6 Replies Latest reply on Apr 15, 2016 11:26 AM by PabloM_Intel

    Trouble shooting Edison power-up on custom expansion board

    drschuet

      Hello,

       

      I'm trying my hand at a first Edison expansion board and failing, however, to get the Edison to power-up. The design is based off of the Sparkfun Edison Base Block (https://cdn.sparkfun.com/datasheets/Dev/Edison/Base_Block.pdf) with the USB interface removed.

       

      Presently  when the custom board powers up I see the 3.3V supply from the Edison come up between 150 to 200 ms after VSYS rises to it's supply level of 3.7V. The 3.3V is on and clean for ~150ms before dropping to 0V. There is no activity on the 1.8V supply. Any insight into what sort of fault this indicates would be greatly appreciated.

       

      A few notes...

        - PWRBTN# is connected to VSYS using the same resistive divider / reset button network that the Sparkfun Base Block uses.

        - DCIN is wired directly to VSYS.

        - FAULT is wired directly to GND.

       

      Any suggestions would be appreciated.

       

      Thanks!

        • 1. Re: Trouble shooting Edison power-up on custom expansion board
          Intel Corporation
          This message was posted on behalf of Intel Corporation

          Hi drschuet,

          Could you please tell us how you’re powering the custom board? Are you using a battery or an external power supply? Sometimes the issue could be related to the power source not being enough for the board. Sometimes, when a feature (e.g: WiFi) starts the boot process just stops or hangs, so you could consider that option and try with a different supply.

          Regards,
          Pablo

          • 2. Re: Trouble shooting Edison power-up on custom expansion board
            drschuet

            Hello,

             

            Naively I'd think there is enough power there. If I watch the VSYS trace along with the 3.3V on the scope I see no ripple or drop-out in either.

             

            VSYS is being supplied by a MIC39102YM low drop-out adjustable regulator  (http://www.micrel.com/_PDF/mic39100.pdf) with a 20kOhm (R1), 10kOhm (R2) restive divider network (fig. 2 of data sheet) with a 10 uF by-pass cap on the output. This network gives an VSYS of 3.7 V. In testing I am using an Agilent programmable power supply to source the input voltage for the regulator (current limit set to 1A). In operation it will be a 3A power brick.

             

            Regards,

            - Dan

            • 3. Re: Trouble shooting Edison power-up on custom expansion board
              Intel Corporation
              This message was posted on behalf of Intel Corporation

              Hi drschuet,

              Thank you for the explanation. If that’s the case, I would suggest you to check the DCIN and VSYS connection once more to see if it’s correct, as you can read in the Edison Module Hardware Guide, DCIN indicates whether the Edison is being powered from a battery or from an external supply. It also sets the voltage level required on VSYS in order to boot. According to your description DCIN is connected to VSYS, but the current behavior that you’re getting sound very similar to this part of the text: “VSYS must rise from 2.5 to 3.5V in 10ms; otherwise the boot is aborted”.

              Regards,
              Pablo

              • 4. Re: Trouble shooting Edison power-up on custom expansion board
                Intel Corporation
                This message was posted on behalf of Intel Corporation

                I would also suggest you to check this thread, you might find some new information, https://communities.intel.com/thread/59916.

                Pablo

                1 of 1 people found this helpful
                • 5. Re: Trouble shooting Edison power-up on custom expansion board
                  drschuet

                  Hi Pablo,

                   

                  Thank you for the help. After checking all the above I discovered the problem. My custom expansion board is a two layer board with power and ground fill instead of ground planes. Solder from one of the SMD connections was bridging the gap between a pad and a piece of power fill that was not fully covered at the edge by the PCB silkscreen. This was causing a short between 1.8V and GND resulting in the Edison's boot failure. Things appear to be working now.

                   

                  One small suggestion / request. I have found very little information on the PMIC used in the Edison, I understand that it is a custom part. A diagram of the power-up sequencing could be useful in debugging problems like what I ran into.

                   

                  In any case, thank you for the help!

                   

                  - Dan 

                  • 6. Re: Trouble shooting Edison power-up on custom expansion board
                    PabloM_Intel

                    Hi drschuet,

                     

                    I’m really glad to know that you were able to solve the issue with our suggestions.

                    Regarding the PMIC documentation, this is a TI component (more specifically SNB9024) and given that this is a non-distributor item there is no documentation available for it. This was their reply to another user some time ago:

                     

                    Thank you for contacting TI Semiconductor Support.

                    Unfortunately, since this is a non-distributor item, there is no documentation available for the SNB9024 device (datasheets, user guides, etc.).

                     

                    Regards,

                    Pablo