Laser shows are cool, so why not build my own? There are a lot possibilities for building a laser scanners. It’s all about speed and accuracy. Some people use galvo meters, some others use speakers.

My first attempt was to use the computer speakers and control this with my soundcard & a text-to-wave-file-converter. But unfortunately this will only create lissajous figures. So that’s why the second approach will be a galvo meter…
3mW Laser
I’m using a small 3mW red laser (class IIIa) unit which will be sufficient for a small laser show. It has a laser dot of about 3[mm] and requires 3[Volt].

When it’s dark outside it’s still possible to see the red dot on the roofs of my neighbors 330 meters away!! I wonder what they are thinking..
Slow servo scanner
Although the aim for this laser project is a quick laser system, it’s also fun starting to understanding servo’s a little better. I also attached a webcam for future use. Well first back to this laser display system. It consists of:
- 3mW laser pointer,
- 2 cheap servo’s, with (cheap) mirrors,
- some acrilic plates,
- Arduino Uno board & pc connection.
Press the mouse pointer somewhere in the black box and the laser will point in that direcition, projected on a wall. So this works, altough both speed & resolution are very low. I tried to play around with the arduino servo library: write() & writeMicroseconds() commands, but it doesn’t really make a difference. This propably due to the cheap AD-converter in the servo.

Also making a circle on the wall, the output is very course.
Laser XY-Scanners: Using speakers
There are a lot of different “XY scanners” around. Some use speakers or other “coil”-devices. These systems are always open loop systems, so less accurate & relative lower speeds but a more simple way to start with! 🙂 Nice examples from: Chez & others:









While I’m trying to understand my first speaker setup, I discovered that’s actually only possible to great lissalous figures (or Bowditch curves). These figures are combinations of the functions: y=cos(a*t+p) & x=sin(b*t), see example.
That’s not what I had in mind.
Upgraded speaker system
Some time later, I found out, how a simple speaker system could work. It’s well written by Rich Olson (NothingLabs.com), which created also a connection to display Twitter messages. Basically the speaker hardware system is completely replaced by very simple electronics. The negative DC signal is completely removed, so the speaker cone works like a spring & the coil will move the cone by a bigger or smaller current. In this way it’s possible to make other graphs then the lissalous figures.
So this means, the coils will also constantly dissipate energy and everything will warm-up. Secondly this is not a closed-loop system. It is comparable with using a (open-loop) stepper instead of a faster & more compact (closed-loop) servo motor. Another negative side effect, is the sounds it produces.
This is not what I had in mind, so let’s figure out, how the real XY galvo scanners work…
Galvo’s…
Creating quick responding laser systems, galvo drives are used. These are basically quick precision rotation actuators with feedback. Some specifications from commercial players:

Here you see some DIY closed-loop galvo meters, by Norm & Chan:



Feedback
Creating a magnetic shaft & coils is not the hard part, but how to create a feed back system? There are some options:
- Capacitive sensors, like Chan
- Optical sensors, like Norm
- Magnetic sensor, like Francois’s
Chan’s approach
Chan uploaded a complete electronics system for both steering the actuator with PID & reading the shaft position with capacitive sensors:

(need to find the other pics..)
Let’s build something…
Okay, it seems to be feasible to build a high speed galvo meter. Let’s build my own, starting with mechanics. Making the axle as light as possible (minimizing rotational inertia), the radial orientated magnets will be enclosed by shaft ends, which will run inside very small bearings. This setup will be inserted inside a thiny brass hollow shaft. On the outside a nice none-magnetic stainless steel housing is placed.



Results & learnings so far:
- The biggest mechanical issue right now, is both axis work like gears! When moving one of the shafts, the other one is turned also by magnetic forces. Maybe this will be solved, by just change the metal mirror holder part, by some other material, but maybe both magnets are still to close to each other.. (All other materials used, are none magnetic, even the bolts.)
- Secondly, when rotating both shafts by hand, it’s producing a trapezium movement instead of a box. So also alignment of the galvo is important. Most likely one of the mirrors is not glued very well..
Electronics… How to apply the +/- 12 Volts very quickly? And what about the closed-loop side?
Software… There are also some options for software control. Some (expansive) XY scanners use serial connection, others LPT or the sound card. If you are using LPT there is a nice inpout32.dll created by Logix4u.net which works under WIN NT4, WIN2000, WINXP.
For my first approach a simple wave file editor is sufficient (speaker system). I found a nice wave file description @ ccrma.stanford.edu, see picture. Then just create a converter which reads numbers from text file (16bits stereo) and convert this with an adjustable sample rate into a wave file.
The second approach will be a little harder. Options so far: Using the LPT port and Visual Basic 6, is not very fast (output = 1000Hz max) and for a nice ILDA test pattern my aim will be 30,000 points per second 😉 This could easy be done with the soundcard which can handle up to: 48.000 Hz!
In this case, the amplifier & closed-loop system needs to be developed. Since 2012, I’m learning to use Arduino’s, which also give more flexibility.

To be continued…
More information:
Geef een reactie