An ARM-Based DSP Modelling Synth

June 14, 2014 by Brian Benchoff 42 Comments

synth

The great analog synths of Moog, Oberheim, Sequential Circuits, and more modern version from Doepfer are renouned for their sound, the sheer majesty of a rack full of knobs and plugs, and of course the price. Analog synths are simply expensive to build, and given that aficionados even scoff at digitally controlled oscillators, require a lot of engineering to build. [Jan]’s DSP-G1 isn’t like those analog synths – it uses microcontrollers and DSP to generate its bleeps and boops. It is, however, extremely cheap and sounds close enough to the real thing that it could easily find a home between a few euroracks and CV keyboards.

plug The heart of the DSP-G1 is a micro from NXP modeling an analog synthesizer with 15 digitally controlled oscillators with Sine, Triangle, Pulse and Saw outputs, a low frequency oscillator, two envelope filters, and a low pass filter, or about the same accouterments you would find in a MiniMoog or other vintage synth from the 70s. Since this is basically a synth on an NXP LPC-810, [Jan] has packaged it in something akin to a MIDI to 3.5mm cable adapter: Plug a MIDI keyboard into one end, an amp into the other, and you have a synth smaller than the MIDI Vampire, an already impossibly small music creation tool.

[Jan] has a few more versions of his little DSP device with varying amounts of knobs available on his indiegogo campaign. The DSP-Gplug is the star of the show, though, provided you already have a MIDI keyboard with a few knobs for the required CC messages. Videos and sound demos below.

Continue reading “An ARM-Based DSP Modelling Synth” →

Faster Benchmarks With Slower Hardware

June 13, 2014 by Brian Benchoff 22 Comments

hardware

The Bus Pirate is a cheap, simple, Swiss army knife of electronic prototyping, capable of programming FPGAs, and writing to Flash memory. The uISP is possibly the most minimal way of programming Atmel chips over USB, using less than $5 in components. Although the uISP is using a slower chip and bit-banging the USB protocol, it turns out it’s actually faster when operating as a programmer for SPI Flash memories.

Most of [Necromancer]’s work involves flashing routers and the like, and he found the Bus Pirate was far too slow for his liking – he was spending the better part of four minutes to write a 2 MiB SPI Flash. Figuring he couldn’t do much worse, he wrote two firmwares for the uISP to put some data on a Flash chip, one a serial programmer, the other a much more optimized version.

Although the ATMega in the uISP is running at about half the speed as the PIC in the Bus Pirate, [Necromancer] found the optimized firmware takes nearly half the time to write to an 8 MiB Flash chip than the Bus Pirate.

It’s an impressive accomplishment, considering the Bus Pirate has a dedicated USB to serial chip, the uISP is bitbanging its USB connection, and the BP is running with a much faster clock. [Necro] thinks the problem with the Bus Pirate is the fact the bandwidth is capped to 115200 bps, or a maximum throughput of 14 kiB/s. Getting rid of this handicap and optimizing the delay loop makes the cheaper device faster.

THP Entry: SatNOGS

June 13, 2014 by Brian Benchoff 10 Comments

NOGS Here’s an interesting thought: it’s possible to build a cubesat for perhaps ten thousand dollars, and hitch a ride on a launch for free thanks to a NASA outreach program. Tracking that satellite along its entire orbit would require dozens of ground stations, all equipped with antennas, USB TV tuners, and a connection to the Internet. It’s actually more expensive to build and launch a cubesat than it costs to build a network of ground stations to get reasonably real-time telemetry from a cubesat. The future is awesome and weird, it seems.

This is the observation the guys behind SatNOGS have made. They’re developing a platform for a completely open source ground station network, with the idea being people an institutions along every longitude and latitude would build a simple satellite tracking antenna mount, connect it to the Internet, and become part of an open source Near Space Network, capable of receiving telemetry from any one of the small cubesats whizzing around in low earth orbit.

Despite being what is probably one of the most ambitious and far-reaching projects in open source hardware, the design of the system is relatively simple: the hardware is a 3D printed alt-az mount, capable of pointing a pair of antennas anywhere in the sky. The stepper motor driver board is based on the Arduino, and the computer running each antenna node is powered by a BeagleBone Black or a WR703N router. The antenna receiver is, of course, an RTL-SDR dongle, capable of listening to all the common cubesat bands. Even the software is derived from open source projects. Tracking a satellite across the sky can be calculated with GPredict, and the team is working on an observation scheduling and management system that combines multiple ground stations for coverage across the globe.

It’s a great idea, crowdsourcing satellite tracking from people around the globe, and something that could be used by hundreds of institutions lucky enough to launch a small cube of electronics into orbit.

The project featured in this post is an entry in The Hackaday Prize. Build something awesome and win a trip to space or hundreds of other prizes.

Call A Cell Phone, Find A Tent

June 12, 2014 by Brian Benchoff 36 Comments

relay Finding your tent at a music festival is a tricky endeavor – not only are there miles and miles of tents exactly like yours, you most likely have a few beers or other substances in you that affect your sense of space and/or time. [James] came up with a great solution to finding your tent by illuminating it with Christmas lights and a cell phone.

The basic idea of [James]’ build is having Christmas light flicker whenever he calls a phone. One feature in nearly every phone that can be exploited to accomplish this task is the backlight turning on when a call is received. Add a phototransistor, a little bit of circuitry, and some Christmas lights, and you have a fully functioning tent finder.

[James]’ circuit is a simple relay driving the Christmas lights, triggered by a phone right on top of the phototransistor. It’s a simple circuit that can be built on a piece of veroboard, and with a few pieces of plastic forming the enclosure, provides a reasonably bulletproof device that will survive the rigors of a music festival. As a bonus, there’s no need to modify the phone to trigger a remote circuit. Video of the device in action below.

Continue reading “Call A Cell Phone, Find A Tent” →

The Development Of A Lightweight Wireless Protocol

June 12, 2014 by Brian Benchoff 30 Comments

BANO [Texane] had been thinking about how to monitor the state of his garage door from a remote place. The door itself isn’t around any power outlets, and is a few floors away from where his server would be located in his apartment. This presented a few design challenges – namely, the sensor itself should have a wireless connection to the server, and being low power would be a great idea. This led to the development of a minimalist framework for wireless communication that allows a sensor to run for weeks without a battery swap.

The wireless protocol itself is based on a simple key value pair; each individual sensor, coupled with a NRF905 radio, has passes an address, a key, and a value. There are allowances for checksums and acknowledgement, but as the PDF says, this is a very minimal protocol.

With the software out of the way, [Texane] turned to the hardware. The microcontroller is a simple Arduino clone, paired with a radio and a coin cell on a small board. The micro spends most of its time in a low power state, with the sensor, in this case a reed switch, tied to an interrupt pin.

There was a problem with the power consumption of the radio, though: when the short 17-byte message was transmitting, there was a significant voltage drop. This was okay with a fully charged battery, but with a partially drained coin cell, the possibility of brownouts was high. A big cap in parallel was enough to offset this voltage drop.

It’s still a little expensive for an all-in-one home automation and monitoring system, but developing a functional wireless protocol and the hardware to go with it is no small feat. It’s actually a great piece of kit that [Texane] is sure to find a few uses for.

SOAP Drama: Another Go At Crowdfunding

June 11, 2014 by Brian Benchoff 73 Comments

SOAP

SOAP, the people behind what was initially a sketchy Kickstarter that turned into something reasonable is having another go at crowdfunding their touchscreen router with every radio imaginable. This time, however, they’ve crossed all their ts and dotted all their lowercase js to turn what was a very…. odd Kickstarter into something really cool.

The original specs of the SOAP router were impressive – basically, a touchscreen tablet with an ARM Cortex A9, USB 3.0, 802.11ac, gigabit LAN, and every radio module you could imagine. The goal, of course, being a completely open, hackable home automation system capable of talking to Zigbee and Z Wave, and X10 modules, all while being an easy to configure 802.11ac router with a touchscreen. It’s a great idea, and considering you could spend $200 on an ac router alone, without all the radio modules and touchscreen.

Judging from the updates to the original Kickstarter, the SOAP guys have come a long way in three months. They’ve moved away from a custom-designed iMX6 board to a Congatec System on Module in a move that could be described as the smartest move in the history of Kickstarter-funded consumer electronics. They’ve also fixed the Ethernet bandwidth limitation of the iMX6, although there’s no word on how that works.

To be fair, the SOAP Kickstarter should be studied by business students as the exact opposite of how you should run a Kickstarter. When the project first went up, there were inconsistencies that ranged from not having a functional prototype to lifting images from unrelated open source projects. In the past three months, though, it looks like the team has managed to pull something together. Whether or not the SOAP router will see the light of day remains to be seen, but the team is now in a much better position than they were three months ago.

Better Lasing With Pulses

June 11, 2014 by Brian Benchoff 24 Comments

laser

The folks at the Lansing, Michigan hackerspace built themselves a 40 Watt laser cutter. It’s an awesome machine capable of cutting plywood and acrylic, and is even powered by a RAMPS board, something normally found in 3D printers. They wanted a little more power out of their 40 Watt tube, though, and found pulsing the laser was the best way to do that.

Unlike the fancy Epilog and Full Spectrum Laser machines, the Buildlog.net 2.x laser cutter found in the Lansing Hackerspace didn’t use Pulse-Per-Inch (PPI) control until very recently. When a laser tube is turned on, the output power of the laser is much higher – nearly double the set value – for a few milliseconds. By pulsing the laser in 2-3 ms bursts, it’s possible to have a higher effective output from a laser, and has the nice added benefit of keeping the laser cooler. The only problem, then, is figuring out how to pulse the laser as a function of the distance traveled.

To do this, the laser cutter must accurately know the position of the laser head at all times. This could be done with encoders, which would require a new solution for each controller board. Since laser cutters are usually driven by stepper motors controlled with step and direction signals, a much better solution would be to count these signals coming from the CNC computer before it goes to the RAMPS driver, and turn the laser on and off as it moves around the bed.

A few tests were done using various PPI settings, each one inch long, shown in the pic above. At 200 PPI, the laser creates a continuous line, and at higher PPI settings, the lines are smoother, but get progressively wider. The difference between PPI settings and having the laser constantly on is subtle, but it’s there; it’s not quite the difference between an axe and a scalpel, but it is a bit like the difference between a scalpel and a steak knife.

It’s an impressive build for sure, and something that brings what is essentially a homebrew laser cutter a lot closer to the quality of cutters costing thousands of dollars. Awesome work.