Announcement

Collapse
No announcement yet.

ECS SF2/661FX no post

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

    ECS SF2/661FX no post

    Another story, the same characters...
    The mobo looks ... GREAT! All capacitors are Panasonic
    Same look : https://www.badcaps.net/forum/showth...ht=sf2%2F661fx
    But... no beeps, no lights, no post. I tried different cpu, RAM, video (onboard, agp or PCI) without success.
    I learned something here, so I checked MOSFETs and seem to be in order, except MF11 and MF12, which are near the ATX connector. They do not give continuity but I remembered what he said @momaka here and I stopped: https://www.badcaps.net/forum/showpo...41&postcount=7

    What should I try?
    Thanks.

    #2
    Re: ECS SF2/661FX no post

    You didn't mention PSU, although I hope that would be the first suspect and you missed it out of the list. Just checking

    Comment


      #3
      Re: ECS SF2/661FX no post

      Thanks. 3 differents psu, same simptoms...

      Comment


        #4
        Re: ECS SF2/661FX no post

        Does the board beep an error code if you run it without any RAM? If NO, then it may not be even getting to the BIOS.

        But before we dig deeper, first let me ask: when you tested the motherboard, did you have the CPU fan plugged in? Some older motherboards have a feature to make the board not boot if a CPU fan wasn't detected.

        Also, while you are at it, check all of the voltages on the motherboard. That is, check CPU V_core, DDR RAM Vdd (2.5 - 2.6V), RAM Vtt (1.25 - 1.3V), AGP and/or Northbridge (usually 1.5V). To do that, it would probably be best (ans easiest) to check the voltage on all of the input/output pins of all TO-252 and SOT-223 devices.

        Comment


          #5
          Re: ECS SF2/661FX no post

          Verification is the first choice every time. Of course this time and I checked it thoroughly mobo, which actually is an ECS OEM for AcerPower F1 / Aspire T310.
          The buzzer is on pcb, no option to mount another extra-buzzer, no sound. At all.
          I did sample it without anything attached, only with onboard video card. I mounted consecutively the CPU, RAM with AGP or PCI video card.
          After each piece added/dismantled I tried to power on the mobo. After all possible combinations the mbo still dont power on.
          Very hard to check the Vcore, Vdd, Vtt or AGP because I dont know the datasheets. Already verifed some IC and I got values. After all I'll put a list of IC measured.
          Thanks again for your suggestions.

          Comment


            #6
            Re: ECS SF2/661FX no post

            I tried to be as concise but I still have to make some additions.
            The bad news comes later ...
            Attached Files

            Comment


              #7
              Re: ECS SF2/661FX no post

              Latest corrections, because I cannot edit the post...
              Attached Files

              Comment


                #8
                Re: ECS SF2/661FX no post

                I was going to use that picture to make you a diagram with the voltage locations, but looks like you already beat me to it .

                Anyways, here goes a looooooooong post...
                Using your latest picture as a reference:
                https://www.badcaps.net/forum/attach...2&d=1459966140

                Power-Related

                ---- MOSFETs "a", "b", "c" (06N03LA) and "d" "e" "f" (part number?)
                These are the MOSFETs that regulate the voltage for the CPU (called CPU V_core or CPU Vcc).
                * MOSFETs "d", "e", and "f" are on the upper side (connected between 12V rail from 4-pin CPU connector [Drain] and CPU V_core [Source]).
                * MOSFETs "a", "b", and "c" are on the lower side (connected between CPU V_core [Drain] and ground [Source]).

                ---- #2, 3, 4 (ADP3418k)
                This three ICs are the MOSFET drivers for the CPU voltage. They turn ON and OFF the upper and lower MOSFETs above (a, b, c, d, e, and f) as required to regulate the voltage for the CPU.

                ---- #1, ADP3180 IC:
                This is the CPU voltage regulator module (VRM) controller. It takes the VID values from the CPU, chooses the right voltage for CPU, and then controls the CPU MOSFET drivers (the ADP3418 ICs), which in turn control the MOSFETs (a, b, c, d, e, and f).

                ---- MOSFETs "m" and "L" (NIKOS P45N02LD)
                These MOSFETs are next to the RAM slots, because they likely regulate the voltage for the RAM. This board uses DDR RAM, and DDR RAM requires two voltages: 2.5-2.6V (often called RAM Vdd/Vddq) and 1.25-1.35V (RAM Vtt).
                Note that there are no inductors or coils in the viscinity of these MOSFETs. That means that each of these MOSFETs is either used as a linear regulator (same principle as a 7805 or 1117 regulator) or as a pass-through switch (the output voltage at the Source is the same as the output voltage on the Drain).
                Based on your measurements, it appears that one of these MOSFETs is a pass-through for the 5V or 5VSB rail. The other MOSFETs then should be stepping that down to 2.5-2.6V for the RAM.

                ---- #8, RT91738 IC:
                This IC is likely used to generate the RAM Vtt (1.25-1.35 Volts) mentioned above (I can't find its datasheet, but I'm pretty sure of that based on your measurements).

                ---- #9, LM324 IC:
                This IC is a quad op-amp. Most likely it is used to control MOSFETs "m" and "L" for the RAM voltage (i.e. the LM324 checks the voltage on the RAM and accordingly drives MOSFETs "M" and/or "L").

                ---- #7, LD1117 IC:
                This is a standard linear voltage regulator. Look for two numbers after the "1117" part. "33" indicates a 3.3V regulator, "25" indicates a 2.5V regulator, "18" indicates a 1.8V regulator, and "ADJ" means adjustable voltage (determined by two resistors usually). Based on your measurements, 1.25V, you probably have either an adjustable type or a 1.2/1.3V regulator. I'm not 100% sure for what it is used on your motherboard, but it could be either Southbridge or SATA ports (if you have them).

                ---- MOSFETs "i" and "j" (APM2055N):
                First, note the round toroid inductor next to the AGP slot - every time you see one of those, you are usually dealing with a buck regulator (basically what the CPU VRM uses, but with just 1 phase instead of 2 or 3).
                Based on your voltage measurements for MOSFETs "i" and "j" (I believe you mislabeled them in your voltage measurement diagram, but I was able to figure it out), neither one of these is a lower MOSFET, because neither one has its Source pin tied to ground. However, also note that black square SMD diode next to MOSFET "j" - this is used in place of a lower MOSFET.

                So, with all of that said, MOSFET "j" and the SMD diode seem to generate a 1.86V rail (as seen on the Source of MOSFET "j") from the 3.3V rail (as measured on Drain of MOSFET "j"). This 1.8V rail is most likely used either by the Northbridge or the Southbridge (I can't say for sure, though).

                From your voltage measurements, we can see that MOSFET "i" takes the voltage from that 1.8V rail through the Drain pin and outputs 1.5V at the Source pin. 1.5V is usually used for the AGP communication and also possibly the Northbridge.

                ---- #10, GS34063 IC:
                This IC is probably the buck controller for MOSFET "j" above.

                ---- linear regulator "k" (APL1085)
                Looks like this regulator takes 5V (most likely from the PSU's 5VSB), and outputs 3.3V. My guess would be that this 3.3V supply is present regardless if the PC is turned ON or not - as long as the PSU is plugged in (check me on that). If yes, then this voltage rail is probably used by the Southbridge. Many motherboards do this, so this "always-on" 3.3V rail is often called 3.3V STB (3.3V stand-by).

                ---- MOSFETs "g" and "h" (NIKOS P45N02LD)
                These appear to be 5V pass-through MOSFETs, based on your measurements. I would guess they are for USB in stand-by and/or possibly WOL (wake-on-LAN).

                ---------------
                Logic and/or NON-power -Related ICs

                ---- #4, IT875F
                This is your LPC/SIO (low-pin count / super input-output) chip, which basically is responsible for the most basic peripherals of the motherboard, such as PS/2 keyboard/mouse, serial and parallel ports, floppy controller, and sometimes hardware monitoring.

                ---- #5 and #6 (GD75232) ICs
                These are used for RS-232 communication for the serial ports. The "232" in the part name and their location (one next to the serial port and PS/2 keyboard connector, the other next to the SIO) is a dead giveaway of what this chip does. Once you know that, you won't even have to look up a datasheet for it next time you see it on a motherboard .

                ---- #12, ICS CA431583 IC
                This is the real time clock (RTC) of your motherboard. It is what provides a reference clock signal output for pretty much everything on the board. An easy way to spot the RTC chip on any board is to look for a crystal oscillator next to it - that would be the shinny component in a metal can right above that chip.

                ---- #13, 93c46w6 IC
                Atmel serial EEPROM. Not sure what exactly it is used for, but probably something for BIOS/CMOS.

                ------------
                And that pretty much covers all of the basics. I left out mentioning the second ICS chip next to the ATX connector. But that probably has something to do with controlling the PS_ON signal from PSU and detecting PWR-GOOD... I would guess.

                Anyways, I hope this is helpful to you - if not for this motherboard, than for some others. Many motherboards from Intel up to socket 775 are setup in a similar way. For AMD, that would be anything up to the AM2 era. The ICs, of course, may be slightly different on different motherboards, and their locations may not be exactly the same either. But once you have a general understanding of what you are looking for (especially in terms of power-flow), it becomes possible to apply the above knowledge to any motherboard .
                Last edited by momaka; 04-07-2016, 10:20 PM.

                Comment


                  #9
                  Re: ECS SF2/661FX no post

                  Three days have elapsed until understand .
                  It's one of the best tutorials for testing motherboard made for beginners or unskilled like me, this little lesson is a great gift.
                  The measurements have made by myself, for one time, based on the datasheets and I dont found issues on basic values (that's my level of understanding). It's difficult to interpret the values that are changed dynamically or within certain limits.
                  The moment when is has happened the disaster with short on pins 5-6 of ADP3418k #2 IC it did more hesitant.
                  The answers for @momaka:
                  - d, e and f driver mosfets are C4442;
                  - RT91738 IC datasheet is on the second image (
                  https://cdn.badcaps-static.com/pdfs/...52312beedb.pdf)
                  - i and j IC are APM2055N; yes, I messed up the order in the list; I asked for a correction without any feedback from forum board .
                  Because time is no longer an ally, I hardly found time to check CPU VRM mosfets. They are ok, so the conclusion is ADP3418 is dead. I'll try to replace him, but I have to do research to find a replacement.
                  Again, the time is not my friend these days...

                  Thanks again, sir.

                  Comment


                    #10
                    Re: ECS SF2/661FX no post

                    Originally posted by alindumitru46 View Post
                    The measurements have made by myself, for one time, based on the datasheets and I dont found issues on basic values (that's my level of understanding).
                    Well, your CPU V_core output voltage is low. Should be 1.75V for Willamate CPUs, 1.5V for Northwood, and 1.4V for Prescott.
                    But this low voltage is probably due to either that bad ADP3418 or a faulty MOSFET... or both. So we definitely need to fix that first.

                    Originally posted by alindumitru46 View Post
                    - d, e and f driver mosfets are C4442
                    Can't find a datasheet for that, though.

                    Originally posted by alindumitru46 View Post
                    - RT91738 IC datasheet is on the second image (http://www.alldatasheet.net/datashee...EK/RT9173.html
                    Ah yes, sorry about that. Now that I think about it, I actually did find a datasheet for it even in my previous post above, and I even explained what that IC did. Anyways, I guess I messed up my post a little, because I did say it is the RAM Vtt generation IC.

                    Originally posted by alindumitru46 View Post
                    - i and j IC are APM2055N; yes, I messed up the order in the list; I asked for a correction without any feedback from forum board .
                    No problems, I figured it out.
                    I'm surprised you didn't get a response. Did you send a PM to one of the mods? (i.e. c_hegge, shovenose, etc.)

                    Originally posted by alindumitru46 View Post
                    Because time is no longer an ally, I hardly found time to check CPU VRM mosfets. They are ok, so the conclusion is ADP3418 is dead. I'll try to replace him, but I have to do research to find a replacement.
                    Let me know if you can't find a replacement.
                    I think I still have some ADP3418/NCP3418 left over as well as ADP3118. Both of these are compatible with each other. ADP3120 may be another option. I had a bad motherboard with a bunch of dead ADP3120 ICs. The ADP3418 worked great in their place.

                    Comment


                      #11
                      Re: ECS SF2/661FX no post

                      In few hours I get back home and I'll check that if I find that ADP3418 and mosftes on my bad motherboards, identically to those concerned.
                      I did not find datasheet for the MOSFET C4442, more try today.
                      To contact the moderators have used the contact form, I thought is not neccesary to ask via PM.

                      Comment

                      Working...
                      X