STM32f105RCT6 - faulty MCUs?

[raf_gc]

Posted on June 07, 2010 at 13:31

STM32f105RCT6 - faulty MCUs?

#11011

[John F.]

Posted on May 17, 2011 at 13:53

Looking at the data sheet,

http://www.st.com/stonline/products/literature/ds/15274/stm32f105rc.pdf I noticed that although PA14 is 5V tolerant (PA14, I/O, FT), PA2 is not! The absolute maximum for PA2 is Vdd+0.3V : about 3.6V for a 3.3V Vdd. You are abusing the part - it wasn't faulty. John F.

[raf_gc]

Posted on May 17, 2011 at 13:53

I didn't said i used PA2 for 5V input, it was left floating. I just said i discovered short between Vdd3, PA2 and PB14 (sorry, i misspelled it with A) with mulitimeter. I bought 2 new MCUs, i checked them with multimeter after i opened the package before any soldering and powering, and those new ones have shorts between those pins i mentioned earlier too.

[John F.]

Posted on May 17, 2011 at 13:53

I'm not sure you can check with a multimeter as I'd expect intrinsic diodes to conduct (if the meter applies more than about 0.5V). If you have 3.3V Vdd in your design, then which pins have 5V on them and are they 5V tolerant ones?

[raf_gc]

Posted on May 17, 2011 at 13:53

I think PB0,1,2 was used for 5V altough they are not a 5V tolerant, but it was not my idea, i just received directions and noticed that later. It might explain voltage leaking, but those shorts happen even between PA2 and PB14, no matter if i use red lead of multimeter on PA2 or PB14. There is short between Vdd3 and Vdd2 altough thare is no connection to Vdd1 and Vdd4, there is no short between Vdd1 and Vdd4 nad there is no short between Vdd1 and any other pin except a 700 ohm resistance between Vdd1 and PA2, PB14 using red lead (+) on Vdd1, when using black lead (-) on Vdd1, ther is no connection between PA2 and PB14.

I want ot use USART1 for programming PA9 and PA10, is there neccesary, to convert 3.3V->5V for max232 input to accept 3.3V as high or max232 can handle it without problem?  According to stm32f105 manual, stm32f105 requires some lines to kept high, like PD6 wchich doesn't exist in lqfp64 package. Does it means, DFU programming is not possible via USB in this package?

[Tesla DeLorean]

Posted on May 17, 2011 at 13:53

I want to use USART1 for programming PA9 and PA10, is there necessary, to convert 3.3V->5V for max232 input to accept 3.3V as high or max232 can handle it without problem?

The spec for the TI MAX232 specifies VIH has a minimum of 2V, VOH from the STM32 should have a minimum between 2.4 - 2.9. It will work.

[power3347]

Posted on May 17, 2011 at 13:53

打酱油。。。看看能�?能�?�中文。。

[Andrew Neil]

Posted on May 17, 2011 at 13:53

'' is there neccesary, to convert 3.3V->5V for max232 input to accept 3.3V as high or max232 can handle it without problem?''

The MAX232 is specifically designed for 5V systems; don't mess about with it, just use an RS232 transceiver that is specifically designed to work in 3V systems! Simples!

[John F.]

Posted on May 17, 2011 at 13:53

PB0 and PB1 are not five volt tolerant (although PB2 is). If you put five volts on PB0 or PB1 I expect the +3.3V supply will be pulled up. Don't do that.

Clive1's remarks are technically correct regarding the TI MAX232 device and you are lucky that both PA9 and PA10 are five volt tolerant so the STM32 should accept the nominal five volt output from the TI MAX232 receiver. In general though, as neil.andrew says, why not use an RS232 driver designed for 3.3V if you can get one. I'd stop trying to see what power pin is connected to what. It's often the case that chip manufacturers bring out multiple power pins which are intended to be powered from the same supply but may not be internally connected - it just depends on the chip layout and pinout. John F.

[raptorhal2]

Posted on May 17, 2011 at 13:53

I have no voltage conversion issues using a MAX3232C that runs at 57,600 baud to a PC over a 50 ft. cable.

Cheers, Hal