Cannot Connect to STM32F405RGT6 on Breakout (ST-Link V2)

honboxuan · ‎2012-12-25

Posted on December 25, 2012 at 14:43

Hello all,

I created my own breakout board for the STM32F405RGT6, but I'm unable to connect to it via the ST Link V2. The error is ''No Target Connected''.

I've tested this with an external 8Mhz crystal (with load capacitors) and without. All other pins are simple broken out. I've successfully connected to the STM32F4 Discovery using the exact same method.

After some searching, I found that pulling the BOOT0 pin HIGH allows me to see if the device is working. Sending 0x7F at 9600bps 8E1 to USART3 returns inconsistent bytes (changes all the time). On all pins of USART1, there is no response. The same thing on the Discovery board returns the expected 0x79 (on the first try, something else on all others).

This is the first time I've developed a breakout for STM32 devices, and it is possible I've missed a connection required for it to work. So far, it is simply VDD, VSS, PH0, PH1, BOOT0 connected.

Any help will be appreciated! Thank you!

Hon

honboxuan · ‎2012-12-28

Posted on December 28, 2012 at 22:03

Thanks for replying! The STM32F405RGT6 I'm using is LQFP64, and does not have the VREF+ pin..

honboxuan · ‎2012-12-28

Posted on December 28, 2012 at 22:11

After putting the 10k pull-up on PB2, the chip responds with 0xFF 0xFF every time I send 0x7F.. Still don't know what this means though..

honboxuan · ‎2012-12-28

Posted on December 28, 2012 at 22:25

I went round and checked all the VDD and VSS pins (there are only a few), and they're all connected.

I tried switching BOOT0 out from HIGH to LOW, no difference..

I read the voltage at VCAPs again, and there's a change. VCAP1 is still at 0.2+V but VCAP2 has changed to over 1V, but fluctuates as I try and read it..

ken239955 · ‎2012-12-28

Posted on December 28, 2012 at 22:29

Normally NRST is pulled up to VDD via a resistor.

NRST is not connected on my board.

honboxuan · ‎2012-12-28

Posted on December 28, 2012 at 22:37

NRST is supposed to have a internal pull-up if I remember correctly, but disconnecting it doesn't change the response on USART3. I've just soldered a new chip to another board, and will be testing it now.

ken239955 · ‎2012-12-28

Posted on December 28, 2012 at 22:43

I know one of the hardest thing is to get a new chip and board up and running for the first time.

I'm not an expert here but just trying to help out.

I just finished building by first STM32F407VGT6 board and in the process of testing it.

My interface is to a STM32F4 Discovery board with only four connections, VDD, GND, JTCK and JTMS. Everything is working fine for me.

ken239955 · ‎2012-12-28

Posted on December 28, 2012 at 23:17

I am not sure what you mean by referring to USART3.

Ok, I just reread your previous post.

You said you are using Sparkfun 3.3V FTDI beakout using a UART or USART connection.

Sorry, I know nothing about this. Don't you need a bootstrap loader programmed into the MCU for this to work?

honboxuan · ‎2012-12-28

Posted on December 29, 2012 at 00:12

There's supposedly an embedded bootloader that allows this to work. Unfortunately I have no idea where to find any additional information about this. It'll be very helpful since it'll might become possible to figure out what the chip is responding with.

With the current setup, it responds consistently. But sending 0x7F as Clive suggests does not elicit the 0x79 that is expected (tested with the Discovery board too).

The second board which has been hastily assembled isn't working as expected either, but that's more likely due to the mess I've made of the soldering..

honboxuan · ‎2012-12-28

Posted on December 29, 2012 at 00:18

Also, the voltages on the VCAP pins have changed again. VCAP2 is now back to 0, and VCAP1 is 1+..

ken239955 · ‎2012-12-28

Posted on December 29, 2012 at 00:45

I think the voltages on VCAP_1 and VCAP_2 are crucial since this is the core's power supply and should read around 1.2V.

You have a working Discovery board. What are the voltages there?