[PMalin]

Have called some vendors on leadscrews, belt drives, etc.. So far everybody is way too expensive, and the belt drive even more so, which I thought would be less expensive. I need some kind of track and drive system from a vendor that is reasonable priced. Maybe start out with a belt drive low cost and upgrade later. Otherwise what options may I be able to put together a track and drive system myself from purchased parts that may be less expensive yet?

In consideration of all this, I have always been undecided on a gear ratio for motor to the drive. If the stepper motor turns x number of degrees per step what sort of displacement on the axis should I try to go for? Also maybe considering 1/2 step or what other motor driver setup may be possible for the stepper motor. What details do I need to work out so I can derive the gearing if any is necessary. Moment load is still an open question because none of this is built yet, so If I have to go with lighter weight materials, I may be able to deal with that consideration later. I need to at least come up with this gearing consideration and then maybe consider the other after that. 10 foot X axis though 8 foot in Y. Mach 4 is the software system I am using.

[TomB]

Just because you have been looking for an answer for a while and I enjoy inventing hypothetical machines that I’ll never have to build and debug I’ll offer a response. First you do not specify how much you are willing to spend and second you do not specify the precision you require so I will be creative and sacrifice precision for low cost. Second you do not specify what you intend to use as the base so I will outline a cheap base approach as part of the problem.

The lowest cost conventional drive mechanism that I can think of is rack and pinion gears. Racks come in 4’ lengths from Amazon for $90 each so you would need 7 to place them on 2 sides and the gantry, i.e. $630 plus the pinions would be about $750. Mount two steppers on the gantry to drive the X axis pinions and one stepper to drive the Y axis pinion and you would have the basic drive mechanism for perhaps $1500. But that is pretty conventional thinking so I assume that is too much for your budget. But skipping the budget concern for a moment just having 7 racks floating in the air is not a machine. You need a frame under those racks and that frame must provide tracks on which your gantry and plasma nozzle can ride.

Home builders do not have the option to make a wood model of the base then take it to a foundry and then get in the mail a cast iron frame. You might go off and consider the merits of epoxy granite frames but you can read that thread for yourself. My suggestion for low cost is structural shape steel. To build a side you will need a flat table maybe 12’ by 3’. Kitchen counter material on a 2x4 frame with lots of legs and shimmed to a stretched wire would probably be flat enough. To build one side start with a 4” x 2” x 10’ channel positioned with one side on the table. Then bolt a 2” by 2” x10’angle to the channel aligned with the one edge on the table and one on the channel flat back. Make the clearance holes in the angle a little lose as alignment will be needed later. Set a second angle on blocks so it is about 10” away from the channel opposite from the flat back and held flat with the top side of the channel. Connect the second angle and the channel with 12” slats at say the ends and quarter points. Then string an alignment wire the 10’ direction and force the channel and angles into a straight line. While the members are held in that position add diagonal members to form truss. I probably would avoid welding the truss members in place as I would fear heat distortion. Instead I’d bolt the members in place and drill, ream and taper pin every connection. Its a labor intensive process but I’m assuming your labor is free. Follow the same process to build one more long side and two more short sides (they will not need the top angle) considering as you go how you will connect the corners together. Add legs and recheck flatness and you have a base and tracks for the gantry.

I would make the gantry just like the short ends but it would go on the machine upside down, i.e. with the flat part of the channel facing down. Then add legs for the gantry, probably a flat plate with blocks set to hold the lower angle and channel as well as the upper angle (where upper and lower are relative to the inverted position of the gantry truss.)

Machine up 12 wheels that will ride on the top of the angle shape and the bottom of the channel and mount them to the gantry legs, spaced so that the gantry wheels will ride on the long side upward facing angle and be adjustable to tighten on the downward facing flange of the channel. At that point you have a frame and a gantry that will roll back and forth, except that the long sides may be bowed in or out and the sides may be out of square. Square the frame by measuring corner to corner and lock it in place with temporary corner to corner aviation cables. Get out the alignment wire and push or pull one long side straight. Hold it there by adding diagonals from the opposite corners to the middle and mounting them so the gantry wheels will clear. These diagonals assure frame is square as well so once they are in the cables can be removed. (If you think ahead about how the cables are clamped you probably can use the cables for the drive mechanism.) Repeat for the other side. Use bolt and pin techniques at all ends and where diagonals cross. Finally take the alignment wire to the top edge of the upward facing long side angles. Adjust to make them straight, flat and parallel then tighten bolts and taper pin near each bolt. Build the plasma platform as a L shaped thing with 4 wheels to ride in the gantry lower facing angle and upward facing channel flange.

I believe that frame and gantry has about 44 feet of channel, 76 feed of angle and 75 feet of diagonals. 200 feet of structural shapes at $1 per foot plus $100 at Fastenal for hardware say you could build the frame and gantry for $300, but check my figures. I guessed at prices and totaled the material usage in my head.

The frame described plus the rack and pinion gears approach would result in a machine that was $2K of less. I think a cable activated machine might cut that price by about $750. Instead of using the rack and pinions build a capstan at one end of the long side and on the gantry. I would make the capstan out of 1” threaded rod with a 1/8” pitch. Mount the motor end on a slide and the other end in a retained, matching bolt so that as the motor turned the capstan assembly moved axially. Then string a piece of 1/8” aircraft cable from the gantry to and around the capstan about 4 time then back to the other end and around a sheive set at a suitable angle and finally back to the gantry. (You may find that with 4 wraps the cable will slip on the capstan during fast accelerations. If so wrap it 8 times.) As the motor turns the cable would wind onto one end of capstan wraps and off the other while the sliding motion assured that the cable stayed locked in the thread and did not ever overlap. Use the same approach on the gantry to move the plasma head holder. That is a big table and the gantry might want to twist. As such I’d use a 11’ long threaded rod on the frame mounted drive with wires on both ends of the gantry but to keep thing cheaper it would likely be two 2’ threaded rod sections connected by piece of galvanized pipe. The cable machine would not be as precise as the rack and pinion but with care and a calibration process it might hold 1/32 over the full table

There is a lot of thinking in the note. I’m sure I’ve made errors and I not really ready to take the time to address them as they come up. I think I given you a good starting point but from that point forward you’re going to have to think through all the details. To paraphrase the old engineering books “Proof of the concept is left as a student exercise and modeling is recommended.”

[TomB]

As I got reading what I wrote I realized I had not addressed gantry stiffness in the x direction. Add some diagonals from the end of the gantry to a point 1' away from the gantry in the x direction plus a member from that point to the gantry. I would put those members on the flat side of the channel. z direction stability would likely require an additional member from the three member intersection to the top of the gantry.

[RCaffin]

Have called some vendors on leadscrews, belt drives, etc.. So far everybody is way too expensive, and the belt drive even more so, which I thought would be less expensive. I need some kind of track and drive system from a vendor that is reasonable priced.

Maybe you need to reconsider what is a 'reasonable' price. It seems the rest of the world does not agree with you.

Cheers
Roger