Transister analog synthesizer
The Transister chassis with one of the voice boards but no front panels
synthesizer project has been in development for several years.
I wanted a synthesizer that was quick and fun to create sounds on but had more sonic possibilities than commercial pre-patched synths like the Minimoog or even the ARP 2600.
Experiences with the Bergfotron modular made it clear that a traditional modular synth was far too slow and tedious to use for my taste.
The patch cords had to go. Instead, a system was needed that made it much quicker to create patches but still allowed for more than the traditional VCOs > VCF > VCA topology.
Many say the patching possibilities should have as few limitations as possible. My philosophy is the opposite: I want to eliminate as many useless combinations as possible, to make it easier to find the patches that I can use to good musical effect. This will have to be based on some assumptions. Firstly, assumptions on what kind of sounds I want to create.
This synthesizer is designed to create wind instrument and lead type sounds. For pads and other sounds, that usually require polyphony anyway, I have other synths already. And for weird noises, I might have to use something else, if I want to create them.
The sounds have to be expressive. There should be midi control over more that just pitch and gate. The synth is going to be optimized for playing from a wind controller but also playable from a keyboard or midi sequencer.
The synthesizer has to be easily portable. The Bergfotron modular is so heavy that I can lift it only with great difficulty. This synth needs to be quite a bit smaller and lighter.
the requirements defined, it was possible to decide what functions
should go into the Transister and which have to be left off. This
process, however, has required a lot of thought and has taken a long
Firstly, I decided on a tabletop case. I had a pair of wooden ends from an earlier abandoned project. These were found suitable.
I decided not to use any wood in the chassis, except for the endbells. Instead I used a combination of aluminum profiles and small brass parts. These are being held together with M2 screws. This turned out to produce a rigid and surprisingly lightweight chassis.
All front panel components are mounted on circuit boards. While this requires four large boards, it makes for a fairly lightweight and, most importantly, reliable design.
The synthesizer voice circuitry is concentrated on two large boards. Again, this saves weight and increases reliably compared to having one board for every module.
I had planned to use a selection of my already designed and tested
AMORE boards for this synth. This
turned out to be more space-consuming than I liked.
Instead, I decided to integrate a selection of the already tested circuits on larger boards. This made it possible to try out different combinations of circuits with the AMORE modules I already had, before committing them to the new boards. It also meant that the circuits on the new boards would be already fully tested and optimized. So less work than doing circuits I hadn't done before.
In the process of testing with the AMORE boards and thinking about the requirements, I realized that I needed a number of additional functions which weren't on the existing AMORE boards. This made it necessary to develop and test these functions. The best way to do this was to make some new AMORE boards with these new functions. This way, I could test the new functions thoroghly and decide if they really merited inclusion in the new synth.
The Transister with two of the front panels in place