Announcement

Collapse
No announcement yet.

XPS 13 7390 2-in-1. No Power On.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    XPS 13 7390 2-in-1. No Power On.

    I've got a XPS 13 7390 (2-in-1 variant) that I've used every day without issue for 2.5 to 3 years now. Specs and failure info are below. This model has an SSD that's soldered to the board with some very important information on it that I need back. I have a bit of experience in electronics repair from years ago (my dad was one of the first-gen computer repairmen, but most of my knowledge is from desktop motherboards with interchangeable components-- not netbooks with microsoldering. Pointers would be very much appreciated.

    It should go without saying that I did only cursory research when I got and mixed up the two versions of the XPS 13. I thought I had a separate SSD or I would've backed up my data when I was literally doing data backups of the family PCs about 2 days before this happened. But I didn't, painful lesson learned. What I've heard from Reddit and Dell forum posts is that, since the SSD is soldered to the motherboard, recovery is unlikely/impossible. If it were in warranty, Dell would replace the whole motherboard and all data would be lost for any error on an integrated component-- processor/graphics/hdd/memory/etc.

    I had initially wondered whether the drive could be disconnected from the board and re-soldered to a standard NVMe pinout, but given the mechanism of failure and symptoms, I suspect the error is in the board and it can be repaired/rigged to work for long enough to boot and copy the data off-- which would be easier.


    Model Info:

    Dell XPS 13 7390 (2-in-1).
    Processor: Core i7-1065G7
    RAM: 16GB 3733MHz LPDDR4x Memory (Onboard)
    HDD: 512GB PCIe NVMe x4 Solid State Drive (Onboard). Benchmarking pre-crash let me know this was the Kioxia BG4 series, I believe the KBG40ZPZ512G. https://business.kioxia.com/en-emea/...t-ssd/bg4.html
    GPU: Intel(R) Iris Plus Graphics

    Motherboard Info:
    Manufacturer: Compal
    Model: Centenario (DDP31)
    PCB: LA-G172P

    Failure Background:

    Last Sunday, the computer was working fine with no issues or indications of a problem. I put it into standby Sunday night (normally I shut it down or leave it on) and left it unplugged. Monday morning I go to turn it on and when I push the power button, I hear a fan spin up and the front indicator light lights up all amber but that's it. Releasing the button stops both. I think, OK, so it's dead. That sucks. And plug it in.

    When plugging into the left USB-C, nothing happens. Same behavior as unplugged. When I plug it into the right USB-C, a fan stays on continuously. The screen doesn't so much as flicker. The front light stays lit all amber for as long as I have the power button depressed but at no other time.

    I took off the bottom panel and totally removed the battery. Attached is is what I'm looking at.

    For reference, because the board is flipped and upside down here, when I plug into port 2, nothing happens. When I plug into port 1, Fan A spins at 100% until unplugged. Also, and interestingly, component X gets VERY hot.

    I initially tried letting it charge for a bit, but smelled one of those close-to-burning-electronics smells so I unplugged it. To test where it was coming from, I plugged in the power and left it on for around 30 seconds. Component X heated up to over 100 degrees F in less than 5 seconds and after 20 or so, it was at ~236 as measured with a laser thermometer. The same happens when the battery is connected and the power button is depressed. So clearly power is shunted through component X early in the process. The temperature also dropped off quickly when unplugged, so I assume that's where the fault lies as no designer would put a part that gets that hot outside of the air channels and leave it to passive ventilation. Above component X are a Micro-SD slot and the non-functional USB-C, which formerly had a hub with power delivery plugged into it. I suspect the hub somehow shorted out a control circuit and that is component X. Since USB-C serves as both the charging port and a data port on this PC, it has to be wired into the charging circuit so my suspicion is that the component that failed at X is now drawing all of the power and preventing boot.

    I've got a schematic request in so I can better start diagnosing what's on the PCB (specifically what is component X) and trace out power flow, but until I can move forward there any suggestions are appreciated.

    Thanks in advance.
    Attached Files
    Last edited by horsebatterystaple; 06-21-2022, 08:36 AM.

    #2
    Re: XPS 13 7390 2-in-1. No Power On.

    Originally posted by horsebatterystaple View Post
    I've got a schematic request in so I can better start diagnosing what's on the PCB (specifically what is component X) and trace out power flow
    Searched by PCB No and found the schematics.

    Still appreciate any suggestions as I move forward. Will update with what I find though.

    Comment


      #3
      Re: XPS 13 7390 2-in-1. No Power On.

      If the metal shield is not soldered, consider to lift it up to review what is underneath. The metal shield appears to be compression fit with the usual fingers. Remove all power and carefully lift up with a spudger or a small jewellers screwdrive but non-metallic is preferred.

      Comment


        #4
        Re: XPS 13 7390 2-in-1. No Power On.

        Originally posted by mon2 View Post
        If the metal shield is not soldered, consider to lift it up to review what is underneath. The metal shield appears to be compression fit with the usual fingers. Remove all power and carefully lift up with a spudger or a small jewellers screwdrive but non-metallic is preferred.
        Thanks very much much, I will do this next. I was comparing the info on the PCB (there's a D462 next to the shield) with the schematics and that component seems to be listed as:

        LMBR4010H3T5G DFN-1608

        I confess I'm a bit rusty on my electronic component diagrams, but the datasheet seems to suggest that's a Schottky Barrier Rectifier and the page it's on in the schematics is for P044 - TYPE-C Port R_PD.

        I suspect I'm onto something and that USB PD trashed it somehow. Will report back later today after I pull it off and see what I'm dealing with.

        Comment


          #5
          Re: XPS 13 7390 2-in-1. No Power On.

          Originally posted by mon2 View Post
          If the metal shield is not soldered, consider to lift it up to review what is underneath. The metal shield appears to be compression fit with the usual fingers. Remove all power and carefully lift up with a spudger or a small jewellers screwdrive but non-metallic is preferred.
          Upon further inspection, the shield around X is soldered, however the top portion of the shield has come loose-- right next to where it gets so hot. It may have gotten hot enough to melt solder when I plugged it in to charge last week.

          Comment


            #6
            Re: XPS 13 7390 2-in-1. No Power On.

            Is the part that gets hot UNDER this metal shield or nearby ?

            Can you take clear pix of this area and identify what is getting hot ?

            Would not be surprised if the USB Type C power delivery controller is getting baked. Use it to cook up some fajitas.

            Comment


              #7
              Re: XPS 13 7390 2-in-1. No Power On.

              Originally posted by mon2 View Post
              Is the part that gets hot UNDER this metal shield or nearby ?

              Can you take clear pix of this area and identify what is getting hot ?

              Would not be surprised if the USB Type C power delivery controller is getting baked. Use it to cook up some fajitas.
              The shield itself is what's getting hot so I believe the component is under the shield.

              Here are two closeup shots of the right side of the motherboard. The part outlined in red gets hottest, though mostly near bottom-center of the square I drew.

              I'm currently investigating the power distribution and power-on functions to try to identify what that might be before removing it. While I can't find mention of S4 as an independent component, it's listed as a a part of the power-on protocol several times.
              Attached Files

              Comment


                #8
                Re: XPS 13 7390 2-in-1. No Power On.

                Using Boardview I've managed to peek under the shield.

                As expected, that's where a lot of USB power components are. The shield is apparently to dissuade probing of the area.

                The big square in the upper-left is called a "BURNSIDE-BRIDGE_BGA_105P-T" and when I Google that I find it's one of two intel USB/Thunderbolt components.

                There are other diodes and chips and whatnot in the vicinity, but given the area that gets so hot, I'm going to operate on the assumption that this is the problematic part. It blows my mind that the retail on this part is under $3.00 and it can just crap the bed and knock out a $1,000 laptop. Then again, the diodes go for $.18 and they can do the same so. Wow.

                I'll be testing power at the 8 or so jump points tonight.
                Attached Files

                Comment


                  #9
                  Re: XPS 13 7390 2-in-1. No Power On.

                  the shield can be removed by removing screws holding the CPU heat sing from the bottom of the motherboard.. Post voltages on PR211 , PL101

                  Comment


                    #10
                    Originally posted by horsebatterystaple View Post
                    Re: XPS 13 7390 2-in-1. No Power On.

                    Using Boardview I've managed to peek under the shield.

                    As expected, that's where a lot of USB power components are. The shield is apparently to dissuade probing of the area.

                    The big square in the upper-left is called a "BURNSIDE-BRIDGE_BGA_105P-T" and when I Google that I find it's one of two intel USB/Thunderbolt components.

                    There are other diodes and chips and whatnot in the vicinity, but given the area that gets so hot, I'm going to operate on the assumption that this is the problematic part. It blows my mind that the retail on this part is under $3.00 and it can just crap the bed and knock out a $1,000 laptop. Then again, the diodes go for $.18 and they can do the same so. Wow.

                    I'll be testing power at the 8 or so jump points tonight.
                    Did you ever figure out the issue? I had identical symptoms, flashing the bios with a donor allowed the device to boot, but also seemed to break the Intel ME and both USB-C Ports. I am able to boot the system to Linux via the microSD card slot, where keyboard, mouse, SSD, WIFI & BT, and touchscreen all work. Both USB-C ports allow charging, but no video or data.

                    Comment

                    Working...
                    X