Re: LaserShark what i have got working.
Posted: Sat May 21, 2016 2:11 pm
Hi guys. First post for me. I'm new to SLA and just doing research. My background so far has been in homebuilt CNC mills and laser engravers. I'm seeing that you are having dropouts in your scan and you are reloading windows and such trying to find the timing errors. In the CNC mill world we had the same issues and it all boiled down to microsecond changes in the time in interrupt as the CPU thread goes off to service another process. I am by no means an expert in the coding issue, but for me it was all solved when I purchased an item called a smoothstepper, that buffers the incoming commands with a DSP and releases them all in a serial steam with a highly predictable timing. My mill now runs perfectly at 10 times the speed I could achieve without the smoothstepper. When I first built my C02 laser cutter/engraver, I was also experimenting with traditional CNC type interface using EMC2 and Mach3, but ended up buying a DSP controller that was built specifically for the task. It seems to me that someone needs to code a DSP specifically for this task and drive the galvos and turn the laser on and off with precise timing. It would take in one page frame at a time and release it with high precision, then while it is doing its refresh thing with the z-axis, it can load up the next page.
I sure do like this project. I still don't know enough about SLA and the right path for my build, but it sure appears that if the goal is putting UV light on the resin, then a UV laser through a galvo should be superior to trying to squeeze the small amount of UV from a projector or from a tablet LCD. With the galvo concept, you are making and using only the wavelenth of interest. Sadly, my experience building my C02 laser engraver won't be of much help here, because it is quite slow in comparison to these galvos. My mirror raster scans across a 30" bed at about 800mm/sec, then has to stop and turn around and advance the other axis by one line. That being said, there are a few related issues like resolution and what happens when alignment is wrong and scan lines overlap.
I did the kickstarter for the TIKO 3D printer, and I hear it will be shipping in a month or two, but after researching SLA, I think my TIKO will be a dinosaur within a year. Perhaps I should sell it on Ebay as soon as I get it. Still unsure. I'll be watching this forum with great interest!
I sure do like this project. I still don't know enough about SLA and the right path for my build, but it sure appears that if the goal is putting UV light on the resin, then a UV laser through a galvo should be superior to trying to squeeze the small amount of UV from a projector or from a tablet LCD. With the galvo concept, you are making and using only the wavelenth of interest. Sadly, my experience building my C02 laser engraver won't be of much help here, because it is quite slow in comparison to these galvos. My mirror raster scans across a 30" bed at about 800mm/sec, then has to stop and turn around and advance the other axis by one line. That being said, there are a few related issues like resolution and what happens when alignment is wrong and scan lines overlap.
I did the kickstarter for the TIKO 3D printer, and I hear it will be shipping in a month or two, but after researching SLA, I think my TIKO will be a dinosaur within a year. Perhaps I should sell it on Ebay as soon as I get it. Still unsure. I'll be watching this forum with great interest!
