I am porting Arduino to STM32F746 board that is in progress yet.
I got a lot of help from this site and thanks for all contributors.
This board has arduino uno pins and extra connectors for motor, uart, can and gpios. it also has mpu9250 for IMU.
It is based on CubeMX project in this site and I am trying to support most of arduino functions.
GPIO/ADC/SPI/UART/PWM/USB Serial test are done so far and it use own bootloader I made through usb serial and using MAVlink protocol.
I also want to share everyhing about it.
Thanks again.
this korean cafe site(http://oroca.org/) about board and there are a lot of information about robots(sorry it’s wrriten in korean) :
http://cafe.naver.com/ArticleList.nhn?s … oardtype=L
source code
– main github : https://github.com/ROBOTIS-GIT/OpenCR
– bootloader : https://github.com/ROBOTIS-GIT/OpenCR/t … bootloader
– arduino core : https://github.com/ROBOTIS-GIT/OpenCR/t … cr_arduino
– download program : https://github.com/ROBOTIS-GIT/OpenCR/t … /opencr_ld
this is the board( it’ll be changed later)
– ADC Test
– MPU9250 test via SPI
– PWM test

Better even: getting a DIY F746 breakout board would be great as the discovery kit is a bit pricey
Is there a place to follow the development in English?
This board seems designed to be used with ROS 2.0 embedded. But I am surprised to not see any Ethernet port on the board. What communication interface will you use?
I have seem that a USB communication is planned according to Morgan Quigley’s talk at the RosCon 2015. So are you aware of any progress on it?
It’s good hear that others are working on the STM32F746.
It’s also good news that ST are releasing a 400MHz ‘746 in late October/November time.
Back in July 2015, I also was concerned about the cost of the F7 Discovery – so I made a simple breakout board that mimics the F4 Discovery board headers, making it a plug-in peripheral compatible but running at 2.5 x the throughput.
- BOB_lowres.jpg (51.39 KiB) Viewed 3483 times
That board looks really good! What package does it support?
I’ve soldered one to a generic adapter but it lacks the power and all other basic connections that are a bit of a pain to connect with wires…
There appeared to be little interest in this board last year – but perhaps with a few more firmware libraries available, plus more overall experience of the F4 and F7 range, that now is a ltime to release this board.
Yes, and now specially with arduino support the interest will become even greater.
If you’ll share it I’ll have some made and tested. I have some chips but on 144 LQFP I think.
The Mega approach is also a possibility, even with F4’s or F1’s it could be done but no one tried it (as far as I know)…
I like very much the standard arduino format with double headers like FRDM-K64F with MCU similar to 746: https://developer.mbed.org/platforms/FRDM-K64F

I have absolutely no idea what I would do with a chip that fast. But its still pretty

But 32k program limit? Bit small isnt it?
Is there a place to follow the development in English?
This board seems designed to be used with ROS 2.0 embedded. But I am surprised to not see any Ethernet port on the board. What communication interface will you use?
I have seem that a USB communication is planned according to Morgan Quigley’s talk at the RosCon 2015. So are you aware of any progress on it?
Thats a very competative price.
If someone can get the core to work on that board (I suspect the HAL MX core should work), it would possibly be the fastest ever Arduino board (well except if someone is running Arduino on a bare metal BeagleBone or RPI board)
Shipping fee to USA only 4.90$.
Shipping to me in Australia is free.
Registered Post Air Mail $5.50
That’s fair enough for registered air mail (I ALWAYS use the registered mail option because it’s always faster – up to 2 weeks and no parcel got lost in the past!).
I totally agree about registered post.
When I buy on Aliexpress, I now look for vendors who offer ePacket as a shipping option.
Some have this as their default, and some just charge a few dollars for this.
I find ePacket gets delivered in around 1 week and is worth it if you are spending $10 or more.
Just FYI…
Just FYI…
The next cheapest identical module is this one.
The next cheapest identical module is this one.
Just FYI…
I am very curious what goes into adding a new board to the “duino” IDE. I noticed how there are several toolchains you can install integrated in the IDE.
I am working congruently on several projects but if anyone is still interested in this I will build hardware and software support for it.
I am assuming I need to build a toolchain and cross compile some libraries. Nothing in life is ever so simple. But this is worth doing.
I would add “duino” style headers and possibly a sensor or connector. I am open to suggestions.
The second image is an example using BOB.zip however perhaps we would use I different MCU.
The first image is an NXP based board.
I do hardware & software for fun. But I am not as good at firmware.
I feel however the main hold up to the STM32DUINO is a lack of boards. And a lack of IDE support.
I have used arm before with Linux however I have never got this close to the hardware. However I have had exposure to AVR and assembly.
My main focus is native USB support and data through put. I will be designing this ultimately for science.
P.S. These boards are very preliminary.
I am very curious what goes into adding a new board to the “duino” IDE. I noticed how there are several toolchains you can install integrated in the IDE.
I am working congruently on several projects but if anyone is still interested in this I will build hardware and software support for it.
I am assuming I need to build a toolchain and cross compile some libraries. Nothing in life is ever so simple. But this is worth doing.
….
I looked into this a lot in 2013 and 2014 however was not excited about the issues with the maple libraries.
I still have not made my own foot prints or schematic (for this project not in general) I’m sort of shopping around for schematic ideas and practice routing boards. After I settle on a design ill produce component libraries, boards, bom all that jaz.
I am open to any and all suggestions about the best STM cores as well.
Stripped down olimex board. stmduino ish thing.
P.S. All FOSH.
This is the fifth practice board in 48 hours.
Once a MCU is selected it will only take a few days to get a full first revision any and all input or direction welcome.
From what I understand from @danieleff, its not actually that hard to port to a new core.
He ported the either the L4 or F1 code to make a F4 core for himself.
But hopefully STM will make their own F4 core soon (if not we may end up doing it for them ;-0)
http://www.st.com/content/ccc/resource/ … 224583.pdf
http://www.digikey.com/product-detail/e … ND/6137870
2M memory? 200+ MHz I need at least 24 MB per second so I need to look into the hardware a little anyway. But I assume this has a very fast SPI buss or two…. lots of timers and dma…
would look something like this
Would still be “arduino style” and would have a lot more IO pins. especially to connect a display…
The existing shields do require +5V compatibility, which is an extra challenge. There are already several boards for the Uno style market area.
For Arduino Mega style, we can leave the +5V compatibility out and have an additional improvement for distributed supply voltage and ground. Instead of just having the pins on left and right, add for every signal a pin for voltage and ground. For the two rows of bottom pins, add a separation between the rows for voltage and ground. Also add two more rows for voltage and ground. After that, the user can connect a 3 wire sensor, servo, or electronic speed controller in left, bottom, and right in a standard way.