[panozeng]

I'm faced with a lot of options on this one, and simply trying to decide which route to go. Here's the problem and what I have to work with:

It's a 90's model 5' x 10' industrial duty plasma table made by Advance / Koike up in NY. It has a beefy steel gantry with steel guide tracks and drive on one side, and a non-guided support roller on the opposite side. I stripped off the old analog servo motors and replaced them with some 1150 ozin nema 42 steppers I already had. It was very easy to convert - a new motor plate for each one, and I reused the 3 to 1 timing belt reductions and rack and pinion drive. These motors will reliably do 1500 ipm on the X (more on Y) at a good acceleration rate. The problem is it's a rather heavy gantry and the further the torch carriage is from the drive side the worse the deflection is from acceleration. I've backed the acceleration way down and can make good parts on the drive side of the table, but really need to tie the other side in so I can put the acceleration back up where it should be.

So, here are the strategies I've been contemplating:

1. Add another motor and drive the far side. I think the most logical thing here would be to take the motor and drive I'm using on the Y off since it matches the other X and use it. I could then use a smaller motor on the Y since it weighs nothing compared to the whole gantry. That would give me a perfect match for the other end of the X. Obstacles - finding matching rack gear (not sure what the specs are for the stuff they used - it's cut from bigger stock than usual for it pitch), having to rigidly guide the opposite side (potential for binding?), and having to rebuild the Y

2. Add a crossshaft at both ends of the table and add some long timing belts. I'm thought of 2 scenarios here:
A. Simply add the crossshafts and belts to tie both sides together but leave the rack and pinion drive alone. The belts would merely hold it square and not actually drive it. This would be a fast, simple modification!
B. Ditch the rack and pinion and connect the motor via 3 to 1 reduction to the crossshaft. This would be a little more involved than A

I've also considered lightning up the gantry with some serious hole saw work. It's ridiculously overbuilt and could get some swiss cheese action without compromising it's strength. Also, I should point out that when it's locked stationary with the motors it takes a LOT of force on the far side to manually deflect it. So I'm really only fighting the springyness of the beam and not backlash or bearing slop. This is why I've been leaning towards option 2A - the simplest.

How would you fix this problem?

[magma-joe]

panozeng,

I was in a similar situation to you. I had an older model Dynatorch machine that was driven with a single X axis motor and a cross shaft system. The cross shaft had a pinion on each end riding on 2 table length gear racks.

The X axis motor had its own gear rack. There were several things with the cross shaft system that I did not like and wanted to improve.

The cross shaft system;

1. The cross shaft gears and rack were very noisy
2. It added extra rolling resistance and vibration
3. It added even more weight to the gantry

4. And the biggest problem was the accuracy of the cross shaft system. In theory the cross shaft pinion gears should stay in sync with each other keeping the gantry straight when traveling from one end of the rack to the other.

On my table I found this not to be the case. When I checked my cross shaft travel, I determined that one side would travel about 1/4" further than the other when going from table end to end. (5 x 10). I am sure it all boils down to gear quality, wear, and back lash.

I decided to upgrade to dual x axis drive motors among many other changes I made to the machine. I have been extremly happy with the performance of the machine with the added X axis slave motor. It is much quieter, smoother, more powerful, and the acceleration is excellent.

The added X axis slave motor will also add rigidity to your gantry. Depending on the condition of your existing X axis gear rack you might consider buying 2 new racks and pinion gears to maintain the accuracy between the master and slave sides. The only other consideration is the addition and extension of the wiring harness to the X slave motor. To answer your question about binding on the slave side; You won't have to be concerned about rigidly guiding the slave side only the master. You can retain your existing roller support wheels which will compensate for slight misalignment in the X master and slave sides.

With a little reading you can easily figure out what your existing gear rack is. You might check out this gear supplier. They were very helpful when I contacted them and they are one of the suppliers who stock 12' lengths of gear rack.

Moore Gear Website

Here is a link to my build; look at the pictures in thread # 82 and 121

http://www.cnczone.com/forums/dynato...h_upgrade.html


Magma-joe

[panozeng]

Here are some videos of the problem. I had the acceleration maxxed out for these - .17g's, and was just bumping the jog button. The first is the worst case scenario - the carriage at the non-driven end of the gantry:

[nomedia="http://www.youtube.com/watch?v=DEiABkBomoM"]YouTube - Gantry chatter[/nomedia]

And here it is at the driven end:

[nomedia="http://www.youtube.com/watch?v=40wA8KshdsY"]YouTube - Gantry Chatter 2[/nomedia]

Joe - I keep thinking that dual motor drive is really the way to go, but it will be more work to do the upgrade. I was focused on the way the existing rack is mounted with the pinion axis in the vertical plane. If I mounted the new rack in the same plane on the other side I'd have to add more guide bearings on that side to keep the gantry from flexing and letting the gears jump. I hadn't thought about mounting the gear rack like you did on your table, which would take care of that problem. Humm...