Here’s a picture of my ‘infinitely adjustable rails’ that I’m trying out. The concept was to make the coupling between the rails and the frame adjustable. I need to do this because most of my wood is a bit crooked and my experience has been that wood has a life of its own in my neck of the woods (North Florida). The leveling part seems to work well, but I only have one rail down so I don’t know if the squaring works yet.
The furniture leveler allows each end to be raised and lowered so that you can level the rails.
The horizontal threaded rod is secured tightly to the board that the unistrut rail is lagged into. The hole through the frame edge and leg is 3/4-inch wide while the threaded rod is 1/2-inch wide. That allows the rod to raise and lower as the furniture leveler raises and lowers the rail. I tried to get the hole in the right place so I didn’t have to drill a real big one and have to use very large washers.
Once leveled, the rod is moved left/right to allow for the rails to be squared and once everything is aligned, the nuts can be tightened to hold everything firmly in place. I’m worried that the process of tightening down the nuts will push the rails out of square so I might have to experiment with that. I think if use the nut I marked as the ‘squareness adjuster nut’ to set it, I can then tighten down the leftmost nut to lock it in place without affecting squareness.
Once I get the rails installed and the lower rider constructed, I plan to level the frame boards (shave it down by 1/8-inch) and top it with 1/2-inch mdf and 1/2-inch foam (as spoilboard). The little block of wood the leveler is resting on (maybe a bit hard to see in picture) is 7/8-inch tall so rails should end up even with the spoilboard… if I did my math correctly.
The rails are nice! I would say don’t surface the table yet. Get a few cuts under the belt, make sure you work out any kinks in your CAM first. These can cause some nasty spots on your table.
I was planning on running a bunch of gcode around the whole frame to get it help get it square and level and resolve any setup issues (binding, etc.) as well to learn the process of zeroing, etc. This setup is different than the maslow, so I’m be relearning how to do it. Then to surface the table frame, I’m going to run the gcode without router bit to make sure things work properly and then do a very light pass first so by the time I’ve stepped down to the full 1/8 inch, I’m in good shape.
Madgrizzle! How I stumbled onto you here as well. I use your we control software with Maslow! I was thinking to build the lowrider2 to replace or enhace Maslow!? Where did you end up getting your pipe? I’m in central Florida! I didn’t know you were in North Florida!
Would you recommend going this route as an upgraded Maslow yet? Or should I start a build like the lead CNC machine from openbuilds? I ultimately want to be able to turn spindles/table legs etc and need a high Z axis…
Can’t say I’m an expert on lowrider as I literally just finished building enough of it to move all the axes. If the Maslow wasn’t vertical (was just taking up too much space in my shed since I have to have it in the middle of th shed to clear the ceiling) and it had the accuracy I have been needing recently, I wouldn’t have built the lowrider. The lowrider just fits my needs at the moment. I’m suspecting we might need a four motor design for the Maslow and I’m looking to make my design for one an outdoor-type machine (self-contained that you can install on some wall mounted hooks when wanting to use it). I’m hoping to cut pieces for it using the lowrider and not pay for someone to make the pieces for me.
As for webcontrol, I am looking at the option of making it support the lowrider… I don’t think it would be terribly hard to do so, but there are distinct differences…
Just some thoughts after getting my build mostly completed.
Most everything went together fairly well, some parts were a bit more challenging than others… first, getting some of the bolts through the locknuts when it was hard to clamp down on the nuts was a challenge. This was particularly an issue on the carriage. I managed with a pair of needlenose pliers, but if its possible to captivate the screws in the 3d print model it would have been helpful.
Second, even though I succeeded, the zipties for the belts were a bit of a challenge to install. I think zip ties is a good idea for tensioning the belt, but if there was a 3d printed piece that facilitated the assembly then it wouldn’t be have been as challenging. Trying to hold everything together to get the zip ties in place was hard. Like I said, I managed to do it without cussing, just think having a piece that makes the connection to easier would be handy.
Yesterday I managed to flatten out the 2x4 frame. took 5 hours … 10 mm wide passes, 2 mm per pass… some areas needed 6+ mm shaved off. I really needed a 3-inch long bit to reach the frame from the router, but I made do with a 2.5-inch bit. I was nervous about how little was seated into the collet, but it worked. After shaving, I installed a 1/2-inch mdf top (so table top is 1/2-inch mdf on bottom, shaved 2x4 frame with three internal cross members, topped with 1/2-inch mdf). I’m amazed how overall flat it is. A straightedge shows less than a mm drop in the middle which is impressive for my carpentry skills (or lack thereof).
I attached a pen-mount and laid down some 48-inch wide craft paper and ran a rectangular test pattern across the board to get it all squared up by adjusting the position of hard end stops to eliminate the parallelogram effect.
With all that done, I decided to exercise the machine to see if there are any issues. I asked my twins to draw a couple or robots on some paper. I scanned them, converted them to vectors and simplified them and scaled them up using inkscape, and generated gcode using easel. I left most of the inkscape scan-to-vector artifacts in place and left the gcode unoptimized to really exercise the lowrider. This is the result after about an hour and a half:
I need a better pen mount than the hack together job I made… something with a spring. I also need to figure out how to better route the vacuum line/power cord because it keeps wanting to get hung up on the router head when the Y-axis is at about 75%.
Kids liked the picture and so we hung it up over their computers.
Overall, I’m very impressed with the lowrider’s performance. The z-axis is blazingly fast (this coming from a stock Maslow CNC) and a straight line is a straight line.
My next steps are to figure out a new case/power management design for the controller/rpi. Right now everything is attached to / hanging from the Y plate and it’s not tidy at all. I was thinking of creating a duplicate plate Y that I can stand off from the existing Y plate and attach everything to that… maybe on the inside of the plate and have hinged on the bottom so I can readily access it. Fortunately, I have a new cnc I can use to cut it out I’m also thinking about some kind of dust shield for the Y plates as well… After 5-hours of shaving the frame flat, there was tons of sawdust all over the place and the z-axis leadscrews were ‘furry’ with saw dust.
That is awesome! I really think this method of surfacing the lower support and then adding the spoil board is the way to go. That is also a great project to get kids excited to draw with the cnc.
What I ended up doing was to attach a small board of ply to the outside of the Y and attach it with a french cleat. I may make more changes to my controller than you, but it was much easier to just disconnect everything and pull it right off. No need to disassemble the Y to get it on or off.
I also think the frame shaving idea is the way to go… glad I read your post suggesting it. I would have been nowhere close to flat without doing it… it also helped me learn about jogging the machine around so that was an extra benefit. I do recommend trying to find a 3-inch long bit to do the shaving, but that depends upon the geometry of your setup. I have my rails planned to be near the top of the 1/2-inch foam spoilboard I’m going to add. To reach the 2x4 frame (1 inch below that), the 2.5 inch bit length was marginal at best, but it was the longest I could find locally.
I’m also thinking about some kind of dust shield for the Y plates as well… After 5-hours of shaving the frame flat, there was tons of sawdust all over the place and the z-axis leadscrews were ‘furry’ with saw dust.