[partsman]

I'm going to be building another plasma table and I'd really like to do away with the open belt reduction design like I have now and see on most machines. Is there any reason why I couldn't just mesh two 20 degree pinion gears to get my reduction instead of two pulleys and a belt? It seems like a small, sealed, oil filled gearbox would work pretty nice and be a lot more compact. I don't see anyone else doing this so what obstacles am I not thinking of? Any suggestions?

[xalky]

Backlash would be the one of the reasons that I could see, not to do it. Expense, would be the other. If you can work it out, why not?

Sent from my SPH-L710 using Tapatalk 2

[jimcolt]

When you get out of the low cost entry level machines and look at industrial cnc plasma machines.....almost all of them use servo drives (wide torque / rpm capability) that are gear reduced using low backlash planetary gearboxes. Industrila cnc plasma machines are machines that are built to use inductrial, high definition (not air plasma) systems and are designed to have superior acceleration, accuracy and are built to run at high duty cycles for 3 shifts daily. Prices typically start at about $75k and up fo this type of machine. On entry level machines belt reduction or (no reduction) drives are used as they are typically less costly. Start adding oil filled gearboxes with low backlash....and the prices increase!
A well designed belt reduction has low backlash and actually works very well in thes applications.
Jim Colt Hypertherm

I'm going to be building another plasma table and I'd really like to do away with the open belt reduction design like I have now and see on most machines. Is there any reason why I couldn't just mesh two 20 degree pinion gears to get my reduction instead of two pulleys and a belt? It seems like a small, sealed, oil filled gearbox would work pretty nice and be a lot more compact. I don't see anyone else doing this so what obstacles am I not thinking of? Any suggestions?

[LeeWay]

You can design covers for them if they are designed right.
That would keep the gunk off them if that was the issue. I love the belt drives I bought from CNC Router parts.

[partsman]

I looked into the planetary boxes but they were expensive & had a lot more reduction than I needed. I thought about just building covers for the belt but decided to go ahead with the gear rout. Here is what I have so far (rough prototype). What do you think? Gives me almost 3:1 reduction which is perfect for my application. Cost me a little less than the popular belt drive kits... as long as I was working for free I'm going to fill them with grease instead of oil though so I dont have to worry as much about leaks.

Attachment 182280

Attachment 182281

Attachment 182282

[LeeWay]

Nothing wrong with that. It looks great. I am interested to see what if any backlash it has.

[partsman]

Nothing wrong with that. It looks great. I am interested to see what if any backlash it has.

I'm thinking that I should be able to take the backlash out with the slotted motor mount. I think that will be a benefit of this style over a planetary setup. We'll see though. Waiting on my motors & electronics from onecncandc.

[LeeWay]

I know the spurs are probably off the shelf and fairly cheap. When I was messing with my 3D printer making gears, I tried some that were spiral type for the extruder. They seemed to work a lot better at taking up backlash than the straight teeth did. They may have just been printed better than the ones I bought with the kit though, so not sure there. Worth looking into though.

[cjtoombs]

How much backlash is too much on a plasma table? I thought the inherent accuracy of the cutting process would hide some. When does it become noticeable in say, cutting out a circle? One thought I had is that the backlash could be taken out by a weight and pulley system providing tension in one direction, thus taking out he backlash. The disadvantage is that it adds momentum and friction to the system, and would likely result in slower movement capability in one direction (not to mention complexity).