I have to give credit for the incredible instructable amandaghassaei posted on laser cut records; this instructable is about my learnings by experimenting with the original code and materials to get better audio quality.
I had intentionally started this project to try to achieve similar results using a CNC router so not only could I work in the Y and X axis but also the Z, giving the record additional information so the quality would be better. Unfortunately I was never able to generate a 3d file with all 3 axis, so I decided to get better audio quality by simple tweaks of amandaghassaei’s code.
Step 1: FIRST ATTEMPT
To start this instructable you must first follow amandaghassaei’s step 7: Make your own. This instructable only concentrates in the last aspect of this step, the processing sketch.
For starters I worked in my university’s digital fabrication workshop where I used a Universal Laser System VLS6.60 with a 60W laser.
My first attempt was with eucalyptus plywood of 3mm. The song I chose was Ain’t No Sunshine by Bill Withers, not only because it’s an awesome song, but because it only last 2 min. therefore less time cutting with the laser (it’s really hard to get an hour in the laser in my Uni., its always occupied), and as amandaghassaei stated out, her code only allows a 3 min song fit in a normal 12” record. Another important aspect is that it was a 45rpm record (I’ll get to that later).
Generating this file was extremely difficult for my computer, and opening the generated pdf was as difficult as generating it, this was due to the size of the file. To make the file smaller I had to divide it in many different files, this option was in the processing sketch, where you can assign the number of groves (the rings or spiral of a record) per file, in my case I used 5 grooves per file, and there were some pros and cons to this. It was favorable that the files were being generated and easy to open but in return I now had 19 different files making it hard to keep track and being organized. In the end I made an Ilustrator file with 19 different layers. This was the only thing I changed for this first print.
int numGroovesPerFile = 20;
int numGroovesPerFile = 5;
Working with the laser was also pretty difficult. It was impossible for the printer to process all the grooves at once from the illustrator file (19 layers each with 5 grooves), in the end I had to print layer by layer, to make the printed file the lightest possible. It was also much slower than I thought, I was only able to print half the song with the time I had.
The end result wasn’t as great as I had hoped. There were two major flaws in this first print. One, the eucalyptus wood, it’s not a great material to start with, it shreds and when I was printing I noticed that one side was slightly tilted upward, leaving less distance between the wood and the laser, this side was notably more burnt. Secondly the space between grooves was minimal, this caused that sometimes two grooves would interlap. When the record was played a lot of grooves were chipped by the needle, but the song was recognizable and even after the chipped grooves it still played well.
Step 2: SECOND ATTEMPT
For my second attempt I decided to change to 3mm acrylic and the space between grooves, from 10 to 15.
float spacing = 10;//space between grooves (in pixels) int numGroovesPerFile = 20;
float spacing = 15;//space between grooves (in pixels) int numGroovesPerFile = 5;
This second attempt was in the end my best attempt out of the three I did. I had asked for more time on the laser after my experience with last time, I was able to print all the grooves in roughly around 3 hours. I also changed the properties of the laser so it would print as fast as possible with the depth I needed for the needle to pass through.
I noticed that the outer grooves of the record sound much better than the inner grooves, this is due that it takes the same amount of time for an outer groove to complete a cycle than the inner grooves (45rpm), but there is a much larger path for the outer grooves to make a whole cycle which translates in more information, thus better audio quality.
Step 3: FINAL ATTEMPT
Giving it some thought there were two possibilities to pass more information through the needle, thus making a better audio quality. One, make the record a 78 rpm record, as it would spin faster more information would pass through, thou this would also translate in a loss of song time from 3 min to around 1 min, and another inconvenient was that my turntable only played 33 and 45 rpm. The second option, and the one I choose, was to modify a 12” record, make it slightly bigger to have larger grooves in the beginning and the inner grooves not so small.
I modified the Processing sketch to make a 12” record into a 13.5”. The inch and a half was the space I had to enlarge the record without touching the base of the arm of the needle. I also changed the space between grooves again up to 17. I used a transparent 2mm acrylic.
float spacing = 10;//space between grooves (in pixels) float diameter = 11.8;//diameter of record in inches float outerRad = 5.75;//radius of outermost groove in inches int numGroovesPerFile = 20;
float spacing = 17;//space between grooves (in pixels) float diameter = 13.5;//diameter of record in inches float outerRad = 6.50;//radius of outermost groove in inches int numGroovesPerFile = 5;
In theory this should be the best audio quality out of the three attempts, unfortunately I encountered some problems while I was printing in the laser. Midway through the record someone accidently tripped and unplugged the laser, the printer couldn’t start off again where it had left off, so I decided to just cut it and have half the song in a better quality. The laser I was using was going to take some time in rebooting and the other laser in the workshop was unoccupied so I decided to cut it there. That’s where the bigger problem occurred, both lasers aren’t exactly the same, the cut file was displaced half a centimeter from the center, when placed in the turntable it doesn’t spin like a perfect circle, more like an egg, making it impossible for the needle to stay in place by its own. I was able to hold the needle in place for some time and it did sound better, I think…but after my failure and no longer being able to get more hours for the laser in my university this semester I decided to at least make a cool album cover for my second complete attempt, and have a complete product.
To make the album cover I just used two colored cardboard, black and blue, with cutout letters and glued with adhesive spray. It measures 31.5*31.5cm (I cut the letters with a box cutter but if I had the time laser would have been a much better result).
I encourage to whoever wants to take on this project to be extremely organized especially with the laser, it can be really tedious to have everything in order. In theory, you could get the best audio quality by making a slightly bigger record like the one I did, 13.5” and making it a 78rpm, even thou you loose song time you gain some because of the bigger size, good luck!