MSP430F5659: ESD causing micro to crash.

Jerry Bucci

Part Number: MSP430F5659

Hello.

I have a problem with ESD causing havoc to the operation of the micro. The design runs fine in every aspect, but occasionally when plugging in a USB cable to the USB connector it causes a transient that I can't squelch. I believe it to be a grounding noise problem. I cut all 3 out of the 4 traces coming from the USB connector to the board, only leaving the ground signal, and this still happens. Maybe, only 1 in 10 insertions of the USB, but we cannot go to production until this is fixed.

I tried adding more capacitance between the supply lines. The board has about 8 capacitors, mostly 0.1uF scattered around the board. I tried adding a ferrite bead between the USB ground to the board ground also.

It's a board that is about 5"x 5" two sided board. It's two sided for cost reasons. I do have ground planes top and bottom side, plus heavy grounds running throughout the board. The design is battery operated from a Li Ion 4.2V battery that is connected to the board through about 3" wires. The 3.0V regulator is a TPS73601 with a 47uF outout cap . There are two clocks, the XT1 at 32Khz and the XT2 at 16Mhz. The MCLK is set at 16Mhz.

I'm using the watchdog timer so when this occurs the program of course restarts.

Any ideas on what I can try, to correct this. I know there are IC's with a bunch of Zeners on all the USB lines to ground, but as I stated this happens with just the USB ground connected!

Thanks

Jerry

over 6 years ago

0 Leonardo Estevez over 6 years ago

TI__Genius 12615 points

Hi Jerry,

Can you check to see if you have a 33 ohm resistor to ground on the ground shield of the USB connector as shown in the MSP430FR5529 launchpad schematic?

BR,

Leo

0 Jerry Bucci over 6 years ago in reply to Leonardo Estevez

Expert 1550 points

Leo,

Thanks for the reply. I had the case floating. However, I have tried connecting it directly to ground and also through a 15K ohm resistor as I saw was done in another TI schematic. None of them seem to make a difference. Besides, most inexpensive Type C USB cables are not shielded, so the case of the connector goes nowhere.

Jerry

0 Jerry Bucci over 6 years ago in reply to Jerry Bucci

Expert 1550 points

Hello,

I guess i need to ask specific questions regarding this. I played with raising the VCore voltage, and changed the 0.47uF cap across it to different values. I know this question has been asked many times before, but I cant find a clear answer, other then 0.47uFd works best? What does that mean? What's wrong with a 10uFd, or a ..01uFd with lower ESR in parallel with 0.47uFd. If there is a large glitch on the ground from ESD, how would it most likely crash the CPU? The 16Mhz clock? I have plenty of capacitance on the 3V supply, spread around, so I would expect a noise pulse on ground would present itself as common mode to both the ground and 3V.

Thanks in advance for any ideas and help on this,

Jerry

0 Leonardo Estevez over 6 years ago in reply to Jerry Bucci

TI__Genius 12615 points

Hi Jerry,

I found another thread on this topic here:

https://e2e.ti.com/support/microcontrollers/msp430/f/166/t/46976?capacitor-choice-for-CVcore-is-the-recommended-470nF-value-critical-

Let me know if this helps.

BR,
Leo

0 Jerry Bucci over 6 years ago in reply to Leonardo Estevez

Expert 1550 points

Thank You.

If there is a large glitch on the ground from ESD, how would it most likely crash the CPU? The 16Mhz clock? I have plenty of capacitance on the 3V supply, spread around, so I would expect a noise pulse on ground would present itself as common mode to both the ground and 3V.

**Attention** This is a public forum

MSP low-power microcontrollers

MSP low-power microcontroller forum

MSP430F5659: ESD causing micro to crash.