Designing A Single Instruction Computer

June 11, 2016 by 23 Comments

Today’s computers are unimaginably complex, and so complicated it’s nearly impossible for anyone to comprehend everything a CPU can do in excruciating detail. It wasn’t always like this – the early CPUs of the 70s and 80s were relatively simple and can easily be recreated at the individual gate level. CPUs can be even simpler, as [Jack Eisenmann] demonstrates with a single instruction computer, the DUO Compact 2, made entirely out of 74-series logic chips and a bunch of memory.

[Jack] has a long history of building strange computers out of individual chips, including a TTL logic CPU and a significantly more complicated single instruction computer. The latest, though, is as simple as it gets. It’s just twenty chips, capable of calculating prime numbers, sorting strings, and everything else a computer is able to do.

With every one-instruction computer, there is the obvious question of what instruction this computer uses. For the DUO Compact 2 it’s a single instruction that accepts three arguments, A, B, and C. The instruction copies a byte from A to B, then jumps to the instruction at C. Is it even possible for a computer to add two numbers with this instruction? Yes, if you have massive look up tables stored in 2 Megabytes of Flash and 512 kB of RAM.

In the video below, [Jack] goes over how his tiny computer works and demonstrates prime number generation (it’s slow), string sorting (also slow), and displaying ’99 bottles of beer on the wall’ on the computer’s LCD. All the files to replicate this computer are available on [Jack]’s webpage, along with an emulator in case you don’t want to break out a breadboard for this one.

Continue reading “Designing A Single Instruction Computer” →

The Almost Useful Machine

June 11, 2016 by 26 Comments

[Alex] is no stranger to making machines of negligible utility. A few years ago he made the Almost Useless Machine, a solar-powered system that cuts through a 20mm dowel rod while you wait (and wait, and wait). Enamored by the internet’s bevy of powered hacksaws, he sought to build a sturdier version that’s a little more useful. Approximately five months of free time later, he had the Almost Useful Machine.

It runs on a wiper motor and a recycled power supply from a notebook computer. [Alex] rolled his own board for controlling the motor with an ATtiny25. The circuit turns potentiometer movement into PWM, which controls the motor through a MOSFET. After the cut is finished, an endstop microswitch  immediately cuts the motor.

Every bit of the chassis is aluminum that [Alex] machined by hand. Don’t have that kind of setup? How about a powered hacksaw with a 3D-printed linkage? Make the jump to see it in action, and stick around for the two-part time-lapse build video.

Continue reading “The Almost Useful Machine” →

The Raspberry Pi Infinity+ Is A Fully Functional Huge Raspberry Pi

June 11, 2016 by 9 Comments

It wasn’t an easy weekend for the rest of the world’s hackers and makers, that of the Bay Area Maker Faire. Open your social media accounts, and most of your acquaintances seemed to be there and having a great time, while the rest were doing the same at the Dayton Hamvention. Dreary televised sports just didn’t make up for it.

MCM Electronics had the Maker Faire booth next to that of the Raspberry Pi Foundation, and since they needed both a project to show off and a statement item to draw in the crowds, they came up with the idea of a 10x scale reproduction of a Raspberry Pi above the booth. And since it was Maker Faire this was no mere model; instead it was a fully functional Raspberry Pi with working LEDs and GPIO pins.

The project started with a nearly faithful (We see no Wi-Fi antenna!) reproduction of a Raspberry Pi 3 in Adobe Illustrator. The circuit board was a piece of MDF with a layer of foam board on top of it with paths milled out for wiring and the real Pi which would power the model, hidden under the fake processor. The LEDs were wired into place, then the Illustrator graphics were printed into vinyl which was wrapped onto the board, leaving a very two-dimensional Pi.

The integrated circuits and connectors except for the GPIO pins were made using clever joinery with more foam board, then wrapped in more printed vinyl and attached to the PCB. A Pi camera was concealed above the Broadcom logo on the processor model, to take timelapse pictures of the event. This left one more component to complete, the GPIO pins which had to be functional and connected to the pins on the real Pi concealed in the model. These were made from aluminium rods, which were connected to a bundle of wires with some soldering trickery, before being wired to the Pi via the screw terminals on a Pi EZ-Connect HAT from Alchemy Power.

Is the challenge now on for a range of compatible super-HATs to mate with this new GPIO connector standard?

We previously covered the 2012 Maker Faire exhibit that inspired this huge Pi. The Arduino Grande was as you might well guess, a huge (6x scale) fully functional Arduino. In fact, the world seems rather short of working huge-scale models of single board computers, though we have featured one or two working small-scale computer models.

Thanks [Michael K Castor] and [Christian Moist] for sharing their project with us.

New CNC Machine? DIY Machinable Wax!

June 11, 2016 by 25 Comments

The folks at Leeds Hackspace have built themselves a shiny new C-beam based CNC mill. As you might expect everyone wants to try the machine out, but there’s a problem. A CNC machine presents a steep learning curve, and a lot of raw materials (not to mention cutting bits) can be used in a very short time. Their solution is simple: mix themselves some machinable wax from LDPE pellets and paraffin wax, then easily recycle their swarf and failed objects back into fresh machinable wax stock.

Making the wax recipe is not for the faint-hearted, and involves melting the LDPE pellets and wax to 130 degrees Celcius in a cheap deep-fat fryer. They bought the cheapest fryer they could find at the British catalogue retailer Argos, you really wouldn’t want to risk an appliance you cared about in this exercise.

Colouring came from an orange wax crayon, though they note recycling of mixed colours will inevitably result in a muddy brown. The finished mixture was poured into Tupperware lunchboxes to set, and the resulting blocks were trimmed to square on a bandsaw. The Tupperware proved not to have a flat bottom, so later batches were cast in a loaf tin which proved much more suitable.

We’ve mentioned the machinable wax recipe before here at Hackaday, but it’s worth returning to the topic here with a description of it being used in the wild. Having watched other environments get through learning materials at an alarming rate with very little to show for their effort, we can see it makes a lot of sense as a training material.

Hackaday Prize Entry: Automatic Digital Microscope

June 10, 2016 by 5 Comments

Ziehl-Neelsen Sputum Smear Microscopy (ZN) is one of most common methods for diagnosing Tuberculosis. On the equipment side, it requires not much more than an optical microscope, although it still needs a trained professional to look through the glass, identify and count the number of bacteria in a sample. To provide reliable and effective Tuberculosis diagnostic to regions, where both equipment and trained personnel is in short supply, [Rodrigo Loza] and [khalilnallar] are developing an automated digital microscope based on computer vision and machine learning, their entry for the Hackaday Prize.

automated_microscope_detection_1They started out gathering images of Tuberculosis bacteria from the internet and experimented with color threshold algorithms to detect dyed bacteria, as well as algorithms for counting individual and clusters of bacteria. This process alone can, according to the team, take a trained professional 30 minutes or more. A graphical interface highlights identified bacteria and reads the bacteria count.

[Rodrigo Loza] and [khalilnallar] are testing their device at the Dr. Roberto Galindo Teran hospital in Cobija, Bolivia. However, getting access to a lab environment is one thing, and being given access to a steady supply of fresh M. Tuberculosis samples is another. Unable to obtain samples, which they need to test their algorithms on live subjects, they turned to another front of their project: The hardware. In several iterations, they developed a low-cost, 3D-printable kit, which transforms a laboratory-grade optical microscope into an embedded CNC-controlled microscopy platform. Their kit comprises three stepper-motor-based axis for the X, Y and Z direction, as well as a webcam mount. An Intel Edison and a custom, Arduino compatible shield control the system to achieve features such as homing procedures, autofocus and bacteria detection.

The team is currently in the process of refining their bacteria detection pipeline, exploring the feasibility of semi-automated detection methods, machine learning and neural networks for classification of bacteria within the hardware constraints. The video below shows their latest update on the Z-axis of their microscope.

Continue reading “Hackaday Prize Entry: Automatic Digital Microscope” →

GNU Radio For Space (and Aircraft)

June 10, 2016 by 10 Comments

GOMX-3 is a CubeSat with several payloads. One of them is a software defined radio configured to read ADS-B signals sent by commercial aircraft. The idea is that a satellite can monitor aircraft over oceans and other places where there no RADAR coverage. ADB-S transmits the aircraft’s ID, its position, altitude, and intent.

The problem is that ADS-B has a short-range (about 80 nautical miles). GOMX-1 proved that the signals can be captured from orbit. GOMX-3 has more capability. The satellite has a helical antenna and an FPGA.

Continue reading “GNU Radio For Space (and Aircraft)” →

Fixing A Broken CCFL Backlight

June 10, 2016 by 8 Comments

When you work at Tektronix and they make a difficult to refuse offer for their ‘scopes, you obviously grab it. Even if the only one you can afford is the not-so-awesome TDS1012. [Jason Milldrum] got his unit before cheaper, and better ‘scopes appeared on the market. It served him well for quite a long time. But keeping it switched on all the time took a toll, and eventually the CCFL backlight failed. Here’s how he replaced the CCFL back light with a strip of LED’s and revived the instrument.

Searching for an original replacement CCFL backlight didn’t turn up anything – it had been obsoleted long back. Even his back-channel contacts in Tektronix couldn’t help him nor could he find anything on eBay. That’s when he came across a video by [Shahriar] who hosts the popular The Signal Path blog. It showed how the CCFL can be replaced by a thin strip of SMD LEDs powered by a DC-DC converter. [Jason] ordered out the parts needed, and having worked at Tektronix, knew exactly how to tear down the ‘scope. Maybe he was a bit rusty, as he ended up breaking some (non-critical) plastic tabs while removing the old CCFL. Nothing which could not be fixed with some silicone sealant.

The original DC-DC converter supplied along with his LED strip needed a 12V input, which was not available on the TDS1012. Instead of trying to hack that converter to work off 6V, he opted to order out another suitable converter instead. [Jason]’s blog details all the steps needed, peppered with lots of pictures, on how to make the swap. The one important caveat to be aware of is the effect of the LED DC-DC converter on the oscilloscope. Noise from the converter is likely to cause some performance issues, but that could be fixed by using a more expensive module with RF and EMI filtering.

This is not an original hack for sure. Here’s a “Laptop backlight converted from CCFL to LED” from a few years back, and this one for “LCD: Replacing CCFL with LEDs” from even further back in time. Hopefully if you have an instrument with a similar issue, these ought to guide you on how to fix things.