I have the LightBurn 8MP-120W but can’t seem to get the right focus or resolution. I can move its bezel to change the focus but it’s less than what I’d hoped for. I did some calculations and realized that the camera may be mounted too far from the bed. So I placed an object at the ideal location from the lens to see if the resolution/blur was any better. It did not. Still can’t pick up on fine details of things I’m trying to get an image of to trace. Namely, doing handwritten recipes on cutting boards.
I have a large CO2 that’s 1000mm x 600mm in size with the camera mounted in its lid at 21.25" above the focused bed. I use a notched rod/stick to be certain the lid & camera are at the same location everytime. That bed height is the same as when I calibrated the camera. BTW, during calibration, my 1, 2, 3 and 4 were just as blurry as the image showing in LightBurn’s own calibration video. That was what I thought was the norm for the camera… somewhat blurry, i.e., low rez. My calibration resolution showed it was at 3264 x 2448.
I can make a mark on the center of the object (on tape of course) with a pencil and there’s no way to see it in the image provided by the camera while zooming in close. I’ve had to use a fat permanent marker to place a dot on the tape before I can see anything.
Large printed letters will be somewhat cleaner but there’s still a lot of jaggies around the edges.
I did open the camera in the Windows camera app and checked out the brightness, selected the resolution and other trinkets the app had to offer, but nothing helped when back in LightBurn.
My thoughts are getting a higher resolution camera would help with the fine lines for a laser of this size. The question is, do I get an Auto-Focus camera or one that’s fixed since my object’s surface height will (or at least should) always be in the best focus concurrent with the laser’s focus.
Have several models in mind but the Arducam Wide Angle 16MP is the only one I see at the moment that has a fixed lens. I’ve read another thread here where the poster says get this camera with that lens and other recommendation I don’t fully understand.
As stated earlier, my calibration was 3264 x 2448, but the Arducam can go to 4656 x 3496 for either the Auto or Manual focus model.
Hi Steven,
I don’t think you are going to find a camera that is going to give you a high enough resolution to produce a good trace on anything that small.
You have a few other options you can use, 1is to photograph or photocopy the image you are trying to replicate and save it as a pdf file . I have been asked to engrave signatures in the past and scanning them and retracing it in Lightburn has worked well however I still had to use the node editor to clean up the trace.
Cheers,
Rob
I wondered about that. Currently I’m using my scanner to get the fine details in a higher resolution when needed.
I haven’t had to do the print and cut yet but I’m thinking ahead for registration issues if that’s ever needed. As it is, I don’t think I could even zoom in to a sharp corner (or detail) reliably enough to trust trying to make a repeat cut if the object has been moved.
I honestly don’t know if it’s low resolution or out of focus. It’s just to pixelated to tell.
I read a post previously where the person was doing some calculations on a set length by counting the pixels over that length to arrive at a resolution they could live with. This could be a benchmark for me so I’m going to try that.
Would be nice if others would do the same and post their results here. Plus provide their resolution they were shown during their calibration if they noted it.
I use a red-dot pointer carefully tweaked to be exactly on the laser’s axis when the head is focused. Before doing anything important, I verify the alignment by firing a test pulse off to the side.
Jog the dot to the first target using the Ruida’s front panel, click on the target in LightBurn, repeat for the second target, and done.
The camera is invaluable for eyeballometric alignment around a millimeter or so, but not to the precision required for good P-n-C results.