SAMD21G burning

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello!
I create this discussion because I am so screwed with my Microcontroller SAMD21G. I designed a board to drive six motors and as soon as I supply the board, a shortcut is created inside the Microcontroller itself. I replaced it three times and it did this everytime. I checked the schematic and I realized that I plugged the VDDANA pin through a EMI filter and a capacitor of 0.1 uF. I checked inside the datasheet, there is nothing said on the fact that not to plug this VDDANA pin to VCC.
I don't understand why it keeps blowing up. My boss tells me that it should just not make the ADC/DAC work but no big deal because I don't use it. 

VDDANA also supplies the inside oscillator but I am using an external crystal.

I don't know what happens and I can't order a new board if I don't know what's wrong with this one...

 

Thanks for your advices

The french man Lambi 

Attachment(s): 

Last Edited: Fri. Jul 15, 2016 - 07:20 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

"Burning up" is usually a symptom of very serious errors. For example, reversed power and ground. Or, many IO pins that are set to be logic high connected to ground; note that IO pins are often current limited but if you have enough that are current limiting, then you will exceed the package dissipation limits and it can get quite hot. There are a few other possibilities, such as exceeding the current limits on many I/O ESD protection diodes but that is harder to do.

 

I would start by checking the polarity of the power AT the IC.

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Quote from the datasheet:

Quote:
The same voltage must be applied to both VDDIN, VDDIO and VDDANA. This common voltage is referred to as VDD in the datasheet.

 

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the advices guys. I checked again the polarity and it is good. 
I also checked for the footprint to make each pins corresponding to the good ones and it's good.

I just did not connect the unused pins on my board.

Except the VDDANA mistake, everything looks fine but I don't think this should be enough to blow up the Microcontroller...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Don't guess. Connect this pin to VCC and try again.

 

There are a lot of things powered though clamping diodes right now. Plus OSCULP32K is powered from this domain, and it is absolutely required for CPU to start running.

 

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'd like to, but I ran out of chips and I just want to figure out why they all blown up.  So I have to guess for this time

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Provided that everything else on the board is fine, my guess is that OSCULP32K is really required. It is used for clocking initial CPU boot process and your hardware is in some unknown sate.

NOTE: I no longer actively read this forum. Please ask your question on www.eevblog.com/forum if you want my answer.

Last Edited: Fri. Jul 15, 2016 - 09:42 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I had to look into a board failure a few months ago where the MCU (one of the ARM chips from ST) was reaching around 120°C, I traced it to a faulty smoothing cap on one of the VDD pins forming a short to ground. As well as drawing the current from the supply rail it was also flowing through the chip from the other VDD pins. So to back up replies above, any error with supply pins is a good place to start, it needs to be right for everything to work correctly anyway.