[Sil_x]

I am finalizing the design for my newest multipurpose table and I have been working on this idea for a while. My idea is to use an acme screw as an rack and hob a worm gear for the movement.

The table I am planning on building is going to be 4x6 and I setup some challenges/goals i want to keep to which are only 1 motor per axis and develop the lowest cost reliable linear movement possible.

So here are my thoughts on the alternatives.
Ball screws - really expensive for longer lengths but good efficiency
Lead screw - cheap (10$ for 1/2-10 6ft lengths) until you get into the larger diameter screws, also possible issues with whipping at higher speeds.
Rack and pinion - medium cost of 50$ per rack + pinon
Belt drive - about the same cost as R&P

So thats when I came up with the thought of using a cheap lead screw for the rack and making my own pinon. Simple right?

Here are some pictures of where I am, i made a hob cutter, hobbed a pinion in delrin. I made a mold from the delrin so i could cast more pinons over aluminum hubs.

So any thoughts, comments, clarifications?
Ive looked around for someone doing something similar but i haven't seen anything, so no stealing my idea and making money off of it

[bobsch]

I've seen curved pinions before (McMaster-Carr probably?) but never used them. I wonder if there will be a lot of wear on the pinion edges due to the difference in radii.

[Sil_x]

Its been a while so here is an update. I have the r&p mounted to my Y axis. It can easily do 600ipm but its doing 350 in the video. It gets a little jerky at lower speeds, i might have to do some filing on my pinion. Enjoy

[nomedia="http://www.youtube.com/watch?v=rMpuspY2IxI"]YouTube - Working Lead Screw Rack and pinion hybrid Prototype[/nomedia]

[eSilviu]

I like your idea.
It's the cheapest R&P solution you can use, and seems to work just right.
It was one of the moments "why I didn't think of that?!!!"

[quazga]

I once did something similar about 20 years ago. I was making a manually operated 3D touch scanner and used the encoders of the old style computer mouse. I had NO MACHINE TOOLS! HAHAHA! =D So I hand carved the aluminum part that you have there in plastic and filled it with epoxy putty and made an impression of a 1/4 inch threaded rod before the putty set up. So these pinions allowed the encoders to rotate via being engaged with the threaded rod. it worked!

Questions I have:
I noticed you cast the timing belt pulleys too! Good thinking! What resin are you casting in and what molding method are you using?
Does it matter at all that the center area of the rod is not connected to anything?
If so, how do you plan on connecting the center area of the threaded rod to the base?

Being that the threaded rod is so cheap you could have multiple racks and pinions allowing for more force and a means for backlash compensation. This would also allow for thinner pinions helping to reduce the problem brought up by bobsch.

Overall, very impressive! Keep the innovations coming!

[Sil_x]

The timing belt pulley is cast in 2 part epoxy. I use the us composites thin stuff. I use either vacuum my molds or pressure cast, the epoxy out gasses at very low pressure.
Here is a thread from me making the pulley:
http://www.cnczone.com/forums/moldma..._critique.html

I am still undecided on how i am going to mount the rod, in the video it is just clamped to my gantry so it supported the whole way. I was working on supporting the rod every few inches with clamps that locked onto 60% of the rod but I am not sure if that's the method I am going to use.


I am considering casting a pinion in semi rigid plastic, that might cut down on some of the jerky-ness.

[quazga]

Thanks for all the info! Very interesting! I read your timing pulley thread too.

Yeah I think going to a lower durometer pinion would help out quite a bit. Also, this would make it inherently backlash compensating.

[quazga]

Hey wait! Look at this! If you look at the specs for this and if those specs prove true, then this is the holy grail of linear motion! HAHAHA!

A Better Belt Drive - 2008-12-05 20:22:47 | Design News

http://www.bell-everman.com/Resource..._Techsheet.pdf

Couple of timing belts, a timing pulley and a couple of idler pulleys. DONE!

[quazga]

Double post error

[quazga]

Well I'll be darned! The inventor posted a thread here about it already!

The thread is here: http://www.cnczone.com/forums/linear...so_myself.html

[Sil_x]

Its been a while but the machine using this method is semi-finished. I still need to paint and do some cable management but I'm having too much fun playing with it. I ended up welding the acme screw to the frame. This worked really well. This machine is a beast and it can move over 1500 ipm. To keep it safe I'm trying to keep it down to 500 :P I will post a video soon.

[harryn]

Hi, your approach is really interesting, but I am still lost on a part of it.

To make the original outside thread shape in the gears, you had to either mold them around that rod, or cut them. You indicated in the first post that you made a hob and cut them.

How do you go about making a hob cutter for this shape ? Is this something you did just by eyeball and experience ?

Thanks

Harry

[Sil_x]

The hob cutter was made by turning down a piece of the lead screw on my lathe and cutting the teeth on my mill. To cut the gear I used my lathe with the hob cutter in the head and the blank on the saddle. Its pretty much like the process that Steve Bedair did on his site Steve Bedair's 9x20 Lathe Site

There are formulas for hobbing gear, I forget what they are at the moment but I remember that my 1.5 inch stock delrin was almost perfect as my blank.

[harryn]

Hi, Thank you for the links and explanations. That is a really interesting approach to making the pinion, and other matching gears for various needs.

[Sil_x]

Here is a vid of my machine
[nomedia="http://www.youtube.com/watch?v=8ilhAtRXwgc"]YouTube - minotaur CNC router[/nomedia]

I am still trying to get the tool length probe to work so that is the reason for the multiple attempts on the cut

[quazga]

Yeah I know what you mean by having too much fun. Having CNC capability is like going from the Flintstones to the Jetsons! HAHAHA! =D

Great work!

[Sil_x]

I can agree with that.

A few notes on the gear material. After about a month of use the epoxy resin i used has started to chip and crack. I am trying a softer and hopefully more wear resistant urethane casting.

I did hob a few gears out of aluminum for my X axis but even though I tried to get them identical, some raking occurred due to the differences in circumference. They work great on my Y though.

[Mike Everman]

I of course love the screw rack idea. I say of course because I've put some effort into it in the last few years, off and on.
Making pinions out of steel was not particularly difficult when you make a simple tool.
Blanks are put in this fixture and a spiral flute tap used to make two at a time. It is not as perfect as hobbing, where the ratio of tap rotation to pinion rotation would be tightly controlled, but it works well. Plastic would be doable with a hand drill and just two pins in a plate at the right center distance.

You just need to calculate the pitch diameters of screw and pinion, such that the pinion pitch is an even number of teeth (obvious, I know, but if side cutting just one you can feed toward the center until you find a spot where the teeth look good; in this case I've nailed the center distance, and machine two at a time so there is no side load on the tap.)

[Sil_x]

Very nice, I like the simplicity of your fixture!