Progress:
We’ve done quite a lot since we last filled you all in, so here’s a recap of what’s happened:
After receiving the Turing and Lightbar boards from the fab house, we populated and tested the boards. To our great satisfaction, they pretty much worked! Zach caught a few software bugs here and there, like the initially reverse-biased 7-segment LEDs and programs running off the end of the tape, but Zach quickly squashed those out of existence. Right now, Zach is working on the programming interface for Turing - or how you’ll write your state machines to run on the Turing Machine. It’s coming along nicely, and he already has a few state machines running, including the busy beaver! Something you might not have known about Turing is that we have included - in hardware - the option of programming your kit wireless-ly [optically]. We’re really looking forward to try it out, but it’s pretty experimental from our point of view.
The oloduino has come quite-a-ways! After going through two revisions, Zach and I have nearly worked out all of the routing mistakes and software bugs. If you don’t know already, the oloduino is an arduino, a power supply, an ICSP programmer, an FTDI cable, all in one! You can break apart the board along its score lines to separate the programmer/ftdi cable from the power supply and microcontroller platform. This allows you to keep your embedded projects small, light, and inexpensive. The most trouble we’ve had with the oloduino was in the programmer/USB to serial [UART] bridge aka USBER. Mainly on the software side - in particular with device recognition and platform independence. Although we’re not quite finished with the USBER, we already have it working as a USB to serial bridge for Typo - Josh’s wireless bidirectional typewriter. He even used USBER to program my microcontroller in the typewriter!
Fungen has gone through many, many revisions - mostly because as we were designing and analyzing one topology, Josh would learn a new analysis technique that allowed us to simplify the circuit while boosting performance! After literally tens of topologies, Josh came up with an extremely simple, functional, and inexpensive topology that is still in development but looks very promising. Simulations that we ran today make it look like we’re going to be able to provide signals higher than 100KHz, which is very exciting for such a simple topology. If you didn’t know, Fungen will provide (at least) saw, triangle, square, and sine signals with an adjustable amplitude and offset. In addition, the frequency will be set with an input voltage (either from a potentiometer on board or an external source), allowing daisy-chaining to make your own synthesizers! Right now we’ve decided that a split +/- 15V supply would make Fungen more flexible for those looking to have more than just a simple synth, and makes the topology easier to understand. Here’s what the VCO stage looks like:
To the left is a 170KHz sawtooth waveform generated by the push-pull constand current sink-source topology to the right. Josh is planning on implementing and testing Fungen next week, so the next post should be exciting!
Ampliflier’s topology has gone through a couple of revisions and has settled to a standard Class-D power amplifier with a Class-A/B opamp preamp. Since the power amp should be able to handle audio, we need to switch extremely fast (above 100KHz) in order to prevent aliasing while keeping component cost low. With the new Fungen topology, we’ve attained super-100KHz sawtooth waveforms, which we can use as an easy-to-filter switching frequency. Though - the closer to 400KHz, the more filtering we can get with smaller components, the less expensive the filter will be!
So there you have it - feel free to let us know what you think, give us advice, or just check in with us at any time!
-Josh and Zach
