Hello again folks!
'Fixing' the issues raised has led to quite a major re-design (it has not been as easy as I thought to increase sensing range whilst maintaining linearity and directionality) .. I think I have now solved the problems, but have [b]another question ![/b]
My new design starts with an almost conventional Theremin 'heart' - it has conventional pitch oscillator and reference oscillator driving a hetrodyning demodulator to produce a primary audio source.. The pitch oscillator also drives the processor, which computes a correction voltage that drives the reference oscillator - this is how I now do linearity correction and set the range (number of octaves spanned) tuning etc. This method also means that there is now only a latency (on the primary sound generator) of about 100us.
[b]Question:[/b] - Operator can set pitch to be highest or lowest close to the antenna or furthest from the antenna - this is no problem - What is not so easy to do, is to set the pitch null-point somewhere between.
Do you ever need to set the pitch so that, say, as the hand approaches the antenna, pitch drops to 0, then rises again when the null point has been crossed? I can add this if it was REALLY wanted - but it is difficult, because everything about the auto-calibration works to give maximum playable span.
[i]My plan now is to have a single PCB containing a complete 'basic' Theramin with linear law etc described before, and having a I2C interface for expansion to other elements such as CV, MIDI, and other sound modules - And to sell this as an expandable starter unit / kit.. I have realised that I will run out of money if I must wait for the complete TherAsynth to go into production.[/i]
'Fixing' the issues raised has led to quite a major re-design (it has not been as easy as I thought to increase sensing range whilst maintaining linearity and directionality) .. I think I have now solved the problems, but have [b]another question ![/b]
My new design starts with an almost conventional Theremin 'heart' - it has conventional pitch oscillator and reference oscillator driving a hetrodyning demodulator to produce a primary audio source.. The pitch oscillator also drives the processor, which computes a correction voltage that drives the reference oscillator - this is how I now do linearity correction and set the range (number of octaves spanned) tuning etc. This method also means that there is now only a latency (on the primary sound generator) of about 100us.
[b]Question:[/b] - Operator can set pitch to be highest or lowest close to the antenna or furthest from the antenna - this is no problem - What is not so easy to do, is to set the pitch null-point somewhere between.
Do you ever need to set the pitch so that, say, as the hand approaches the antenna, pitch drops to 0, then rises again when the null point has been crossed? I can add this if it was REALLY wanted - but it is difficult, because everything about the auto-calibration works to give maximum playable span.
[i]My plan now is to have a single PCB containing a complete 'basic' Theramin with linear law etc described before, and having a I2C interface for expansion to other elements such as CV, MIDI, and other sound modules - And to sell this as an expandable starter unit / kit.. I have realised that I will run out of money if I must wait for the complete TherAsynth to go into production.[/i]
