mike125 published a new build:

Read more about this build...

 

This router uses a 15mm wide htd5 belt system on all axes. The larger belts should keep belt stretch to a minimum.

 

I haven't decided yet. It will probably be either Kickstarter or Indiegogo. When the time comes, I will let everyone know about it. I want as many people as possible to be involved so that the project will be a success.

 

It definitely should run with a good dust collection system. I figure it will run better open, but yes significant buildup could lead to tolerance errors. Still debating on whether or not to double up the gantry beam and just beef it up in general. I have high hopes of this router to be able to run a somewhat powerful spindle.

 

It works the same way as the x and y axis. The only difference with the z is that the motor is fixed in the z direction and the z beam moves up and down. The x and y is opposite in that the motor moves in there respective axes and the endpoints are motionless. I did a bit of head-scratching myself while working the concept out, but I know it works, because my 1st router build uses the same setup. It is probably one of the few good things about my 1st router. Lol. I had a request for some close ups on the z axis. Here is a link to that conversation. http://www.openbuilds.com/conversations/cheetah-33254-cnc-router.99/

 

I plan to make changes to the y rail on the gantry. I foresee a lot of flexing with just the single 20X80 v slot beam. I will probably do a solid aluminum beam with openrail at the top and bottom of it instead of the v slot. I have put this project on hold temporarily because I am working on a skate bearing design that will be more affordable to build. I plan to pick this one up after I finish the one I am working on now.

 

The system uses 12 tooth pulleys. I use stepper drivers capable of micro-stepping, and resolution isn't a problem at all. If there is any loss, then it is to mechanical advantage, but this setup works just fine with 270 oz-in motors on a similar machine that I have that is 24" X48"X5". Sorry about the link, I have found those pics, and I will repost them here.

 

Sorry again for not getting the pictures up. I didn't realize that I had push the reply button after posting pictures.

 

In my experience with the one machine I have built, the microstepping has been completely reliable. I haven't had any missed steps. The current setup has plenty of torque to cut wood. I haven't cut any aluminum yet. I haven't seen any need to have a gear reduction. If I find that I need more torque to cut harder materials such as aluminum or phenolics, then I will use a higher torque motor like a 425 oz in. I like the speed capabilities of the direct drive system.

 

Yes, the z will drop when the controllers are turned off. I used a very similar belt system on the first router that I built. I usually just put a block of wood under the spindle before I shut off the controller. 2 beams on the z would be wider and would decrease the travel on the gantry axis. Dust does get on the belts, but it didn't seem to cause any problems on my first router. I'm sure that the smaller belts in the v groove collect dust too. As for running the belts under the table, my first router had a single belt down the center under the table and it collected dust as well. I didn't have a dust collector on that router. I'm not sure if the dust caused any significant accuracy issues, because my first router had a lot of slop in the linear slides, so it was only accurate to + or - 0.03 in. I plan to report the accuracy of this machine once I get it built. My guess would be that dust on this larger pitch belt would be less significant than dust on a finer toothed belt.