STM32F105 no boot loader connect via USB

[peters]

Posted on June 28, 2010 at 12:13

STM32F105 no boot loader connect via USB

[peters]

Posted on May 17, 2011 at 13:56

Hi Chikos,

sorry for the delay. >If you don't have correct voltage on it you won't have any communication on bootloader. >Are you using a USB filtering/ESD circuit between the STM32 and USB plug ? >Verify its outputs in this case. There is always no voltage from the PC on the data lines. I changed the USB cable and used different ports on my PC, but there where no changes. >Do you measure 5V at the second pin of the 1K resistor (or when you remove the resitor) ? There are definatly 5V across the resistor, but the PA9 pin 'sucks' the current:  I measured 120mV on the pin, which results to 20 Ohms resistance. When I start the MCU not in boot mode, the pin shows a normal behaviour. This puts me into confusion..... Regards Moardy

[chikos332]

Posted on May 17, 2011 at 13:56

Hi,

PA9 is mandatory to have USB IP function correctly. If you don't have correct voltage on it you won't have any communication on bootloader. Are you using a USB filtering/ESD circuit between the STM32 and USB plug ? Verify its outputs in this case. Do you measure 5V at the second pin of the 1K resistor (or when you remove the resitor) ?

[peters]

Posted on May 17, 2011 at 13:56

Hi Chikos,

here comes an update: I tried the following: 1. First I put the USB cable into its plug on my protoboard with power off. 2. Then I switched on the power for the MCU. I could see a burst package on both data lines. I tried this several times, allways with the same result: bursts on both data lines, but only one time a got an error message from my PC:  'installation failure'. I fear, the problem is my PC, not the MCU. Regards Moardy

[ealbach]

Posted on May 17, 2011 at 13:56

Moardy,

We worked for a couple of weeks trying to get an STM32F105RCT6 (64 pin, date code 948) to connect to the PC with every possible combination of pull-ups etc.

According to the errata at that time it should have worked. The USART would connect but USB never would. We had the same high current problem on PA9 too.

ST told us many things to try but nothing worked. Finally they gave us a new 952 date code part and it worked immediately.

I would suggest that you get the newest part, based on our experience. We wasted a lot of time on a part that was never going to work. ST got the part from us so they could evaluate it. They should update the errata sheet to say that parts before 952 may not work with USB boot. Now we are still trying to get the USB library to work fully but it seems ST has not fixed the issues yet. At least we can connect to the PC via USB and program the flash now though.

Eric

[peters]

Posted on May 17, 2011 at 13:56

Eric,

thanks for this information. We also tried a new F105 with Datecode 023 and it works immediately, but only with a 25MHz xtal, not with a 8MHz one. I just tried it again with the older F105, Datecode 923. With the 25 MHz xtal, it worked immediately. BTW: be shure to use an USB2.0 certified cable. With USB1.1 cables, the connect is a case of luck. Regards Moardy

[peters]

Posted on May 17, 2011 at 13:56

[Fleck]

Posted on May 17, 2011 at 13:56

Moardy, Eric,

we see the same problem with high current into PA9 here, as well. Datecode is 011. I have previously posted about this [DEAD LINK /public/STe2ecommunities/mcu/Lists/ARM%20CortexM3%20STM32/Flat.aspx?RootFolder=/public/STe2ecommunities/mcu/Lists/ARM CortexM3 STM32/STM32F105 DFU - how much current does it draw on PA9 pin&FolderCTID=0x01200200770978C69A1141439FE559EB459D758000626BE2B829C32145B9EB5739142DC17E&TopicsView=https://my.st.com/public/STe2ecommunities/mcu/Lists/ARM%2520CortexM3%2520STM32/AllItems.aspx?Paged%3DTRUE%26p_DiscussionLastUpdated%3D2010]in this thread but got no reply. We are using an 8MHz crystal and hav enot found any solution yet. Glad to here others have the same problem, though. ST support kept insisting it would work if we kept the relevant other pins at a high or low level. We do this, even tried experimenting with all combinations pull-up/pull-down resistors, but still no go. I assume the reason it draws so much current is that PA9 is configured as UART TX pin by the bootloader. Can Errata 2.5.6 be used to explain this problem? Best Regards, Jens

[antonveretenenko]

Sorry for digging this old post, but I failed to run usb dfu bootloader on stm32f105 datecode 721 and were searching for a solution for several days already.

So I was using 8 MHz crystal and were getting windows error of device detecting, reading descriptor etc.

A working solution is to use 25 MHz crystal, it then just works.

Also in AN3156 there is a note:

"For Connectivity line USB DFU bootloader, the device first tries the 25 MHz configuration,

then, if it fails, the 14.7456 MHz configuration, and finally, if it fails, the 8 MHz configuration.

In case of fail, this operation is repeated with a large timeout value (the three configurations

are tested again). If the second trial fails, a system reset is generated."

So I guess there is some timeout issue with detecting device while bootloader tries 25 MHz and 14.7456 MHz settings first.

Related to stm32f105, stm32f107 connectivity line mcus.

#stm32f105rb