Jonathon Duerig published a new build:

Read more about this build...

 

Not yet. I'll get some video of it tomorrow. It is actually somewhat boring to actually watch after a minute or two. In order to grind at the high quality I showed pictures of, it takes 10 minutes to move the glass about 500 mm. And then 5 minutes to move it back to where it started. The slow patience and consistency required is the big advantage of doing it by machine in this case.

 

@JustinTime For my case, all the cuts are the same angle and all the panes I am grinding are the same size. So I didn't try to add the complexity of an adjustable angle. You could change the angle by swapping out the acrylic jig for one held at a different angle. But I am sure that there is a way to have an adjustable angle by having a hinge at one side and an actuator at the other to raise and lower it. The tricky part would be that the edge near the grinder might also need to move (a Z-axis) so that you still get a clean edge.

It would be faster if I had a z-axis and could take it in multiple passes that are closer each time. But that adds more complexity. And turning on a machine that I can leave alone for 15 minutes while I do other things is fast enough for me. So I went with the simplest thing that worked for me.

@Mark Carew Thanks. I am using a gShield and Arduino for control. I've used a Shapeoko in the past and so I knew that it would work for something this simple. Then I just use one of the g-code senders to run the super-simple program to grind a piece of glass. At some point soon, I will probably hook up a Raspberry Pi with a touch screen as a UI so I don't have to use my laptop any more. I figure the UI only needs four buttons: 'Jog left', 'Jog right', 'Start Grinding', and 'Stop!'.

@Rob Taylor I agree. There is so much you can do with T-slot and V-slot that I hope it really branches out. As cool as cnc routers and 3d printers are, it was fun to build something totally different. I don't know what the deflection properties of V-Slot are, but it is the wheels/gantries which are the limiting factor here. The reason I went overkill and had two separate rails with two gantries each is because an earlier jig use just one rail (normal v-slot not c-beam) and one gantry and there was a huge amount of deflection when it came in contact with the grinder.

The only advantage of doing it in one pass without a Z-axis is that it made the machine simpler/cheaper. In previous iterations when I was moving it back and forth by hand, I did have a z-axis because I couldn't move it slowly and consistently enough to do it in one pass. Right now it takes me 15 minutes a pane. If you added a z-axis, I bet you could cut that time substantially with equivalent quality. Another possibility for the z-axis is to use two hobby grinders instead of one and have one grinder a little further from the glass than the first. This way you would get two passes at precise distances without the added complexity of a real z-axis.

The holding jig is all laser cut acrylic. As long as I only need one angle, it will work fine. If I ever need a second angle, I might be able to get away with swapping out between the two jigs. But any more than that and I'd have to figure out how to make it adjustable. Either by hand or by adding a rotating axis of some kind. Since it is T-slot, the jig can go on and come off pretty fast. The only slow part is doing the fine adjustment so that when it moves back and forth the glass doesn't change its distance from the grinder. With multiple jigs, I think you'd want to somehow mark each one's exact position so you can re-attach quickly.

-D

 

@Rick 2.0

I've attached a video to the build. It took me a while to figure out how to mute it (glass grinding doesn't sound very nice) and add some subtitles. Take a look here:



-D