I've worked on the commercial ones using HECAD in UV. UP expand the beam using a specialized scanning collimator, pass it thru a beam shaping mask, then aim into a F-Theta after the galvos. This gives you a very long working distance. You'll need a F-theta or a cheap(er) two lens approximation of one.
~
If some one pesters me long enough, I might be inclined to scan in the basics of scan lens design from a old General Scanning publication. However, I am busy as a one armed,one legged, one eyed, wall paper hanger named "lucky" so you might have to bug me quite a bit.
~
Corrected scan lenses are not that hard to build, but a true f-THETA takes a lotta math. There are some two lens, long working distance, approximations for corrected scanning that work pretty good, they just need a up-collimator ahead of the scan pair and final lens.
~
I would not try to do this device at 30 cm without optical tangent correction... Even with a low cost home-made lens quad, your still a long ways away from a commercial machine's resolution. Commercial machines now use digitally encoded optical galvo position sensors, but old machines with analog optical or capacitive position sensors kept the galvos at a slightly heated constant temperature to ensure sensor accuracy. (For those who are oldies, think GSI-120DT or G330DT galvo, where T means thermal blanket and thermistor inside the galvo)
~
US patent 5048904 is a start, but still does not fully tangent correct across a wide field. Just nobody does it that way anymore, as its hard to change the mirrors on the scanners. Asymmetric mirrors on scanners quickly wear out the bearings, too.
~
And before someone says, "Oh, I'll just correct for that in the software", I'd say don't be a fool... If it were that easy, every time I scrap a old marking system, I would not find huge arrays of the same pre-objective scanning lens scheme, even if the final objective is not F-Theta. The downside is the optics design is somewhat wavelength specific, so buying a 1064 F-theta from Ebay is not going to work that well at 405... Nor is the correction lens fitting into a small space. They always need a long optical bench in front of the scan pair, but after the laser.
~
Thorlabs now sells the near UV correction lens for galvo scanning, but they are 600$ each...
~
Your not shooting for high scan speed, and when you do the math, 12 bits is pretty good for a galvo due to some physics limits(When printing, galvo final resolution is strongly dependent on wavelength) , and 16 bits is often exceptional unless you have good optics. The OLD machines I worked on had fine and course dac chips for each galvo, with the signals summed prior to going to the amp. The scanner amps are not tuned for the very Fast ILDA 12K or 30K PPS in marking or SLA applications. They are tuned for accuracy and that both scanners have the same speed, so that exposures are equal on diagonal lines. Remember, your X and Y mirrors have different inertias in most cases.**
~
That is not hard to do, you just need a 20 Mhz dual trace oscilloscope looking at the position signalsl from the amps, and a square wave generator at 30-60 Hz. You need the scope tuning method for galvo calibration, which is in a Cambridge Technology publication. It will save you huge amounts of time instead of printing test objects. Plan to spend a lot of time printing test objects anyways.. Most older commercial machines have four precisely spaced photodiodes with a pinhole in front of them, to calibrate the scanners, under software control. A fifth PD is mounted on the tank carriage to find focal home in the Z axis.
~
~
~
Before you ask me why all this is done, Think about the following: "Some Times it is Best if the Aquarium Owner does not explain to the Turtle how the Tank Filter works" Dilbert...
You have a lot of reading to do.. But I would not encourage you to quit.
~
Search Keywords: P.J. Brosens, Jean Montague...
~
** Ultimate scanner speed limit is roughly expressed as the maximum specified average, safe, scanner torque divided by the system inertia , which is the sum of the scanner shaft and mirror inertias)
~
~
Steve
Last edited by mixedgas; 03-25-2016 at 08:54.
Qui habet Christos, habet Vitam!
I should have rented the space under my name for advertising.
When I still could have...
3D SLA Printer Build
Reply With Quote
Originally Posted by VDX
I've been looking at the commercial builds like the Lightning II one from Cambridge technology (
