SirMixALot DYI F7 Fully Loaded Flight Controller

User avatar
TonyFPV
Elite Mods
Posts: 22
Joined: Mon Apr 15, 2019 9:03 am
Location: Perth
Contact:

Thu Aug 15, 2019 9:31 am

If you use a programed chip or not it does not matter. If you put the stm32 into DFU it will always show up. If it does not for whatever reason, you can flash it with an ST-Link.
jrowe88
Posts: 16
Joined: Wed Jun 26, 2019 3:30 am

Mon Sep 02, 2019 4:31 am

Update...slow going here. Got new boards with the correct footprint. Success in that my Bat->5V regulator and 5->3V are all working well! Nice clean power. Ran into a little trouble with the USB though. USB-C to USB-C requires the CC1 and CC2 pins to determine what is supported and apparently I didn't realize that when I designed the board. I was able to use a USB-A to USB-C to connect successfully to the computer. Second face-palm moment was when I realized that I didn't look at the button specs correctly and I had a permanent short on the Boot pin. I removed the button and was able toa actually flash the chip!! Yee ha.

Well, kind of....after flashing the chip, I rebooted and nothing happened. No lights, no nothing. A little help from the Slack forum suggested that it was hanging in initialization for some reason. They suggested I use a debugger, so I ordered an STLINK-V3SET https://www.st.com/en/development-tools ... v3set.html. I tried flashing the 4.1 dev build of the STM32Fxx firmware to no avail either. This apparently is pretty benign and should run on about any board, including the ST demo "Nucleo" board. Well, any board except mine!

This opened a whole can of crazy debugging, including downloading a bunch of software and some insanely fine soldering to attach to the debug SWCLCK and SWDIO pins. (See the pics below)

Miraculously I was able to set up the debugger and actually step through code (thanks https://blck.mn/ for the pointers -- OCD and GDB ended up working, though I never got debugging directly connected to VStudio). Apparently there is a whole bunch of initialization before the code ever gets to "main()" and it is failing somewhere in there. After spending the better part of the day stepping through esoteric code, i decided I'd try another board and go with a minimal setup with just the power and MCU and try again. Well, after screwing up that board somehow, I am back to the drawing board.

I laid out a new board with pads to make it easy to debug directly. I'm also going to pour over some schematics to see if I did something obviously dumb. Stay tuned. I posted my latest board layout with some minor tweaks and the debug test/solder points. Red circles are new/changed stuff (test points, CC resistor, Diode on 5V/VBat selector)

I've posted all my photos and new diagrams here: https://photos.app.goo.gl/ceajJaWq5XVEXWPk8
User avatar
TonyFPV
Elite Mods
Posts: 22
Joined: Mon Apr 15, 2019 9:03 am
Location: Perth
Contact:

Mon Sep 02, 2019 7:10 am

You might have messed up the USB_CONNECT wiring to the f405.
User avatar
TonyFPV
Elite Mods
Posts: 22
Joined: Mon Apr 15, 2019 9:03 am
Location: Perth
Contact:

Mon Sep 02, 2019 7:11 am

opps, sorry i mean the f722
User avatar
TonyFPV
Elite Mods
Posts: 22
Joined: Mon Apr 15, 2019 9:03 am
Location: Perth
Contact:

Mon Sep 02, 2019 7:14 am

This is how I have mine set up. Might be different as this is on an f405 chip.
jrowe88
Posts: 16
Joined: Wed Jun 26, 2019 3:30 am

Mon Sep 02, 2019 7:31 pm

I think I have the same thing... 10K voltage divider. I'm just using a resistor network instead of individual resistors.
usbconnectPNG.PNG
usbconnectPNG.PNG (18.85 KiB) Viewed 963 times
I did end up finding a couple of things that could be causing the issue. I had all (baro, acc, osd, memory) of the SDI and SDOs connected incorrectly. Instead of MOSI to SDI, I had it connected to SDO, and MISO to SDI. I also discovered that for SPI3 I was mixing pin groups -- I'm not sure if that is OK or not, but maybe it wasn't finding the AF's correctly.
jrowe88
Posts: 16
Joined: Wed Jun 26, 2019 3:30 am

Fri Oct 25, 2019 9:05 pm

Hello Everyone, quick update with more to follow. I have successfully completed the board and it fully works -- USB-C, OSD, Accelerometer, Baro, LEDs, etc. Had to figure out a few final bugs due to my misinterpretation of the STM32F722 documents, but we are good to go! I'll post more later, but if you are interested in trying it, I'll share the Oshpark files, BOM, and some tips.
User avatar
Raj??ishKumar
Posts: 1
Joined: Fri Aug 30, 2019 6:04 pm

Sat Nov 02, 2019 5:58 am

Can you share the files and bom i want to try this board and want to make one of my own too
morbius
Posts: 1
Joined: Sat Nov 30, 2019 4:14 am

Sat Nov 30, 2019 4:21 am

Has your testing included a test flight?
Post Reply