Maker spaces and ‘FabLabs’ are popular and accessible in many areas. 3D printers are on the cheap, and powerful laser cutters are in the range of the fearless hobbyist. You can get dirty-cheap PCBs from China in less than a week (ok, probably not right now because of Corona virus) and it is easy to SMD solder parts these days with a DIY SMD PnP machine and OpenPnP. With the right equipment and skill set it is possible to build many cool projects. It is very rewarding and a great learning thing. Blog about it so other can learn too. And it even could get featured on Hackaday.
But: The risk is that someone might send you letter about a ‘Copyright Infringement’. Sadly, this is what happened to me for one of my recent projects. I don’t think that ‘take down’ letter was justified, but I learned a great deal what I should have done differently to avoid that situation. So in the end, it was a learning opportunity, which I believe is worth to share. In essence: what can a maker or educator do?
Hackaday: building a giant meta-clock made of smaller clocks (image: original image from Hackaday)
The tinyK22 board with the NXP K22FN512 is a bread-board-friendly small board with a 8 MHz external oscillator:
This tutorial is about how to use the NXP MCUXpresso Clock configuration and configure the board to the maximum clock frequency of 120 MHz. The same steps apply to many other boards, including the FRDM-K22F one.
Eclipse with its CDT managed Make system makes it easy to build projects, because it can handle a lot of the background tasks and settings between the project and the build setting. It can get a bit difficult if I want to include a library or other sub-source project for which I have to add extra compiler #define or add extra includes path settings.
Eclipse project settings for include paths
This article show the different ways I have found to make such imports (and exports) easier.
I really love clocks. I think this is I am living here in Switzerland. Beside of that: clock projects are just fun :-). After I have completed a single clock using stepper motors (see “DIY Stepper Motor Clock with NXP LPC845-BRK“), I wanted to build a special one which is able to show up to four different time zones: Below an example with London (UK), New York (USA), Beijing (China) and Lucerne (Switzerland):
In “Eclipse JTAG Debugging the ESP32 with a SEGGER J-Link” I used a SEGGER J-Link to debug an ESP32 device with JTAG. I looked at using one of the FTDI FT2232HL development boards which are supported by OpenOCD. The FT2232HL is dual high-speed USB to UART/FIFO device, and similar FTDI devices are used on many boards as UART to USB converters. With OpenOCD these devices can be turned into inexpensive JTAG debug probes. This article shows how to use a $10 FTDI board as JTAG interface to program and debug the Espressif ESP32.
Bootloaders are a fine thing: With this I can load any applications I like. Power comes with some complexity, and a bootloader alone is a complex thing already. But this applies to the application part too: I need to link the application to a certain offset in the memory space so it can be loaded by the bootloader, plus the application typically needs to add some extra information to be used by the bootloader. This article describes how to build a bootloader application with Eclipse (MCUXpresso IDE) using the MCUXpresso SDK.
Stack overflows are probably the number 1 enemy of embedded applications: a call to a a printf() monster likely will use too much stack space, resulting in overwritten memory and crashing applications. But stack memory is limited and expensive on these devices, so you don’t want to spend too much space for it. But for sure not to little too. Or bad things will happen.
The Eclipse based MCUXpresso IDE has a ‘Heap and Stack Usage’ view which can be used to monitor the stack usage and shows that a stack overflow happened:
Heap and Stack Usage
But this is using the help of the debugger: how to catch stack overflows at runtime without the need of a debugger? There is an option in the GNU gcc compiler to help with this kind of situation, even if it was not originally intended for something different. Continue reading →
The Espressif ESP32 devices are getting everywhere: they are inexpensive, readily available and Espressif IDF environment and build system actually is pretty good and working well for me including Eclipse (see “Building and Flashing ESP32 Applications with Eclipse“). The default way to program an ESP32 is to a) enter UART bootloader by pressing some push buttons and b) flash the application with ESP-IDF using a USB cable.
That works fine if the ESP32 is directly connected to the host PC. But in my case it is is behind an NXP Kinetis K22FX512 ARM Cortex-M4F microcontroller and not directly accessible by the host PC. So I had to find a way how to allow boot loading the ESP32 through the ARM Cortex-M which is the topic of this article.
The new semester is approaching in a very fast way, and so is the new lecture and lab module ‘Advanced Distributed Systems’ at the Lucerne University. For that module we are going to build a new ‘Sumo’ style robot with WLAN capabilities using the ESP32 chip. It will be a new robot PCB, and below is the current robot (based on NXP K22FX512) with the WLAN module connected to it:
In my previous article “Seeed Studio Arch Mix NXP i.MX RT1052 Board” I described how I can use and debug the Seeed Arch Mix Board. But so far I only had things running in RAM. Ultimately I want to use the QSPI FLASH memory on the device with my firmware and running code on it. This article shows how to get from RAM execution to SPI FLASH in-place execution (XiP).
The Seeed Studio ‘Arch Mix’ board is a small and versatile development board with an NXP i.MX RT1052 on it, and it costs only $29.90. So I was not able to resist and just have ordered one so I can explore it.
The ‘standard’ binary files for many tools are S19, binary or Intel Hex files. Especially for S19 and Intel Hex it can be useful to control the amount of data per line. By default, the GNU objcopy creates files with a line length of 44 characters:
default objcopy binary file line length
But it is possible to have Intel Hex files with an custom line length using the SRecord utility, and this is what this article is about.
A few days ago NXP has released a new version of their Eclipse IDE flagship: the MCUXpresso IDE v11.0.
NXP MCUXpresso IDE V11.0.0
The previous v10.3.1 was released back in Feb 2019, and the 11.0 now in June this year matches up with the Fall university semester. I appreciate that the releases are about every 6 months, so this gives me time to use it in my university lecture material and lab work. I had the weekend for trying it out, and I’m very pleased.
With the cost of an single pin, many ARM Cortex-M boards including the NXP i.MX RT1064 can produde SWO data: think about a pin able to stream data out of the chip in realtime. For example interrupt activity which otherwise might be hard to capture: