Announcement

Collapse
No announcement yet.

MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

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

    MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

    This was my friend's motherboard. He bought a 2nd hand computer which worked for a couple of days after he got it, then died. I troubleshooted it down to the motherboard and he bought a new board, which fixed the problem.

    Now I decided to try fixing his old one for fun.

    Problem: No POST, No beeps, power button won't shut board down after power up, have to power PSU off at mains.


    After having problems with my own machine locking up and failing to POST, which I eventually traced back to leaky power\reset buttons on my case, I started by desoldering the onboard reset button and checking the RESET line (FP_RST#) which turns out to be high (3.3v). So that at least seems OK to some extent.

    I noticed that FP_RST# also feeds the "SB700 & RS780 POWER GOOD CIRCUIT" (Page 27)


    In this circuit, I measured Q38 to have 5v on its gate, and the output of Q38, the SYS_PWRGD signal is LOW.

    This seems to be because Q37 has no base drive, and there's no voltage either side of R209. NB_VCC1P1 rail seems to be down.

    I notice that NB_VCC1P1 measures 19.6 Ohm to GND. Is this a short, or normal? The schematic (Page 28) suggests that this rail supplies 10.2 Amps.
    Attached Files
    Last edited by Agent24; 12-09-2014, 06:35 PM.
    "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
    -David VanHorn

    #2
    Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

    More information:

    U23 is getting 12.3v to Vcc Pin 5
    U23 is getting 1_2VREF voltage - 1.214 to VRef Pin 7
    U23 Feedback Pin 6 shows nothing

    Q42 is getting 5.08v to its drain
    "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
    -David VanHorn

    Comment


      #3
      Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

      Anyone know about these or have any ideas? Is that 19.6 Ohms on the NB_VCC1P1 rail normal or too low?

      Here's the datasheet of the buck converter and MOSFETs

      Edit: Pin 4 of the uP6103 (Lower FET gate drive) is at 12.24v and the Upper gate drive Pin 2 is 0v

      Datasheet of the uP6103 suggests that if the output was shorted, I should see some small voltage spikes on the output as it tries to turn on and shuts down over and over. I am getting nothing at all on the output. The Upper Gate drive Pin 2 never seems to even attempt to turn Q42 on at any point.

      Since all the inputs seem OK, it looks like the uP6103 maybe bad?
      Attached Files
      Last edited by Agent24; 01-20-2015, 05:16 PM.
      "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
      -David VanHorn

      Comment


        #4
        Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

        This is weird. I started going through all the rails and references voltages to see if there are any other problems.

        I found 5V_DIMM, VCC3_SB, +1.2VSB are all OK (no real surprise) but when I got to U16 which generates some reference voltages, things go funny...

        U16, apparently a UP6261BM8 is getting 5v OK, and the 1.8v Ref output is fine, but the 1.25v and 1.2v outputs are dead, either in standby or with the board started. I notice the 1.2v Ref feeds Pin 7 of U23 and this is now gone also (it was 1.2v when I first started testing, now it's got nothing)

        I wonder if U16 is the problem and it is in the process of slowly dying? I can't find a damn datasheet for it though...

        EDIT: U16 also appears to have an enable signal (EN_HT), Pin 8. This is only 0.05v - perhaps this is why nothing is happening... Interestingly this is derived from the VRM_GD signal, which comes from the ISL6323CRZ VRM controller (Pin 37, VDDPWRGD). Hmm...
        Last edited by Agent24; 01-20-2015, 08:12 PM.
        "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
        -David VanHorn

        Comment


          #5
          Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

          I have now discovered that the ISL6323CRZ will output VDDPWRGD to the VRef IC U16, If the CPU is unplugged, and the 1.25v and 1.2v Refs then come up fine.

          With the CPU unplugged, VCCP is 0.952v. With the CPU plugged in, it's 0v. The CPU is NOT shorted and works fine in another motheboard.

          Interestingly, U23 and the NB_VCC1P1 don't work whether the VRef IC U16 is enabled or not.


          So, I don't know. U23 is either being disabled somehow because the VRM controller doesn't like it when a CPU is plugged in, OR U23 circuit is bad and disabling the CPU VRM somehow?

          Anyone can shed some light here? This is my first time trying to troubleshoot a motherboard at this level.
          "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
          -David VanHorn

          Comment


            #6
            Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

            Very interesting.
            My pc
            CPU : AMD PHENOM II x4 @ 3.5Ghz
            MB : ASUS M4A89TD PRO USB3
            RAM : Kingston ValueRAM 16gb DDR3
            PSU : Cooler Master 850W Silent Pro
            GPU : ATI Radeon HD 6850

            Comment


              #7
              Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

              You know, I have a similar problem with a s775 Shuttle motherboard - when the CPU is inserted, CPU VRM outputs 0V, and without CPU, I think I get normal voltage. Haven't figured it out, and did nowhere as much circuit tracing and troubleshooting as you did.

              Anyways, let's keep it with your thread.
              Originally posted by Agent24 View Post
              Anyone know about these or have any ideas? Is that 19.6 Ohms on the NB_VCC1P1 rail normal or too low?
              I would lean towards normal. Chipsets, CPUs, and even RAM can have low resistance readings like that on their Vcc rails.

              I'll try to have a more detailed look at the schematics this weekend.

              I wonder if this could all be due bad BGA in the CPU socket. Given that both AMD and Intel have had a retarded heatsink retention mechanism since their s939 and s775 sockets, I wouldn't rule it out. Does the board look bent or curved under or near the socket area?

              The other thing you can check is for bad pins in the CPU socket. But to do that, you would have to remove the CPU socket top part. It's not very hard to do IMO, but often times I do end up breaking the plastic on the side. It doesn't affect the function of the socket afterwards, though.

              Originally posted by Agent24 View Post
              So, I don't know. U23 is either being disabled somehow because the VRM controller doesn't like it when a CPU is plugged in, OR U23 circuit is bad and disabling the CPU VRM somehow?
              Well, I think the question is more about why is the CPU VRM output getting shut down with a CPU? See what the CPU VRM controller has as input pins, and then try to determine which input pin is causing it to become disabled. The only snag you might run into is if the VRM controller has a input (receive) or bi-directionnal serial communication with either the CPU or the NB. If that's the case... well, I don't know what I would do at that point.
              Last edited by momaka; 01-21-2015, 01:19 AM.

              Comment


                #8
                Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                Thanks for the input. I saw you posted somewhere about NB etc having low resistance but I wasn't sure what a 'normal' resistance might be...

                The rear of the board has a pretty sturdy-looking backing plate. The board doesn't seem to be bent much - and I have other boards which are obviously bent more and still work, though they may not be BGA - I can't remember. But I guess bad BGA balls are always a possibility...

                The thing was shipped assembled to my friend, so who knows if it was knocked around. Then again, maybe the reason it got sold in the first place was that it had issues. I do know it worked for my friend for a few days then quit altogether and hasn't worked since.

                I'll have a look inside the socket if I can figure it out.

                Good point about the VRM controller, I'll have to read up on the datasheet for that and try to figure out what should be going on. Attached that here too.
                Attached Files
                "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
                -David VanHorn

                Comment


                  #9
                  Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                  Well, I had a look in the CPU socket but all looks OK. Removing the top part was fairly easy. I just slid a thin scalpel blade down inside the latch grooves and pushed out slightly. Managed to get the top off without breaking anything. All the contacts look fine.


                  But, I think I found out why the VRM controller doesn't output VDDPWRGD with the CPU plugged in. And I don't think it's because it can't run, but because it doesn't get its enable signal.

                  The enable signal for the VRM controller is derived from VCORE_EN# which is inverted by Q1 to become PWM_EN and which is fed to Pin 24 of the ISL6323CRZ.


                  That VCORE_EN# signal itself comes from the output of 4 transistors Q6, Q7, Q8 and Q9, and Diode D2 (See page 27) which make up some kind of custom discrete-component-based logic gate and will only output a low on the VCORE_EN# line when VCC5_SB, VCC_DDR, VDDA_25 rails are OK, AND, it gets the PWR_OK signal from the PSU - and that circuit has everything OK except for the VCC_DDR rail.


                  What I found is that with the CPU plugged in, the VCC_DDR rail (which should be 1.8v) is down at 0.38v so VCORE_EN# never goes low and the VRM controller does not start. Why on earth VCC_DDR does not come up with the CPU attached I don't know. That rail is controlled by U7 which drives Q24 and Q25, all of which work fine when the CPU is NOT plugged in.


                  U7 has its output controlled by its VRef pin, Pin 7 - which is brought high by the 1_8VREF rail from U16 to turn it on. This voltage input is in turn controlled by Q15, Q13, and Q12 in another discrete-transistor-based logic gate, but U7 always gets that 1_8VREF voltage just fine, with the CPU attached or not. It also gets Vcc, and, its MOSFETs get 5VDIMM just fine too, and there is no other control or enable signal. It just simply doesn't output the correct 1.8v when the CPU is in the socket...


                  And none of this explains why U23 doesn't produce the NB_VCC1P1 rail - either with OR without the CPU.
                  "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
                  -David VanHorn

                  Comment


                    #10
                    Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                    The story continues...

                    Since there was nothing else I could see possibly stopping U23 from running except itself, I decided to replace it and see what happens.

                    I was considering buying some uP6103 ICs on Aliexpress but have never bought there before so am a little wary of fakes and a probably nonexistent refund procedure... Anyway, I recently looked on eBay again, and saw there were some for sale now. For about $8 NZD I bought 10 of them from http://myworld.ebay.com/icmarket2009 as they have good feedback and I have bought from them before.

                    I replaced U23, and got the NB_VCC1P1 rail back! - but only when the CPU was not plugged in, since NB_VCC1P1 won't come up unless the VCC_DDR rail is up, and that doesn't work with the CPU installed.

                    VCC_DDR is created with another uP6103, U7. Since it's the same chip, one already seemed to have died, I had 9 more... I replaced U7 too..... and now the board POSTs!


                    I guess U7 and U23 both failed around the same time - how exactly, or why, I don't know. I wonder if they were knockoff or cheaper clone parts - the replacement uP6103s I received from eBay have larger printing on them. Or maybe they're just cheap ICs from a cheap manufacturer nobody's heard of, and they don't last.

                    I hope anyone following this thread learned some things about motherboard troubleshooting, I know I did
                    Attached Files
                    "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
                    -David VanHorn

                    Comment


                      #11
                      Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                      Cool! That would be a nice board now that you got it going.
                      I love putting bad caps and flat batteries in fire and watching them explode!!

                      No wonder it doesn't work! You installed the jumper wires backwards

                      Main PC: Core i7 3770K 3.5GHz, Gigabyte GA-Z77M-D3H-MVP, 8GB Kingston HyperX DDR3 1600, 240GB Intel 335 Series SSD, 750GB WD HDD, Sony Optiarc DVD RW, Palit nVidia GTX660 Ti, CoolerMaster N200 Case, Delta DPS-600MB 600W PSU, Hauppauge TV Tuner, Windows 7 Home Premium

                      Office PC: HP ProLiant ML150 G3, 2x Xeon E5335 2GHz, 4GB DDR2 RAM, 120GB Intel 530 SSD, 2x 250GB HDD, 2x 450GB 15K SAS HDD in RAID 1, 1x 2TB HDD, nVidia 8400GS, Delta DPS-650BB 650W PSU, Windows 7 Pro

                      Comment


                        #12
                        Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                        Very nice
                        My pc
                        CPU : AMD PHENOM II x4 @ 3.5Ghz
                        MB : ASUS M4A89TD PRO USB3
                        RAM : Kingston ValueRAM 16gb DDR3
                        PSU : Cooler Master 850W Silent Pro
                        GPU : ATI Radeon HD 6850

                        Comment


                          #13
                          Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                          Originally posted by Agent24 View Post
                          This was my friend's motherboard. He bought a 2nd hand computer which worked for a couple of days after he got it, then died. I troubleshooted it down to the motherboard and he bought a new board, which fixed the problem.

                          Now I decided to try fixing his old one for fun.

                          Problem: No POST, No beeps, power button won't shut board down after power up, have to power PSU off at mains.


                          After having problems with my own machine locking up and failing to POST, which I eventually traced back to leaky power\reset buttons on my case, I started by desoldering the onboard reset button and checking the RESET line (FP_RST#) which turns out to be high (3.3v). So that at least seems OK to some extent.

                          I noticed that FP_RST# also feeds the "SB700 & RS780 POWER GOOD CIRCUIT" (Page 27)


                          In this circuit, I measured Q38 to have 5v on its gate, and the output of Q38, the SYS_PWRGD signal is LOW.

                          This seems to be because Q37 has no base drive, and there's no voltage either side of R209. NB_VCC1P1 rail seems to be down.

                          I notice that NB_VCC1P1 measures 19.6 Ohm to GND. Is this a short, or normal? The schematic (Page 28) suggests that this rail supplies 10.2 Amps.
                          ok so did you get This Motherboard Up and going because ive got the same mobo. the cpu works in another mobo but nothing shows on screen.

                          dunno if Capacitors or something else is shorted on board.

                          i live in south africa western cape lakeside

                          Comment


                            #14
                            Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                            Yes I got it running. It's been working without issue for 3 weeks now.
                            Can't say the same for the Windows 10 preview I tried out on it though!

                            Start by checking all the voltage rails. Maybe start with NB_VCC1P1 and VCC_DDR since those two were at fault on the one I fixed.

                            EDIT: Actually, first check the board carefully for any broken\burnt parts, traces etc. Just in case.
                            Last edited by Agent24; 04-13-2015, 04:04 PM.
                            "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
                            -David VanHorn

                            Comment


                              #15
                              Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                              Originally posted by Agent24 View Post
                              Yes I got it running. It's been working without issue for 3 weeks now.
                              Can't say the same for the Windows 10 preview I tried out on it though!

                              Start by checking all the voltage rails. Maybe start with NB_VCC1P1 and VCC_DDR since those two were at fault on the one I fixed.

                              EDIT: Actually, first check the board carefully for any broken\burnt parts, traces etc. Just in case.
                              not Eactly to sure about the Thing you are talking about. i really dont understand anything. is there someone i can take it to ... that is able to do this kind of thing?

                              Comment


                                #16
                                Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                                You'd have to search around and find someone in your country that does component level computer motherboard repair, good laptop repair places are likely to have that capability and may do it for you, but if you wanted to get a professional to do it, it may make more sense to just buy a new board considering the possible repair cost.

                                You could perhaps explain to them about this thread so they can test for the same problem which would speed up the diagnostic process and make it cheaper... *IF* your board has the same fault. But a no-POST issue can be caused by many different things that could go wrong, so it could well be yours has something else wrong with it.

                                Without you doing an inspection, checking voltages etc, it's impossible to say if yours has even remotely the same problem as mine.
                                Last edited by Agent24; 04-13-2015, 10:48 PM.
                                "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
                                -David VanHorn

                                Comment


                                  #17
                                  Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                                  Originally posted by Agent24 View Post
                                  You'd have to search around and find someone in your country that does component level computer motherboard repair, good laptop repair places are likely to have that capability and may do it for you, but if you wanted to get a professional to do it, it may make more sense to just buy a new board considering the possible repair cost.

                                  You could perhaps explain to them about this thread so they can test for the same problem which would speed up the diagnostic process and make it cheaper... *IF* your board has the same fault. But a no-POST issue can be caused by many different things that could go wrong, so it could well be yours has something else wrong with it.

                                  Without you doing an inspection, checking voltages etc, it's impossible to say if yours has even remotely the same problem as mine.
                                  mine starts up but nothing on screen. when it starts up the cpu fan stops spinning and the ZIFF Socket doesnt heat up O__0

                                  strange ey ?

                                  Comment


                                    #18
                                    Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                                    yeah The Motherboard was Starting UP........ but now its not starting...... it hasnt Started before and then after leaving it or i use to take it to a guy in a internet Cafe with experience but he only use to Swap RAM Modules and use the cpu on His old Boards and sometimes take out the CMOS Battery and then it would work with a Different Power Supply..... SO i dunno if This is Same Problem you were having but This is 1000% A COMPLETE un-deniable motherboard Short would you say so ?

                                    its happened like more than 6 times.

                                    Comment


                                      #19
                                      Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                                      I don't know. Without testing all the voltage rails and checking them for shorts, checking the control signals from the SIO chip etc, it's very hard to say. If it runs at all, I would expect there is no short, but I can't be sure. Sounds more like some other, intermittent issue.

                                      If you don't know how to check these things then you either have to learn, or just take it to a good repair shop, which is more advisable. A motherboard is probably not the first thing you should attempt to repair, the diagnostic is tricky enough, but more to the point, you need good soldering skill to deal with the SMD components properly.
                                      "Tantalum for the brave, Solid Aluminium for the wise, Wet Electrolytic for the adventurous"
                                      -David VanHorn

                                      Comment


                                        #20
                                        Re: MSI DKA790GX \ MS-7550 v1.21 - Do I have a short on NB_VCC1P1?

                                        Hey Agent24...

                                        Hopefully you're still around. Thanks for the post on this m/board. There may be some hope for me.

                                        I have the same board and the same problem except It will power down by holding the switch. Fans and drives power up. CPU makes heat. No display on either the integrated or PCI vga.

                                        My question is, if it's not too much of a hassle, where do you check these voltages on U7 and U23 and can it be diagnosed without a scope?

                                        Thanks.
                                        Lee H.
                                        WI/USA

                                        Comment

                                        Working...
                                        X