Hardware cassette mellotron

#1

Hey!!

I'm new here!

I still didn't recieved my axoloti, but I allready started to pop some tiny and interesting ideas in my head.

I'm more into electro-mechanical things.

I had an idea, no clue if it will be possible, but I'm naive by nature, so bear with me.

Hardware MIDI controlled mellotron ( based on cassette)

-Record long sustained sounds on cassette (just one note)
-The hardware can be something like a servo (360 degrees), attached to a hacked tape player mechanism (walkman for example)----- here I'll just use the gears and the tape head.
-Make a tape head pre amp for go to the Axoloti audio in, and there apply a vca and a adsr axoloti modules for control the amplitude
- throw some mellow FX via axoloti patches (chorus, vibrato, tremolo, reverb).
-Make dedicated potentiometers for the most revelant Fx parameters, and for the adsr.
-Make a beautiful enclosure.

I know this is a very ambitious project, but I like the challenges.

My actual knowledge is:
Industrial and mechanical design

One axoloti object I guess I will need to create is a thing that relates midi incoming notes to servo velocity (so the pitch of the tape changes)

So if C1 is played, servo goes x fast, If F it's played, servo goes x + a bit more faster...increasing the pitch of the tape.

What do u think...doable?

sorry for my english.

CHEERS !

#2

Hi, this is very interesting topic, something I also want to try myself. I did some research previously and found these :
http://www.crudlabs.org/
http://www.mysterycircuits.com/melloman/olmelloman.html

I think that Axoloti is really a perfect platform to develop such interface, with all the necessary control/effects objects already created.
Hope you successfully make it to final instrument!

#3

Wow!

Didn't know about that!

The only thing that I see, the mentioned instruments don't have an adsr controlled vca.
This is important for being able to place this instruments in a musical context.
Being able to bring the audio from the tape head pre amp---to axoloti in, and then apply there the amplitude control via an ADSR (all of this digitally with axoloti modules), it's much more appealing in a "real world" use a.k.a making music. And then process this inside axoloti with careful patched effects.

What do u think ?

I'll start just making an a examination of the inner mechanics of an old walkman, but I think this it's gonna be the easiest part, even if I scrap the totallity of the walkman, I can design a better mechanic solution for acommodate servo's or even stepper motors (3d printed). The thing that will need more attention i thing it's the play head, this part, obiously needs to be taken from a cassette player.

The thing that I don't have an idea if it can be done it's an axoloti object to translate midi notes to servo rotation velocity. In my mind it seems simple, but who knows...I don't have programming knowledge.

#4

You can drive two standard servo's with the gpio/out/pwm t4 servo object, but you need to modify your servo motor from position control to speed control/continues rotation.
I'm not sure the rotation of a hobby/rc-grade servo is stable enough to drive audio tape, I imagine it 'd have a lot of jitter. Stepper motors tend to have a lot of step-frequency vibrations, you'll find many floppy-disk/scanner/matrix printer synth projects on the 'net, plenty of noises before audio tape is involved...

Some possible variations on this theme: optical discs/photodetectors: http://www.umatic.nl/tonewheels_historical.html
or maybe a floppy drive mechanism can be used to play/record analog waveforms on floppy disk tracks, the floppy disk is like a multitrack tape loop.

I'm sure a lot of fun can be had with hybrid digital/analog/mechanical synthesis!

#5

Hey!

Thanks for the suggestions!

I worked a lot with stepper motors (i designed the high resolution stereolitographic 3d printer Stalactite3D ( https://www.youtube.com/watch?v=lqQLZ1ZW_ZI ), and the steppers in tandem with a proper coupling, there's no virtual jitter. If they are used for give motion to a well placed gear system for make the cassette spin, mechanically can be very precise. They key it's the microstepping (resolution), second order damping and precision sine/cosine current references .... and the motor doesn't need to much torque....so a very nice candidate.

I guess servos are equally useful, it depends of what mechanism u implement for transmit the motion to gears that drives the cassette.

I'll do a mechanical study about a cssette player drive and gears, so I can stop specaulating and see if it can be done.

I'll encounter difficulties in the software side of things, but if anyone is willing to help, we can then make the project open source, and start to see lot of "Axolotron's" in people live rigs and studios.

Dreaming is free.

Cheers and thanks for the very interesting answer !

#6

Sounds like a great project, I hope you keep this thread updated as you progress with it. I absolutely love Mellotrons and any oddities related to or inspired by them.

Here's a video and link in case you haven't seen them, and there's schematics and stuff at the link:
http://gieskes.nl/instruments/motor-keys/tape.php

#7

Some time ago I stumbled upon something called the "TRS Drawbot".
The idea is it uses Audio that simulates very similar the signal required to drive a servo. In the "TRS Drawbot" instance, it drives 2 servo's to draw images, but after reading your goal, maybe Axo objects can be formed to do something similar. With the correct LFO etc in place, it seams like something very possible. If you Google, "TRS Drawbot" you can read all about the idea etc from many sites. "Makezine" has a good article on it.