Jigs / absolute positioning

Been GF’ing for about a month now and I’ve found myself in a jam a couple of times with needing to re-cut/score parts after the fact and have not had the success I’d like to. I’ve found the lack of precision on the camera dewarping to be unhelpful other than to broadly place parts around voids (and even then it fails).

One example, I did a little holiday set piece that had a doormat-like thing. Forgot to set the “Merry Christmas” engrave parameters. Pulled the parts (and the wasteboard) before realizing the text was missing. So, I used the wasteboard as a jig, put the parts in “centered” for the kerf and taped them. I luckily had the file still loaded so I re-run stuff, but could not 0,0 it so everything was off and worse, it had a 2 degree rotation relative to the first cuts. Was it because the crumb tray moved? Was it because I wasn’t quite square to the edge of the CT? Was it because my holddown pins weren’t an accurate guide? Maybe all of the above?

I’ve cruised the forums on jigs, but I haven’t seen anyone address the fact that the crumb tray is not fixed / finding “0,0”. The gantry obviously has a “0,0” and so does the crumbtray, but because the crumbtray has play in it (I don’t understand the design choice), this results in any movement of the crumbtray, or material on it to lose reference.

I have seen the suggestions around 12x20 and that many people use jigs that are 2 edge-aligned on the CT. I tend not to do that since I have seen problems with through-cutting vs. material warp. So, I’m pretty diligent about pinning my material down - at least on 3 edges.

I haven’t double-stuck the tray down as a) it will affect focal height, b) the first time I move it, I will lose reference again.

Love any tips from people who have successfully done multiple passes between lid opens on the same part. Ideally, I’d like to find a workflow that lets me do this arbitrarily and not require jigs. Most of the time that I’m needing this, it’s not a pre-conceived, multiple part thing. For example, last night built an architectural model for a home and decided that I wanted to score the garage door after the fact. It’s a big part with some other detail on it and didn’t want to waste the wood. But, finding a way to get the alignment - X/Y and rotational - is escaping me.

Thanks in advance!

You might be able to find that but it won’t be dead perfect.

Some people have done permanent reference rulers, etc. I find that to be unnecessary, and that jigs are very quick. I juts cut a corner and go for it, it’s rock solid.

As for your tray moving around, get @timjedwards boots going, your tray wont move and its height is not affected. Link incoming.

3 Likes

Boots:

Corner jigging:

As with all jig techniques, how you do it will depend on your requirements in terms of precision. You can get it down to sub-kerf accuracy if you do a kerf corrected jig, but in general 0.007" translational and sub-0.1 degree rotational is plenty accurate.

As Jimmy DiResta says: If it looks straight, it is straight.

6 Likes

Pretty simple, but most folks miss it…if you have to add something to an existing cut and you haven’t moved the material on the bed, when you align the addition, align it to the pink design on the screen without moving the original pink cut lines.

Wherever they appear to be after the cut job, leave them there, and center your engraving or whatever on them. Then set the original pink cut lines to Ignore in the thumbnail column.

(Ignore the fact that it looks like the design shifted on the screen. Do not move the cut lines trying to match it to the post-print picture of the bed.)

When you run the addition, it will align correctly with the cut you formerly placed on the bed.

Try it a couple of times on some scrap to see what happens. :slightly_smiling_face:

You can add whatever you need to after the fact, you just need to align it to the already processed part of the design without shifting anything first.

6 Likes

That statement is incorrect.
Also, use “set focus” every time before you place your artwork and this (assuming your camera is calibrated) will get you within a hairs width of where you need to be everytime.

2 Likes

@beerfaced, this isn’t my experience at all. I always do set height or focus and the extents are considerably off. Here’s an example where I haven’t lifted the lid. I’ve seen other people mention how inaccurate the de-warping is as well. I assumed everyone experienced and worked around the visual drift.

Are you really able to get the four corners of your design to show near-perfectly aligned with your cuts? Maybe there is a tolerance problem on my machine somewhere causing this. I have been caught by several mm drift in either placing new parts on old material or trying to re-hit existing parts.

@Jules - thanks for this, and yes, this technique works… but only if you have not removed the material from the machine or the previously cut design from the GFUI. I’m talking about doing a second round of parts, essentially with a “clean sheet”, or trying to accurately place artwork on an object or sheet that’s been previously cut with higher accuracy.

I think one thing I’m going to start doing is putting all of my artwork on 12x20 templates so that I at least have that as a reference point. The issue with that of course is that I routinely will move parts around to work around hold-down pins or an imperfection in the wood, or to leave added scrap on a design I didn’t gang very well.

@evansd2 - Thanks for the boot reference, I hadn’t come across that. Stabilizing the tray is a huge part of the equation here. I also decided to try using magnets rather than pins so that I can get the material and parts square using two edges of the CT as a reference.

Something you can try, but not ideal if you’ve taken your material up…i run a score of what i’m needing to add using a piece of paper-goes pretty fast. Then i place my material on top of that paper and align that way. It’s more accurate than guessing and hoping:-)

3 Likes

I think there may be a couple of different concepts being interchanged. The short answer is you can’t always trust what you see in the browser, but you can always trust a placement jig.

Focal height, set focus - the laser beam is an hourglass shape with the narrowest point being where the most power is concentrated. For engraves, how far above or below the surface of the material the focal point is set has a detrimental or beneficial effect depending on personal taste and some other technical factors. For cuts, it’s more about how wide the kerf will be and how much charring there will be. Changing the focal height, using set focus, also adjusts the dewarping of the lid camera and makes the view displayed in your browser change. It does not affect where the laser head travels.

Alignment - all about placement of your file on your material. Some machines are better than others, but the lid camera is probably never going to provide pixel perfect placement unless it’s coincidence. For that you need to jig. Using set focus or an accurate manual height entry will help, but perfect is unlikely.

For something you cutout you can always make a placement jig. If you have cut something out and then notice you forgot to engrave/score/cut something place a piece of cardboard in your machine. Cut the object from the cardboard. DON’T do any interior cutting/scoring/engraving - it’s a waste of time and cardboard is too flammable for a bunch of lasing. Remove the cutout piece of cardboard, without moving the larger piece of cardboard. Place the piece you forgot to do something on into the hole in the cardboard. Set focus on the piece and then do what you forgot to do. The image in your browser will show that the alignment is off - ignore it and use the force Luke. I have pretty good alignment on my machine for the initial file placement. After that, all subsequent placements are shifted a bit to the right. The placement on the material, however, is always perfect.

If you do not find this to be true one of three things is happening:

  1. I have not properly communicated how to do a placement jig.
  2. The artboard in your file is not 12"x20" (not sure if this still matters, but it did a couple years ago and it’s just habit for me.)
  3. There is something wrong with your machine.
4 Likes

Why will focal height not change if I add height to the tray? You’re saying that I enter .125" for material height whether i have thick double sided tape on the tray, no tape, or no tray? If the machine relies purely on autofocus to determine height then why am I entering material height for each job and each step?

It uses that entered material height to adjust the view from the lid camera. It needs to know how far away from the camera the material is to de-fish the image properly. It is FAR more accurate to use set focus instead of entering the material thickness manually. Use set focus and allow it to adjust the camera view then place your artwork. If you have performed the camera calibration and have your unit sitting flat or shimmed so that the lid closes properly and fully without touching on either side the placement of the artwork to where it actually cuts should within sub-millimeter.

I’d change that to read that it is potentially more accurate. It’s possible to measure your materials quite precisely.

What the set focus does better is ensures that the specific spot you focused on is as accurately represented by your lid camera as it can be.

Jigs will always be the most accurate way to position materials. The lid camera is at best a good approximation – sometimes it’s accurate enough for your needs, but if you need positioning accuracy to the sub-kerf level, jigs win hands down.

I think we’re talking about different things here. Based on what I’ve seen, the autofocus / broad material height setting is for the lid camera - it just sets the parameter for the de-warping algorithm to display your material and help you align your artwork. This in my experience, no matter what you do is insufficient when trying to deal with extents.

The material height in the per-part settings (cut/engrave/score) sets the laser focus for the operation.

I believe if I change the height of (with tape for example) or remove the tray, I will have to compensate for the focus height. This isn’t a huge ordeal, but I can’t believe that adding some beefy double-sided tape below the tray is not going to affect the laser focus.

Am I mistaken?

True. It also sets material height for cutting.

It does, but it works as an override. If you leave it blank it’ll follow the material height setting or the set focus value.

Yes you will, for both cuts and also for the camera algorithm.

Set focus would neatly sidestep this, as it will measure the height of whatever you put in there. But then so will using the boots from before with no changes to tray height.

Tape is a poor solution for a few reasons. Try the boots, your tray wiggle will be gone. Jigs will still be your best solution for extreme placement accuracy.

1 Like

A) you don’t need to.
B) entering a manual focus height different than the set focus is most commonly done to achieve a defocus effect. Smoothing out an acrylic engrave, getting a fatter score line or better cutting through a thick material with a second/third pass for example

I have found that the thickness of the crumb trays is not exactly the same from GF to GF. My local library has one and it is 1/16" less tall than mine. If I put material that is exactly 0.125 inches thick measured with digital calipers on my bed it will be 0.0625 inches closer to the lens that if I put the exact same piece of material on my library machine. That’s why I believe using set focus is a much better tool than using digital calipers.

1 Like

Each machine is calibrated at the factory so I suspect that that discrepancy is already accounted for. Also, 1/16" sounds pretty extreme, it actually sounds like they didn’t have the tray in the divots or that there was some other mistake in measuring.

Almost everyone’s tray height (tray, not lip) is right around 1.35.

How to properly measure tray height:

I’d be curious whether applying that technique to both machines – being very sure that the tray is properly seated – would yield 0.0625" difference in measured height.

1 Like

We measured the trays out of the machine sitting on a flat marble table with a machinists level from front/rear and using digital calipers from the machined surface on the level to the table.

1 Like

I have no idea what any of that means lol. You sound pretty detail oriented though so I am going to assume that was accurate. If I found my tray was a full 16th off from “normal” I would have asked support about it and pushed for a replacement. That sounds like a defect.