Yeah I was thinking a replacement tray with this built into the gap below comb would be a good way of accomplishing the upgrade.
Do industrial machines have that type of feature?
I haven’t heard of such a thing, but it wouldn’t surprise me. It was one of those ideas that come to you right before you fall asleep. Haha.
Would it be possible to detect the IR scattered by the floor of the forge with a stationary, but wide angle IR detector ?
Might need a layer on the floor to scatter more light.
You’ld have to allow for the crumb tray interrupting the beam, but mechanically a lot simpler.
not sure how that would work. below the crumb tray would be looking at a solid bottom of the crumb tray. otherwise the crumbs would fall straight through the honeycomb and land in the bottom of the machine.
Pretty sure they’re talking about a gap between that tray, and the honeycomb grid…
Yes, There’s a 1/4" gap, or so, between the bottom of the honeycomb and the metal base of the crumb tray.
I’ve already got an idea for converting that space into a vacuum bed !
Ithought about this back when I had my K40, and a lot of cutthrough problems. I think one problem is the honeycomb will interrupt the beam, and that would read as a penetration failure. It would have to know where to expect the interrupts. Maybe this could be done with the bed camera.
I’m way out of my depth here, but could a combination of the vector design with a stored pattern of the crumb tray subtracted from it, be used as a comparison of the output of the proposed ir detector ?
It would mean determining the registration of the cutting path with the tray, but perhaps the inclusion of some sort of ‘virtual’ registration marks into the file could be the answer.
Getting a bit wooly now !
You’d also have to run without rear masking…
Why would you need to run without back masking? You’d see it getting cut through or not, right?
That being said, I’m not sure how to do the ‘seeing’ without some thing dramatic, like replacing the crumb tray with something with a camera looking up at the bottom of the material?
I’m not sure about industrial engraves, but I work on industrial metal cutters and we use a system that senses a change in the light wavelength to determine when the beam gets through the metal. Not sure if this could be accomplished the same way because non metallic materials don’t reflect light the same way.
Just thought that the smoke under the honeycomb would scatter all the ir anyway, so no need for any additional material.
2nd thought, which may be a killer to this idea - the apparent random cutting sequence.
It’s already been noted that the head seems to often takes a weird path, so matching the expected ir appearance under the bed with the original file might be problematic, even on a good day !
I’ve done some cuts where the material is cut fine but the back masking is not reliably cut through. So a sensor that relied on cuts all the way through the stack would give false negatives.
This was actually the original idea. When we couldn’t reliably detect cut-through, we gave up and created material that was consistent enough to cut through reliably - and thus was born Proofgrade.
Oh, that’s really cool! Man, I would love to see a mini documentary/podcast about you guys from start to now to hear the whole story. Startup by Gimlet would make for an intriguing episode.
I had lunch with Alex (the CEO of Gimlet) a few months ago and we were brainstorming about ways to work together. But that podcast covers more exciting stories than us… Glowforge was pretty low on the drama scale compared to most startups.
No way! Haha, thats awesome.
I beg to differ! Being such a huge crowd funding campaign, the drama of the delays, the way you guys handled the concerns of customers and made up for things with different bonuses along with how huge of a community has come out of this… I think there are a lot of things that the listeners of Startup could learn from Glowforges journey.
But I digress. I’ve just been on a podcasting binge as of late.
How about the Tested team?