The team sponsored and owned by the Dutch State has released an image of the engine dubbed "deep state" Hexastorm.
The engine is shown below;
A maxon motor drives an octagon prism. The festo tubes are to put air in the air bearing. The copper element is for water cooling of the laser tubes. The above image is from 2017. The engine apparently hasn't changed since then. You can see they use a Spartan 6 FPGA chip.
As cool as the image is, they still have the following challenges; - build a robot to mount the aspherical lenses ( they now have a crazy screw system) - get circular spots They assume special asperical lenses can be made which circularize laser beams. I doubt it and even if it's possible this would be so useful it is a business case in its own right. I would immediately buy these lenses. Formlabs, Envisiontec (Desktop Metal) would probably buy these as well.. - they have to show their engine has a competitive advantage with respect to Hexastorm. I am quite sure a lot of people will prefer not to deal with patent holders. - achieve uniform exposure with all lasers and facets This engine is for sure not able to do that. I tested it.
Hexastorm has the following fundamental advantages; - their engine needs to be tilted, mine does not - my engine is simpler as it uses a single laser bundle. The original engine violates the KISS principle / Occam's razor. - my engine is more cost effective, the price of the actuating unit with prism should be less than 35 euro's per laser.
Anyhow, great they put the image out there. I even saw, one of the original inventors Jacobus Jamar is back on the team. I thought it was nice to share with the folks from Hackaday. As you now start to understand, how I "hacked" and "open sourced" a "deep state" technology.
NWO grants around 250K euro's funding to AMSystems in Takeoff 2 financing. As said, AMSystems works on an earlier version of Hexastorm which uses a plurality of lasers per prism. The idea is that these systems have increased throughput. There are multiple issues with this technology. As you can see in the image below.
The laser spots are not uniform. This can be seen directly. They have not the same power and are "elliptical". The copper elements is used for water cooling. The screws are used for "aligning" the laser. Back in 2017, they planned to fix alignment with a robot. They actually build a robot but I guess they will build a new one.
I have recently ordered 40 pieces of prisms and plan to start selling mounted prism on mirror motors online soon. This allows makers around the world to build their own "deep tech" and "deep state" open hardware laser scanner,
After my last post, @Gravis wondered if it would be any help to have a motor driver that compensated for being slightly out of balance?
In short I think so yes.
The prism rotor is a key component of the laser scanner. It might be good to go over the assumptions surrounding the rotor. The project start with the assumption that it is not possible to get a laser grid. There are ways around this, e.g. laser with a DMD device. There is also research into nano-lasers; which could possible form a laser led screen. All these technologies have their limitations, due to cooling, diffraction or focusing of the laser. A way around this is moving the laser bundle either by mirrors or prisms. Rotating prisms have a lower cross-scan error and require different optics then rotating mirrors. The Netherlands is especially interested in lithography due to the regional importance of ASML. The original prism can be seen in the image below. There are other images but this is the only image in the "public domain". The original prism is relatively large, in the order of 20 cm and rotated using an air bearing. It makes a lot of noise.
After studying the laser market for some time. I realized that if there is market for prism scanner with a plurality of bundles, there must be for a single. The rotor design present in laser printers is thought out extremely well. The current prism dimensions are simply based on these constraints. This probably needs to be changed in the future. I could imagine the prisms are made smaller. This partly depends on manufacturing capabilities of the prism and the desired line length. The prism could be lifted by an air-bearing or electromagnetic field. It could be kept in glass housing with a lower drag gas like Helium. In such a configuration, I imagine that the position of the prism can be optimized during rotation. I still need to look into how prisms are rotated by lasers. But I can imagine that a disk which can be rotated by an optical field can simply be attached to the prism. Another option would be to add coils or magnets to the prism which would allow it to be rotated without contact.
I balanced a prism using the single-plane 2 run method. Let's look at a plot of the polygon motor with AN44000A chip and an unbalanced 30x30x2 mm square prism.
A clear peak is present at 50 Hertz for both the acceleration and infrared spectrum. The rotor is pulsed at a frequency of 10 Hertz. It apparently spins five times faster.
Before adding a weight the following measurements are obtained; Adding a weight at the exact opposite location reduces the accelerometer amplitude. The rotor tends to jump to different frequencies. The NBC31111 is more stable. For a frequency of 30 Hz the NBC31111 needs to be pulse at 180 Hz. The AN44000A needs to be pulsed at 5 hertz to achieve 33 Hz.
Mirror motors have 5 pins; PWM, MOTOR_EN, GND, Voltage and LCK_PIN. PWM is speed control. GND is ground. Voltage is the supply voltage, ideally 24V. LCK_pin is false if encoder and signal are not in sync. I cannot confirm the function of LCK_pin and MOTOR_EN. Changing them or looking at their input does not reveal any activity. I will try to expose with the AN44000A. So far I only exposed with NBC31111.
Finally, Max who bought the first prisms found some interesting links.
Baraja Baraja is building a LIDAR scanner which uses a prism with an uneven number of facets. They do not rotate the prism but change the frequency of the laser. https://www.baraja.com/
Saudi Aramco got granted patent US10018817B2 in 2018, Adaptive optics for imaging through highly scattering media in oil reservoir applications, in claim one the following is read "two pairs of lenses with two galvanometric scanners"
I claim the whole patent but then using a similar system with "two pairs of lenses with two prism scanners" which should allow me to circumvent the patent. Competitive advantages; lower cross scan error and increased throughput.
I started with improving the optical layout of the Hexastorm. Luckily, there is a company from Colombia, cihologramas.com, who made an open-source optical package for optics and also provided some instructions on how to integrate this into FreeCAD. You can seen the result below. I will explain the whole simulation in a different video.
Anyhow, what is even more interesting is that they also seem to user laser writing to a substrate to create holograms. See the video below
They report a resolution of 25400 dpi, which is 25.4 (mm)/ 25400 = 1 micron. Production speed is 55 cm^2 per hour, i.e. a square sided 7.41 cm. These are 74100 lanes with a length of 7.41 cm, which is 1.54 m/s . Most likely, they use a galvo scanner. It might be done much faster with a prism as well, but I would have to recheck the whole design. I will outline in the next video, how you can make a design, in the first place.
I think prisms are suited for this domain. An optical wave is split in two. One part is used to illuminate a sample from the bottom with a plane wave. The other part illuminates the sample at a specific point. The waves interfere and the reflected light is measured by a camera. A prism could be used to to scan this spot. It would require a very fast camera but these do exist.
It can be used in reverse as well for triggering reactions like polymerization. This requires something like a digital micromirror device, these are typically slower. So that might not be a good fit with current technology.
There is a very nice old video on holography on YouTube
I created a new video, where you can see the exposure and the result.
This makes it much easier to see where the project is at. It is really easy to add measurement, see this excellent video.
My current aim is at bringing the module to production. I also plan to add confocal. A unique property of laser scanning is that it can read and write at the same time. This is a real advantage compared to other projection techniques like LCD.
Just a quick update; I am now finally able to make exposures with the new nmigen gateware.
There are a lot of improvements in the new tool chain with respect to ldgraphy on the beaglebone. - Slicing can be done on the Raspberry. - Motion can be done with acceleration/jerk profiles - Steps per scanline can be altered and is not fixed - Whole code base is one language; Python - It is much easier to align the optics;
the design of the laser head is improved and there are algorithms to fix the alignment and calculate; spotsize and cross scan error
I will make a video for the complete process soon.
Hexastorm
The laser spots are not uniform. This can be seen directly. They have not the same power and are "elliptical".
A clear peak is present at 50 Hertz for both the acceleration and infrared spectrum.
Adding a weight at the exact opposite location reduces the accelerometer amplitude.
The rotor tends to jump to different frequencies. The NBC31111 is more stable. For a frequency of 30 Hz the NBC31111 needs to be pulse at 180 Hz. The AN44000A needs to be pulsed at 5 hertz to achieve 33 Hz.
Anyhow, what is even more interesting is that they also seem to user laser writing to a substrate to create holograms.