Making jigs is a bit pointless if we don’t know exactly where to put them on the bed and…
[quote=“thomas.alessi.jr, post:12, topic:3386”]
That’d be silly. Extremely imprecise.[/quote] …correct.
The placing of the cut-diagram / art using a visual cue from the camera is a great novelty and certainly very useful for certain tasks and fun. But cameras have barrel and perspective distortion. It could of course be corrected up to a point (and I’d bet my bottom dollar a lot of development time went into refining this element). However a typical CO2 laser is apparently around 20 micron in diameter and like @thomas.alessi.jr implied, it would be imprecise if we need to ‘point and shoot’ every time we start a repetitive job.
I think it is important to have some orthogonal-XY (physical) absolute reference point.
Things will probably be much clearer once we get the unit but it would be nice understanding what we’re getting.
Btw, the GF staff other than Dan doesn’t routinely answer questions on the forums, but the community at large has an amazing body of knowledge that you may find handy.
[/quote]The problem with the ‘amazing body of knowledge’ argument is that while it may be well and true, none of the community here has a Glowforge (except 2(?) beta testers). I find it frustrating when a question like this gets lost in speculation and the original question does not get answered because we have created a nice camp fire with marshmallows.
[quote=“dwardio, post:4, topic:3386”]Expressing a desire to have GF staffer answer your question isn’t typically effective…
[/quote] Bit of a shame if the question is relevant I would think.
Despite some unclear answers and a dubious reason for the delay (hang in here with me…) I am personally very impressed with how the GF team have handled things and I am still exited about the product. But for a product that is supposed to ship within the next six weeks, there seem to be quite a bit of uncertainty as to the capabilities of the machine. I think a simple query like this deserves a response from the team - how can we do repeat jobs with consistent accuracy without using the ‘point-and-shoot’ methodology. Would we be able to all?
It’s interesting that they all (except mr. Kopel) only received beta tester status 7 days ago. What is the point of that I would wonder if the machine’s are about ship? (man, I should become a Wall Street analyst… oh wait…)
I think the big question is what @kennethclapp asked. I have been assuming, perhaps wrongly, that if you load an SVG (or other equivalent type) file and don’t do anything to it in the interface, it will zap in the same place every time. If that’s true, then we get to make jigs and rulers to our hearts’ content, and things will get cut/scored/engraved where we expect them.
If that’s not true, and every iteration of an untouched-in-the-interface cut file gets zapped on a slightly different region of the bed, then yeah, we’re in trouble. This would take only a minute or two (ha!) for @dan to clarify.
that is the day I asked the forum at large why all the new folks posting beta projects hadn’t been updated with public beta-tester badges. All were added/updated in Discourse later that day.
Indeed this is the question. That ‘zap’ you refer to is what I thought would be included in ‘Feature 8’: Optical Alignment The dual cameras align the laser head with the frame, with your design, and with your material. Glowforge realigns with every cut and engrave, adjusting timing and position, so every print comes out perfectly.
What this promises is what I am expecting but from recent responses by GF it seems we won’t be getting all the functionality on the feature list - or perhaps just a limited version of it…
The impression I’ve been going under (perhaps wrongly) is that when you just zap you get the underlying alignment of the machine, and when you select bits for manual positioning you get only as good as the camera can give you. There was an implication in 8 that if you put your material in randomly there would be some magic that translated and rotated your design to match, but that’s pretty clearly not there (although if/when the double-sided alignment happens, at least part of that functionality will be there).
I forgot to mention, but something I use very often on my laser is the test fire button which will fire the laser while being held down. This lets me fire a quick pulse to determine the exact location of the beam. This can be replaced by a red dot (preferably by using a beam combiner).
I’m wondering if anything like this will be on the gf
Sort of yes, sort of no. The amazing body of knowledge is not about the GF specifically (except perhaps for the beta testers & staff participants in the forum) but about laser cutting, design and a host of other things. Tips & tricks that apply to design for most any laser applies to the GF as well. Ditto materials experience, etc.
So general knowledge outside the actual GF users about the GF is low, although somewhat higher than the general public. But there are a number of people here with deep design experience and/or laser operations.
At the end of the day, this thing is a 40W CO2 laser and the body of knowledge about machines of that class is relevant. You can discover that knowledge through trial & error or get it from people with decades of experience with lasers.
For specific GF questions that you have a sense of urgency around, you’re better off sending an email to the Support team at GF and not use the forum. You could share the results here but that’s a direct link to GF staff, this is an indirect link.
It seems that when you pick your design in your design library, it gets dropped in what could be considered 0,0 (top, left corner.) If it drops it there every time, doesn’t that work for dropping in job after job?
not having a consistent 0,0 makes me really uncomfortable…
optical alignment of the head can change over time due to changes in camera positioning from things like hinges not closing the way they used to, plastics deforming, etc.
If I make a part that breaks down the road, and I need to remake it to less than mm accuracy, not having the ability to physically line something up will make that so much more difficult.
In a machine that is supposed to be designed for laser point accuracy why would anyone choose to remove that from the equation?!
Because end stops are a $%# pain. They’re just one more bit of electrics to get bollixed up, or optics to get obscured, and if you can avoid that point of failure you’re just that much better off. (Speaking as someone who has “adjusted” the endstops on a 3D printer endlessly and replaced both inductive sensor and motherboard when electrical noise fried them.) Btw, if I understand the process properly, isn’t it the narrow-angle camera that tells the head where zero is? That thing should be ridiculously close.
There’s only the Big Button so far as on-device controls go isn’t there? I suppose GF programmers could implement that sort of feature in the UI and the Big Button would do the firing but an on-device button for that purpose would be a lot more convenient. Just guessing, but a dedicated manual pulse button doesn’t seem like the sort of feature that one could home brew very easily.
If end stops are the same thing as limit switches there is no reason that they should be a problem for something like a GF. I have a 3D printer, CNC lathe and CNC mill, all with limit switches and none of them have been problematic in 8+years with the mill, 2+ years with the printer and over a year with the lathe. What sort of problems have you had?
If end stops are different than limit switches, ermm - never mind, though I would like to know what they are.
They’re essentially the same as limit switches. And they’re fine as long as everything goes right. But they are a point of failure. I’ve had them fail in one 3D printer because a little crumb of plastic got in the wrong place, on another because a cable bundle dragged on the switch bar, and on that same printer because electrical noise caused the switch sensor to fail. And for best results on one printer I have to adjust the vertical end stop by some fraction of a millimeter between winter and summer.
None of these are huge issues, but they’ve been annoying, and they’ve required me to know my machines pretty well to deal with them. So I can totally see someone at GF saying, “Hey, we have this high-precision camera for doing autofocus, why don’t we use it for setting our zero position as well and get rid of a bunch of components.”
come back to the forums and increase the available accumulated knowledge.