STM32L152V8T6 behaving weird

[dhiraj]

Posted on January 25, 2017 at 06:04

Hello everyone,

I am using stm32L152V8T6 in my project with following peripherals enabled:

1. 36x4 segment LCD with external VLCD supply

2. 12 GPIO's configured as follows:

   a.6 pins are configured as input pull up

   b.4 pins are configured as output open drain

   c.1 pin is configured as input pullup with external interrupt

   d.1 pin is confugured as output pushpull

3.PVD interrupt enabled for low voltage indication

4.uart2 in asynchronuous

5.TIM2 enabled

I have used CUBEMX code generator  and openstm32 ide for development, the problem I am facing is the device works fine for few days and then some of the pins get permenantly pulled down/up(specially the pin configured as external interrupt gets pulldown).Is this a the hardware issue? Because the code runs fine for few days.I am keeping the device always on in stop mode and wake it up by external interrupt. I have replaced 3 MCU's from the pcb to confirm this issue and its a tough job to solder them again and again :\ .

device current consumption:

356uA in run mode

21uA  in stop mode

Waiting for the response,

Regards,

dhiraj.

[Imen.D]

Posted on January 25, 2017 at 16:20

Hello dhirajp15,

I suggest you to start with an example within your firmware package (STM32CubeL1) relevant to the device that you are using to confirm if the problem here seems related to the hardware environment.

Also, have a look to this 

http://www.st.com/content/ccc/resource/technical/document/errata_sheet/88/d3/7b/04/85/0f/46/05/CD00278726.pdf/files/CD00278726.pdf/jcr:content/translations/en.CD00278726.pdf

 to confirm if you apply the same conditions as described and the 

limitations have an impact you.

Regards

Imen

[dhiraj]

Posted on January 27, 2017 at 08:15

Hello Ma'am,

 I went through the errata sheet ,but could'nt find anything that affects the working of gpio pins. This is a very critical issue I am facing ,the next IC i soldered and it stopped working after 12 days. The problem I found is that the internal pullup for one of the pin was showing 0.42V instead of 3.1V(VCC) due to which the pin was read as low and was getting triggered continuously.

Waiting for the response,

Regards,

dhiraj

[dhiraj]

Posted on January 27, 2017 at 08:46

Also I want to add few more observations:

If I try to give external pullup to the pin ,it consumes 1.12 mA. I am now sure that this is a hardware issue , please guide me if there is any manufacturing defect in this chip ?

[dhiraj]

Posted on February 12, 2017 at 17:52

I have I quick product release and this isse  is causing me to wait , can someone help me out?

[Tesla DeLorean]

Posted on February 12, 2017 at 19:17

Maybe you should start by assuming the issue isn't with the STM32, and more likely your design. If pins are getting physically damaged you need to look if you have adequate protection on the pins or you are driving signals into outputs.

[dhiraj]

Posted on February 13, 2017 at 14:30

If this is the case , can you suggest the protection circuit required? I have 6 gpios configured as input pullup which gets connected to 4 gpios initialized as output open drain through switches forming a keypad design.

[Tesla DeLorean]

Posted on February 13, 2017 at 15:30

Are specific pins failing? Could they be subject to ESD or otherwise exceed the Absolute Maximums defined in the Data Sheet?

Clamping diodes, or other ESD protection should be considered. If the keypad assembly has a metal frame this should be grounded so when touched the ESD will go there.

Have the design reviewed with an ST FAE associated with your account.

[dhiraj]

Posted on February 13, 2017 at 15:46

The GPIO input pins are failing, not all together but one or two at a time ,If I try to give external pull up to same pin ,I observer a 1.1mA currrent consumed by failed pin meaning it gets pulled to ground internally .The keypad has carbon contacts along with plastic case and not metal switches. The STM FAE person asked me to add 330E resistance in series to the pin for protection , also a capacitor in parallel with the switch.