STM32F427IIH6 Boot Problem

[forevermarvellk]

Posted on September 18, 2015 at 06:26

Hi, currently I am using STM32F427IIH6 for my application, which is a 176+25 UFBGA package chip with its minimum circuit. When I am trying to upload my bootloader to the chip, the debugger fails to recognize the device. The debugger I am using is Black Magic Probe following the steps from (http://smaccmpilot.org/hardware/blackmagic.html) and it keeps giving me the failure message ''Target Voltage 3.3V \\ SW-DP scan failed!'' when I run the line ''monitor swdp_scan'', which is to scan the device using serial wire debugger. However, the process can be done with other devices as STM32F407 and STM32F103. 

The default boot address is the user flash with my current circuit and I have also tried to boot from the system memory and the SRAM. All of the methods I tried but get the same reply of ''SW-DP scan failed''. I am not sure what may cause this problem, which may be the schematics error, soldering problem or any other reasons. I have checked the schematics for several times and no error detected yet. Now I am totally at a loss what to do with it. Please help to suggest me.Thank you very much!! #stm32

[Tesla DeLorean]

Posted on September 18, 2015 at 14:40

Take a look at what you're doing with the BYPASS_REG pin.

Check you have 1.25V on VCAP pins.Check the level of NRST

[forevermarvellk]

Posted on September 19, 2015 at 12:32

Hi, clive1,

I have checked the items you told me:1) I have done nothing with the BYPASS_REG pin. The pin is floating;2) I have checked the level of VCAP pins. Both of them do not have 1.25V, the level only reached 0.3V. I think the problems come from here.3) The NRST pin level is 3.26V. So what could be the probable reasons for wrong VCAP pin voltages? Maybe wrong capacitors value? Since my boards using very small package capacitors, so I cannot remove them and change to the correct ones. The only way is to send the boards back to the manufacturer. Thank you very much for helping me! You are really awesome.

[Tesla DeLorean]

Posted on September 19, 2015 at 16:04

Well it's indicative that the internal regulator isn't turned ON, the BYPASS_REG pin should be pulled down (examples I've seen 10K to ground).

With it OFF, you're going to have to provide 1.25V (1.2 - 1.3V for 168 MHz) from an external supply. Tack solder to the chip side of the VCAP capacitors.

[forevermarvellk]

Posted on September 20, 2015 at 10:25

Hi, clive1,

Thank you very much! I think that was exactly the problem. I also read through the datasheet and found that the internal 1.2V voltage regulator is not ON and the internal reset is not ON. The ''Regulator ON'' option is only activated when BYPASS_REG set to VSS and ''Internal reset ON'' option is activated when PDR_ON set to VDD. This will make change to my hardware PCB design. For implementation, do I need to attach a 10K pull up/down resistor to these 2 pins?Many thanks and this confuses me for several days. Seems I have to scrutinize the datasheet next time when designing the circuits.