I’m 99.7% sure the lid camera employs only digital correction and does not have any sort of electro-mechanical focus…
2 Likes
Understood. I knew from comments that the camera could be focused by messing with the mechanical focus ring (not an approved adjustment). I was confused by a bunch of earlier comments from multiple people where it seemed to suggest that the overhead camera changed it’s focus depending on the manual entry of material height. It might, but I don’t see any evidence that it does. I believe it’s a fixed focus setup. Especially since a few of the early machines had manual focus issues. The S/W for the camera does a lot of distortion correction depending on the material height.
But I could be wrong of course. I did a quick test with small text directly below the head camera. Tried different material heights entries and actual distance tests within the 1/2" nominal range. Didn’t see any change in text clarity at the 500% zoom for any setup.
4 Likes
I recall from my photography days that wide angle lenses have an inherently large depth of field, so focus would be sharp over a relatively wide range, as you’ve discovered.
5 Likes
I know at least one of the multiple people was attempting to provide an easy to understand description of the functional differences between the two operations without diving into the mechanical hows. 
1 Like
It doesn’t change focus, but it does make a change to the way it corrects for barrel distortion when you change the material height. It makes a pretty big difference in aligning your design to the material. I personally still have some alignment issues with materials and design, even when the height is set spot on, but it’s a lot better when the height is entered correctly. I’m sure the newer models are much better at this.
I’ve resorted to aligning my piece to a quick perimeter score on the tape I’ve put on the bed to make sure my piece is where the gf thinks it is.
8 Likes
Thanks. This was my belief from the beginning. Maybe just a semantics misunderstanding. Thought others were suggesting that the overhead camera focuses on the material depending on the material thickness entry. Didn’t want to correct anyone without confirmation. Believe it or not I have been known to be wrong. At least the wife said that.
3 Likes
Man, the same thing happens with me. Once I thought I was wrong, but then I found out I was mistaken.
7 Likes
I could be wrong as well. The elements in the camera could be moving slightly which would change the focal range a slight bit, causing the field of view to be variable, and the camera could just have a large depth of field. Id have to look at a couple of pictures to really say, but Im not anywhere close to a GF for a couple of days. focal range is quite different than focus however, and the only thing that im aware of that really changes focus is the ring that you have to turn manually.
1 Like
Seems like a DOF of 1/2" wouldn’t necessitate an “auto-” focusing mechanism, but I’m amazed it can also focus on the top of the head (which must be a good… 3-4" higher).
Speaking of how much higher the top of the head is in comparison to the expected material height, anyone with a PRU willing to measure from the floor to the head for me? I’m looking for the height from the top of the bottom of the Glowforge to the bottom of the bracket that holds the head. I’m just curious about the maximum height an object can be before it’ll physically interfere with the movement of the head. (Not the maximum height for cutting or engraving, the maximum height something placed in a crumb-tray-less Glowforge can be without worry of the head colliding with it.)
1 Like
I think you will probably find your answer in what @Scott.Burns said. That wide angle of a lens must by necessity have a large aperture so enough light can get through to allow you to see what’s on the edges. That tends to lead to a very large depth of field. I don’t know if that could account for the entire distance from the top of the head (logo) all the way to the bed of the laser, but I bet it could get close, especially with a little help from the software.
2 Likes
Earlier it was said that the lens was, like, totally super-fancy and stuff. Although I’m not convinced it needed to be, I am still amazed that they have made it work. DOF increases as the aperture gets smaller. Perhaps they just have a pinhole aperture in their camera.
I hope they aren’t “focusing” with some kind of software thing. I do recall seeing an Adobe demonstration where they were working on a technology that would let you “focus” a blurry photo, but I’m pretty sure they were using some kind of black magic to do it. I’m not sure if it’s made it into any of their products yet. I bring up all this because I’m not so sure that “correcting” focus is something that can easily be done with software at this point.
2 Likes
The head hangs down lower than it’s mounting bracket, so assume you want the bottom of the head. And it’s not really the head that you need to worry about, it’s the air assist port that is much, much lower. The bottom of the head is just over 1-1/4" (31mm) to the top of the crumb tray. But the air assist port is only ~0.51" (13-14mm) above the crumb tray.
Grossly rounded measurements… Air Assist port to bottom ~1.91" (49mm) Head to bottom ~2.77" (70mm)
You caught me in a good mood. Generally won’t answer questions based on general curiosity.
3 Likes
Don’t believe they are focusing the overhead camera at all, mechanically or with S/W. Just doing some fancy S/W lens distortion corrections to make the bed image the correct size and rectangular for design placement.
3 Likes
Here’s a sample DOF calculation using a common 1/3" CMOS sensor style camera, the 2mm Focal Length is common length for small fisheye lenses of this type. Using a smaller f/16 aperture puts literally everything into focus and no focusing mechanics would be needed. Increasing the aperture to f/8 changes the nearest focus to 1.8" and the furthest focus to 8.8", and that’s still probably more range than needed.
Makes sense why there’s so much internal lighting though.
4 Likes
Cool! Is that calculator online?
Ope, I think I found it…
http://www.cambridgeincolour.com/tutorials/dof-calculator.htm
Here’s another…
This one also gives you the FOV angles (and dimensions at a distance), which is nice.
4 Likes
LOL, you are absolutely right, I said that totally backwards. Needless to say I haven’t been spending much time with my camera lately 
Oh, well, if that’s all I got wrong today, I’ll take it!
2 Likes
LOL for me as well… I honestly didn’t even notice that… 
(I had to go back and reread your post up there a couple times to see the problem.)
I think I was focusing on the “enough light” part of what you said (which is totally right) and didn’t notice the “see what’s on the edges” part. And the fact that I said the opposite as a response was more of a coincidence than anything else.
1 Like
Hahaha! Well, hey, the larger number IS a smaller aperture so depends whether you’re talking about the number or the aperture itself. OK, technically it is a fractional number, where the “f” is the focal length of the lens, so F/16 for 2mm lens is .125mm. Yowza! That’s tiny!
1 Like
No worries! In practice we handle this on a case-by-case basis.
This was kind of awesome when it happened. Pictures of problems make our lives much easier. 
Great suggestion for the hopper! (cc @tony)[quote=“rpegg, post:19, topic:7739”]
is the overhead camera actually refocused or is the lens a fixed focus?
[/quote]
The lid camera lens is fixed focus.
The lens should be glued in place. You can permanently destroy the camera by adjusting it (depending on how well stuck it is), so it’s something we only do if it’s critical.
Correct.[quote=“Hirudin, post:29, topic:7739”]
I’m looking for the height from the top of the bottom of the Glowforge to the bottom of the bracket that holds the head. I’m just curious about the maximum height an object can be before it’ll physically interfere with the movement of the head.
[/quote]
The lowest point is actually the air assist fan, tucked away behind the head, not the carriage plate directly under the head.
…wait, what he said.
11 Likes