MSPLife 2 Posted August 3, 2016 Share Posted August 3, 2016 Dear everyone, I made six pcbs, but only one pcb well-worked. I just wrote a program to test LED and Interrupt on button with LPM4. My problem is only one pcb has about 3.3V on RST (well-worked), another about 2.2 to 2.8V. The lower one only work for a short time, after that it turns to be uncontrolled. I take off the chip and measure on RST pad (with out MCU), it is 3.3V. So I hope your help to figure out where is my failure, why the RST pin was dropped voltage, was my MCU broke? Please give me some suggestions to solve my problem. Thank in advance! Quote Link to post Share on other sites
cubeberg 540 Posted August 3, 2016 Share Posted August 3, 2016 Any chance it's a weak battery? A weak battery could measure 3.3v without load and lower with a load. Is the resistance from RST to VCC the same between the boards? Pictures might help as well. Quote Link to post Share on other sites
MSPLife 2 Posted August 3, 2016 Author Share Posted August 3, 2016 Thank for you reply. I use LM1117 3.3 to drop down 5V to 3.3V from USB port. I power directly from computer also try on a adapter. Quote Link to post Share on other sites
cubeberg 540 Posted August 3, 2016 Share Posted August 3, 2016 I'd suggest checking for shorts as well, in addition to checking RST resistance. It might also be good to check current across JP11 to see if it's higher on the non-functional board. Quote Link to post Share on other sites
Fmilburn 446 Posted August 3, 2016 Share Posted August 3, 2016 @@MSPLife You don't say whether or not you are using surface mount components. But if you are then carefully check adjacent pins on the microcontroller for solder bridging. Even if you cannot see them, check with a multimeter. cubeberg 1 Quote Link to post Share on other sites
MSPLife 2 Posted August 3, 2016 Author Share Posted August 3, 2016 Firstly, I can program the board, the code flashed without any error. I take off the chip and measure the pad, it is 3.3V, so the board can not be shorted. I face with 5 pcbs with same problem. I used a soldering iron without adjustable heat. I wonder if the MCU was dead by overhead, but after all, I can flash the firmware so .... Quote Link to post Share on other sites
MSPLife 2 Posted August 3, 2016 Author Share Posted August 3, 2016 @@MSPLife You don't say whether or not you are using surface mount components. But if you are then carefully check adjacent pins on the microcontroller for solder bridging. Even if you cannot see them, check with a multimeter. As my topic title, I used a TSSOP 20 pin Quote Link to post Share on other sites
cubeberg 540 Posted August 3, 2016 Share Posted August 3, 2016 @@MSPLife - check for shorts and resistance differences anyway - many of us have been bitten before by similar issues. I've had a few DOA ebay parts as well (odd resistor values, etc.). I once had a short on a TSSOP that ended up being beneath the pins in a way that couldn't be clearly seen. It could be a fried MCU as well (a short between VCC and GND resolved when I pulled the chip off) - although I've never seen one that programmed correctly afterwards. Quote Link to post Share on other sites
yyrkoon 250 Posted August 3, 2016 Share Posted August 3, 2016 SOrry I got that backwards, We tied reset to vcc through a 47k resistor. Anyway, the datasheet or users guide. I forget which. States that you can leave reset float if you're using the on die watchdog. We found this to be incorrect. As our software was using the watchdog, with reset floating. The msp430 would not work until I put a 47k ohm resistor between reset and vcc. I was getting odd, and inconsistent readings too. But it turns out the MSP430 was not even running, or if it were it was in a constant state of reset. I never checked with a scope. Quote Link to post Share on other sites
cubeberg 540 Posted August 3, 2016 Share Posted August 3, 2016 @@yyrkoon - based on the schematic - I see a 47k resistor on RST yyrkoon 1 Quote Link to post Share on other sites
spirilis 1,265 Posted August 3, 2016 Share Posted August 3, 2016 RESET trace isn't too close to a GND trace or plane right? PCB fab shops do have tolerances for the clearances between traces, drill sizes, etc... if you're violating those tolerances you may have unpredictable variation between boards. Just a thought Quote Link to post Share on other sites
yyrkoon 250 Posted August 3, 2016 Share Posted August 3, 2016 @@yyrkoon - based on the schematic - I see a 47k resistor on RST Ah right, ok i se it now thanks. Old eyes . . . that are used to reading someone elses schematics cubeberg and spirilis 2 Quote Link to post Share on other sites
MSPLife 2 Posted August 3, 2016 Author Share Posted August 3, 2016 My question is, if I can flashed firmware to MCU, was my MCU have ability to be dead? Quote Link to post Share on other sites
cubeberg 540 Posted August 3, 2016 Share Posted August 3, 2016 My question is, if I can flashed firmware to MCU, was my MCU have ability to be dead? I personally wouldn't rule it out. Best advice I can give - rule out potential problems, one by one and don't make any assumptions that parts are viable (could be a bad LDO, especially if you're not 100% sure it's genuine) Quote Link to post Share on other sites
yyrkoon 250 Posted August 3, 2016 Share Posted August 3, 2016 I wonder how much current you're sinking on P2.2 Quote Link to post Share on other sites
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.