I bought two 6-foot gear racks from an auction ($14 each for Boston Gear L2016-6). My desire is to modify the rollers on the Lowrider to ride on and be driven from the rail rather than pulled along the belts.
Has anyone gone down this route or has any suggestions? My hope is to make the long axis more rigid compared to a long belt length.
Not tried this myself, so this is very much a theoretical concern.
Any chance the gear rack could be mounted with the teeth pointing down, or maybe to the side? I’d be worried chips building up on the teeth could cause movement problems. I don’t have any idea how complicated that might make mounting the stepper motors.
I’ve considered orientation options. One way or another there will be a dust guard around this as I already have issues with chips impacting the flat track.
I’ve considered sandwiching the side assemblies between the existing flat bearings and this rack turned upside down with the teeth facing down. This would constrain the sides both in the Y linear direction and Z.
I’m fully aware of the current design considerations and challenges of constraining motion without binding.
Interesting options are to replace the flat bearings with gears and rails. Flipping the drive upside as I suggest above. Turning the drive on its side (sandwiching the drive across the gantry). Or maybe an angled drive that applies constraints in multiple directions.
Clearly you’ve thought way farther ahead than I have. Best of luck with the project.
This might give you some ideas:
To avoid backlash, the pinion will need some radial pre-load so it engages both sides of the teeth. The weight of the LR3 is one way to achieve this.
The issue that I see (apart from dirt) is that the rack/pinion is doing multiple duties in supporting loads in Z and Y and torsion in the Y/Z plane, while also driving the Y axis. If any pinion is slightly eccentric for example then it will translate into inconsistent Z height, or inconsistent Y/Z tilt. And I’m not sure what to expect as far as stiffness in the radial/vertical direction.
There is also the question of how you plan to constrain the X axis, if you are not using the tube on the right side.
I think it is safer to keep the motion separated from the drive, so the rack acts like the belt to drive the Y axis but not a kinematic support for other axes at the same time.
I think a rack and pinion setup is much better suited to the LR2. Put the rack on the underside of the table, and change the orientation of the Y motors 90°. This keeps the rack clear of chips and debris, and also serves to help the Y axis track true.
The rail is enough to keep the LR3 tracking true, and the belt provides excellent motion accuracy.
The new mod that I am basically finished creating basically turns the Lowrider 3 into a hybrid between a LowRider3 and a LowRider 2. It might be possible to adjust it suited to allow you to have those gear rails on the outside edge of the table facing outwards, sideways, with the Y motor positioned on one of my secondary Y plates. Here’s a link.
I’m currently doing it on only one side but it perhaps could work for both sides or maybe one side with something else on the other.
So you’re suggesting in my configuration that my racks be laid on their sides and essentially replace your aluminum track that you’ve tucked the belt inside. That might work well. One side could still ride on the conduit like the stock LR3, but extend the drive servos out a bit and geared to the rack.
Yes, essentially that. The rack you have would replace the rubber belt in my situation, and your rack would be mounted to the table edge, whereas my belt floats in air. But we are essentially saying the same thing, I think. I’m just nitpicking. The superstrut is not aluminum, but rather steel. 
