Amazing! Great idea and thanks for sharing!
Ha! With a little engraving you’ve got one of those rolling pins that makes cool shapes! Very nice.
Brainstorming here - once we figure out how to reliably spin the objects, one could wire up an arduino with a light sensor/photogate and have the GF self-index by passing over a certain part of the bed, and use the software to get it there, either by having a “dummy” point or some feature in the “cut order” menu.
System is completely separate from GF, no fiddling with Y-axis, etc.
I also thought along the Arduino lines but… then we hit “non-open sourcery” alerts… The GF laser path sequence is based on some efficiency algorithm. This means (apparently somewhat ironically) that the laser jumps from one part of the cut / engrave pattern to another. If it were all ‘sequential’ and determinable, we could simply adopt the rotary pattern like you suggest for a given design. But since this is not the case, while it’s not impossible I am sure, it would be a lot harder.
A lot would be possible with the GF if users could get access to the source code or even just the G code of a certain design. But then we’d also need to be able to feed whatever mods back into the machine somehow so we’re now moving more towards a dev kit… Will they do it? Time will tell.
My thinking on this months ago was to use Arduino, but manually advance remotely using my phone or add a pushbutton that will advance the rotation by 1 increment.
I haven’t been keeping up with the GFUI all that much, but I’m fairly certain we can set/re-order the operations. This feels ‘hacky’, but if I have a drawing layer with a separate color, it is it’s own operation. If that layer contains a single short line/point, say, in the top right corner, and place my sensor there, I can create a series of features that always returns the cutter head to approximately the same place. (and not laser my sensor haha) So regardless of what the efficiency algorithm is doing with the rest of the job, those drawing interrupts would bring the head back to the index position.
[index punch 1] // photo sensor LOW, rotates n seconds
[index punch 2]
Maybe it could be a combination. Once we understand a pattern we could automate within a given ‘sequence range’, using the benefit of the GF’s autofocus - i.e. allowing for some curvature - and just automate the incremental index rotations just like @karaelena has done manually.
It would be very cool if the GF software allows us to create ‘sequence block overrides’ (man, I should work in Hollywood). This could solve the whole thing for some projects. It might look something like this:
- Simply draw boxes over the design in increments that equate to the diameter of the object we’re engraving.
- Place and align the object on the rotary indexer.
- GF completes the first sequence, parks the head in the corner and give a little beep sound.
- The rotary indexer knew how long each ‘block’ would have taken and then rotates one set increment.
- After a per-determined pause the GF resumes, doing the second block.
- Engrave, Rotate, Repeat
@cbarker, I think we just said a lot of the same things at the same time…
The only thing is that ‘re-ordering’ of the operations. My understanding is that we’re pretty closed source at this point.
I wouldn’t be surprised if GF have something cooking on the rotary front already. I think it was one of the early ‘hopper’ items.
I’m disappointed I didn’t get to use ‘Sequence block override’ though.
Here’s a quote from Dan
But it’s what you were thinking, right?
Ok, awesome. Thanks, I missed that! That might be enough to give us what we need here!
Yeah… but the best I could do was drawing interrupts. I need to go back to technobabble school.
It would be nice if we could have a generic sequence block override that allows positioning the cutter head without having to use a drawing feature, but as it stands I think this would work!
Thinking about this more, would need to have a little ablative “tower” that is adjustable to the height of the work piece in the index-zone to give the camera something to focus on.
I’d say you’re awesome, but some things are too obvious to be repeated Very cool!
Here you go, youtube is your friend.
As long as you can specify the order of operations in the UI, you could just do: main score/engrave, trigger engrave to set off the mechanism, negligible-power engrave off to one side to delay the head while the indexer is moving. Rinse, repeat in order as many times as you need.
Or, if you’re willing to risk fire and explosion, you could have a wireless thingy running the indexer (particle, onion, pi etc) that would wait until the engrave was done and then tell the GF interface “repeat last job” and trigger a robotic button-pusher outside the case…
So cool! This is something I’d buy from the design catalog!
and voila! Another functionality that not even GF could foresee (or maybe it was in the hopper )
That is super clever!
WOW! Amazing work, and I can’t wait to buy this off you from the catalog! I"m so thrilled it works!