do not entirely follow your advice.
I will be building it essentially stock initially. It will go on a very basic table on the floor and I will use that to cut the parts for the torsion box that will serve as the top for the new workbench, as well as cut many of the components for the table saw, dust collection, router table and downdraft table that are also going in to this bench. So I will get a good feel for how the machine cuts and performs before I implement my changes.
I think the only thing I have ever built close to stock is the Annex K3. I am aware of the pain this sort of thing could cause.
This total gantry redesign for the v2.4 was my last major project and was hundreds of hours of work. GitHub - MakerBogans/Fourbie: AWD 2.4 Mod with VZBot belt spacing on X gantry
It went through multiple generations including a ground up redesign, and then several iterations after that.
I say this not as a brag(apologies if it at all comes across that way) but to demonstrate that I understand where this is coming from, and the kind of impact and work it requires when you take something known good and change it. Especially a system like this that is relying on good design to make the best use of minimal/good value components.
I do have some idea about why the design and engineering choices have been made the way they have in this instance.
I also understand the want of people in the community to protect the community from people introducing new mods that change the way the machine works, and that other people may replicate - making the experience for new users worse, or make the machines worse.
So to anyone reading this - if you do not understand the compromise of any mod, please do not do it. I am knowingly and meaningfully changing the design in ways that reduce the performance of the machine to make it a better fit for very specific use case and design goals.
Iām also going to keep trying to think of solutions that allow me to keep the stock geometry of the Y rail and belts. Iāve yet to come up with anything I like, but there is still plenty of time for that!
One thing I am considering is just setting it up as a half or even a third sheet size by default with a rail extension(printed joiner for the rails, perhaps threaded?) that can be installed and the ability to move the belt mounts for full Y travel as needed. This is yet another compromise. But by far the easiest to implement.
Haha! Yes, in fact my concept for full sheet CNC prior to deciding to go a lowrider based design instead was to build an almost vertical CNC using a 2020 and steel framed on an 80 degree angled stand. Sized to fit in a 60cm wide cubby and just roll it out when I want to use it.
However I really want it to just be ready to go at all times if possible. So I will play with this concept - and if it fails Iāll have learned and will pivot.
So it looks like I would need to drop the Y rail by 27mm, and the belts by 36mm to get them both even with the table surface.
My initial impression is thats not great, but not the end of the world if I donāt go too overboard with spoilboard thickness. The effect is roughly like adding a very thick spoilboard.
I had been hoping to take a bit off the bottom of the LR core to compensate, so I could mount the router a little higher to compensate. That seems tricky due to nut clearance for the rear bearings. Itās already quite space efficient.
Iāll think on that, there may still be a way.
The other thing I can do to help compensate for the reduction in stability is giving up some Y space efficiency - which I can afford as my table is already oversized - and move the Front and Rear rail rollers further apart as I move them down. That doesnāt compensate for moving the belt down but it at least helps with the roller position.
I may be wrong, but my gut feel is the reduction in performance will not be earth shattering. Perfect from an engineering standpoint, no. But possibly an acceptable compromise.
To compensate for the lowered belt position, so far my only solution is to make the Y drives stiffer - more material or stiffer material so the extra leverage doesnāt cause deflection of the parts. May need to do the same thing for the YZ plate as it will be taking the brunt of that extra leverage once the Y mount is stiffened up.
Have not explored how I would approach the design of that yet - that is for another night.
If you look at the belt holders for the LR3, if you extended the rail rollers downwards by the 27mm, you could make the belt holders the same 27mm higher, and simply remove the belts in order to use the table saw.
In fact, youāll need to remove both belts anyway. It might be an idea to have both belt holders a bit lower, as the normal belt holders leave a stub above the table surface.
At least they are very quick to removeā¦
@DougJoseph has a mod to put the belts below the table surface that might also work, if you donāt want to remove them, however I find it to be a quick process when I want an assembly table.
Iāll definitely consider keeping the belts removable.
If I find it takes away from the intent Iām going for I can always change it, so long as I account for that possibility in the table design. It seems swapping Y motor mounts is fairly trivial, so definitely room to try different things.
Iāve looked at @DougJoseph mod, and it seems very good and well tested, well liked by the community.
I think Iād only need to drop them just below the surface rather than bringing them down further and nestling them in the strut. The strut doesnāt work initially with what Iām thinking for the table saw fence. I could possibly change that. Iām thinking similar geometry for motor placement. I could probably get away with dropping them less than required to put them in the strut and add more more material for stiffness since I donāt need to cantilever the idlers back inside the strut. But we will see, more CAD time will be very informative.
On the rail side, I could extend the zy plate down further and use the stock mount or an almost stock mount potentially. This makes it harder to undo if I decide to go a different path. Probably not worth the effort if Iām likely going to do something similar for lr4 when it lands
The imminent release of LR4 definitely supports the idea of just modifying the front and rear rail mounts and leaving the belts where they are
So where I have ended up is modifying the side plate and essentially just mirroring the Y drive and rollers so the rail is on the outside and the belt is on the inside just below the table surface.
The fornt rollers moved forward 30mm and 27mm down, the rear rollers moved back 30mm and 27mm down, the motor mount moved down 38mm. Rollers spaced out like that to lengthen the wheelbase to counteract the larger moment caused by dropping the rail 27mm
Then I just need to do rail mounts and belt mounts/tensioners and I can start printing and cutting and building while I finalise the rest of the table resign.
May yet happenā¦ we will see. Depends on rail mounts I wonāt compromise a rigidly mounted rail for this, but if I can make it work I will - simple is good
Need to mock up the router, add a spoil board and see if I need to lengthen Z at all/move the z motor mount down. The rails and leadscrews I have need to be cut down to size anyway, so should confirm before I do that.
I have decided to use a smaller pipe for the Y rail - 21.3 as the way the bearings sit on it should be more stable than with a 25mm pipe
One of the key design elements of the LR series is the single EMT rail for Y axis. This leaves Y travel unconstrained on the far side and thus not prone to binding. The belts are what position the axis, the single EMT rail forms a line for Y. There isnāt a need for the opposite Y rail to be perfectly parallel, because there isnāt a 2nd rail.
What youāve done is to introduce a need for your two Y rails to be perfectly parallel, and moreover perfectly straight (and perfectly retained).
This is workable, but thereās a reason the design is what it is. In my opinion, youād be better off modifying the non-rail side to have an offset so the rollers can ride below the table, and forego the tubing on that side.
Unless I am misunderstanding you, I think you have misunderstood me!
I am not adding a second Y rail. I will only have a rail on the right hand side of the gantry.
The reference to mirroring is me mirroring the roller and the motor mount so the motor and belt are on the inside of the YZ plate and the rollers that roll on the Y rail are on the outside.
I am not adding a rail to the left side. That would require me building a steel or alu frame to mount the rails to - wood expands and contracts too much and too inconsistently to maintain alignment well over these spans. I understand this. While I am confident I could align it by designing rail mounts with fine adjustment - I am not confident it would stay that way.
The only change I am going to make on the left side of the gantry is to move the belt so it runs just below the surface of the table. See the last picture.
The goal here is not to run two rails, just to move the belts and rail to be flush with or just below the table surface, so I can pass stock over the top without hitting the rail or belts.
on the left side, the only change to stock will be to the Y motor mount lower so the belt sits just under the surface of the table. I have not done the cad for the part yet, but this render shows the rough positioning.
I had considered that before I even started this post, but honestly I donāt think even rollers that run under the surface are necessary to maintain stability in Z - with a 1500mm gantry and gantry weight, gravity should be sufficient for the level of precision expected of this machine.
Like two rails, again it would require really good tolerances, in the thickness of the wood and doesnāt account for expansion - unless there is some kind of compliant mechanism or material(like the roller blade wheels used in earlier versions of lowrider) to deal with these variations that is also not a great idea.
Happy to let gravity do the work.
TLDR for the below explanation of changes;
Y rail moved flush with table surface and put on the outside, belts put below the table surface and close to the table. Minor geometry change to the ZY plate on the right hand side to increase stability, counteracting instability introduced from moving these components lower.
To be clear, the changes I am intending to make to the base low rider design are as follows
New belt mounts/tensioners that bring the Y belts just below the surface of the table.
New semi adjustable rail mounts that allow me to side mount the rail to the table and fine adjust it so it is straight, parallel to the side of the table and most importantly flush with the surface of the table
On the right side of the gantry mirror the belt mount and Y rail rollers so that the belts are on the inside 4mm from the table edge and rail is on the outside, 6cm or so from the table edge.
modify the ZY plate on the right hand side so that the front roller is 30mm further forward, the rear roller is 30mm more rearward, and both of them are 27mm lower so they can ride on the new lower than stock Y rail.
On the right hand side the motor mount is moved 38mm down to lower the belts below the surface. I recognise this will introduce minor instability as it means they are no longer in-line with the spindle I will attempt to increase the rigidity of the parts to help counteract this.
on the left hand side of the table the motor mount is redesigned so that the belt runs in a similar position to the right hand side belt - that is 4mm from the table edge, moved 38mm lower so it sits just below the table surface.
Please donāt be offended if I do not entirely follow your advice.
