STM32F4Discovery - external VDD supply

[prabhu]

Posted on June 04, 2013 at 13:52

Hi All,

Is it possible to use STM32f4Discovery without powering it’s ST Link section and associated LEDs.  External 5V (at P2 pin 3,4)  will power up ST Link section and associated LEDs.  Is it safe to remove JP1 and apply external 3.3V at VDD pin?  In this case some on board devices(IC U1, U2, U3, U5, U6, U7, U8, U9) using 3V supply will not receive power. Will this have any adverse effect on STM32F407 operation (latch up, ADC performance etc)?

[Tesla DeLorean]

Posted on June 04, 2013 at 15:13

I certainly wouldn't want a bunch of unpowered logic backfeeding current.

Wouldn't it be easier to just get a breakout board that doesn't have all the chips/functions you don't want?

http://www.ebay.com/itm/Core407V-STM32F407VET6-STM32-ARM-Cortex-M4-Development-Board-Full-I-O-Expander-/261045971937

[prabhu]

Posted on June 04, 2013 at 16:28

Thanks Cilve!

I was also thinking of Core 407V and other possibilities, but having started with F4Discovery wanted to explore if it could be used for low power battery operated application. In standby mode VDD current is <500 uA, whereas current drawn by discovery board varied from 35 to 45mA. I believe most of it is from ST link section and wondered if there is any way to reduce this overall current drain in standby mode.

[Tesla DeLorean]

Posted on June 04, 2013 at 17:25

The supply looks to be reasonably partitioned, but you'd want to keep the 3V domain powered, lest the pull-ups on PB6/9 cause a current flow into the 3V domain via U7, and to other chips. There may be other paths, I just scanned the schematic. You could check for a voltage on the 3V domain with your currently described connectivity.

I would probably just start by breaking JP1 and placing the current meter across it instead. Then also look at removing R26, and examining the VBAT current. In all practicality you'll also want to add the 32.768 KHz crystal and caps so the RTC will run in STANDBY.