One of my hobbies is hardware development. I enjoy to go through all steps from the first idea to design a circuit and finally design the PCB, populate it with electronic parts and finally get it running after one or more debugging cycles.
With now 6 Raspberry Pis lying (hanging) around inside and around my server rack I’ve started to look for a nice and clean way to mount them in my server rack. On thingiverse I’ve seen a few 3D printed Raspberry Pi brackets and I immediately liked the idea; but as I wanted a simple design with the ability to add a heat sink, I had to start from scratch. Also, at the time I started the design, there weren’t any brackets for the newest Raspberry Pi 4. So, the result of my effort can be seen here:
After a about half a dozen iterations over the last two months, the following 3D print is ready to be equipped with heat sinks and mounted in my server rack:
I’ve prepared two versions of the RasPi Rack holder, one for the the older Raspberry Pi 3 Model and one for the Rasperry Pi 4. The STL files for 3D printing are available for download and licensed under CC-BY-SA 4.0: RasPi_RackMount.zip
The heat sink was bought on ebay (link) and is the upper half of a passive cooled aluminium Raspberry Pi case.
The 2 HE rack panel was bought at Musicstore.de, but I’ve also seen them on ebay and Amazon. They also sell small panels to cover 1 or 2 empty slots, if necessary. The 2 HE panel can take up to 10 modules (i.e. Raspberry Pis).
The RasPi’s are currently all powered from the back side, each with their own power supply. I’m currently planning on inserting a 5x USB charger into one of the module slots to power up to five at the same time.
That’s what happens when you’re drinking a few beers and being in a nostalgic mood: Already over a year ago I had designed a generic Covox Breakout Board (Wikipedia link) based on an oldwell established design.
The whole design is based on two TTL ICs (74HCT373 and 74HCT164) and an R-2R resistor ladder (7.5k and 15k Ohm). The result is a simple DAC with 8 bit parallel and serial input and a few control pins. I’ve uploaded the PCB (back then I still used Eagle Cad) and the Gerber files to my Github Covox Repository.
Sure, I could have taken a cheap audio DAC, address it for example via SPI, and let the dedicated chip do all the heavy lifting. But that would not have been half of the fun of designing the board and (bit)banging the audio signal… 😉
The design is pretty generic and when I built it I thought I could use it in combination with a Sharp PC to generate audio output. I started with a few lines of code, but in the mean time other private tasks became more important. So I’ve reduced my testing of the Covox card to connecting it to an Arduino board and output simple (square and saw tooth) waveforms on my oscilloscope. Maybe if there is enough interest I will try to create a video with some mod-tracker like audio output…
I finally completed the free USB IR Toy v3 PCB I got over a year ago (May 2015) from Dangerous Prototypes.
It took so long because I had to order some of the parts from Digi-Key – and I wanted to wait until I’ve a longer list of parts to order.
I’ve used a PicKit 3 to program the PIC microcontroller. The trickiest part was finding the setting to power the USB IR Toy with the programmer. (I could have powered both devices via USB, but only had one appropriate cable at my hand at that time.)
So far I’ve only verified that the USB IR Toy is detected as serial device and shows its version number in a terminal window. It looks like the build was successful… 🙂
Half a year ago I’ve started to use KiCad for new PCB designs I’m working on. I already wanted to try out KiCad for quite some time. Its release 4.0 and the latest changes in EagleCAD (annoying ads and recently being bought by Autodesk) were enough pressure to switch. And what should I say: after dealing with the rather unhandy library management and some cryptic error messages I really now enjoy KiCads workflow.
Exactly 30 years ago a great disaster struck the region of Chernobyl: a nuclear accident occurred that released a large quantity of radioactive particles into the atmosphere. And it is only five years ago that, with the Fukushima Daiichi nuclear disaster, a second similar catastrophic event has taken place.
These anniversaries did not directly let me build a PIN Photodiode based Geiger Counter, it is more or less a coincidence. The main drive to build such a device was my curiosity and (please forgive me) a fascinating green glow I’ve seen on various fluorescent Uranium minerals under UV light. But in this context it should not be forgotten that at present there still is a significant increase in background radiation in some regions and some agricultural products due to these events.
There are lots of descriptions of how to build such a device; even cheap commercial products (e.g. the Smart Geiger) use such a design. Especially two sites caught my attention: OpenGeiger.de and Das Gammastrahlen-Mikrofon (German). The design I’ve chosen is based on these sources but I’ve begun to further modify it. In this post I’m showing the design I’ve started with. It mainly relies on two BPW34 (Vishay Datasheet) photodiodes connected in parallel, and two transistors to amplify the voltage fluctuations of beta and/or gamma rays striking the diodes. A 9 Volt battery was added to increase the pulse height.
The common approach to protect the photodiodes from light is to use one layer of tin foil and connect it to ground. This should also protect the circuit against electromagnetic radiation. I’ve started with something different and dipped the diodes three times into liquid rubber (Plasti Dip). My hope was to at least allow some beta particles to reach the semiconductor material.
So far I’ve tested the basic design shown above and had mostly noise on my microphone input. I’d say that sporadic crackling has more to do with the 1 hour hacked together design than beta or gamma rays. The liquid rubber seems to block of light, but the simple design is sensitive to electromagnetic radiation. Waving your hand or even movements in about 1 m distance is visible in the sound profile. An additional tin foil shield connected to ground did not change the noise profile, although the EMR influence was reduced. I’ve tested it with two different sound cards (microphone inputs).
I’m currently redesigning the whole approach and expect better results. So stay tuned…
I’m currently trying out a lot of new software and hardware. My next project will not only be designed with KiCad instead of Eagle CAD for EDA. Its basis will be an ARM microcontroller (probably a STM32F103) instead of an Atmel AVR.
I’ve already set-up a respective GNU/ARM toolchain (with Eclipse as an IDE) and OpenOCD for programming and debugging. The typical “Hello World” LED blinkie code was working within less than half an hour… Nice. 🙂 So no comes the real work…
This is partly a review of the Dirty PCB manufacturing service (“Dirt Cheap Dirty Boards” as they call it) as this is my first order. I usually let my boards being manufactured at Seeed Studio or iTeadStudio, but hearing a lot about Dirty PCB lately made me curious and so I ordered a small RFM26W breakout board.
They provide all you need (including Design Rules and CAM export) for Eagle CAD on their web side. For $14 you not only get around 10 boards (12 in my case) but also 6 different colors to choose from.
The connector fits good enough for my purposes. If necessary removing a bit of the plastic case left and right of the pins improves the connectivity as the replacement connector is a bit broader.
There is also a version with mounting holes (HIF6B-60PA-1.27DSL) which I will also try to get my hands on (currently not in stock).
My original solution was to use a 2×30 1.27×2.54 pin header as shown in this post, but the narrow space between the pins led to serious constraints in designing a new interface board (more about that when it’s ready).
Two years ago I made a rather simple circuit board to be able to program Atmel ATtiny microcontrollers with an Arduino board as ISP. I shared my excessive PCBs and made the design open source. The design has proven to be pretty successful, and I was asked multiple times to make a more flexible follow-up version. So I recently started to design the revision 2.0 which should combine ISP and HVSP/HVPP features.
This is a preview (i.e. the design is still buggy) but the final goal is to support the default “Arduino as ISP” option as well as HVSP/HVPP programming modes. The new feature can be useful to recover ATtiny (and ATMega) controllers with incorrect (broken) fusebits settings.