Matching VRM MOSFETS on a motherboard

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • mockingbird
    Badcaps Legend
    • Dec 2008
    • 5484
    • -

    #1

    Matching VRM MOSFETS on a motherboard

    I have in my posession a Foxconn G45A01 out of a Dell tower...



    I've already filled in the empty polymer capacitor positions with exact equivalents (Save the batch number obviously), and have replaced almost all the off brand electrolytics.

    Above the CPU socket, there is a row of three encapsulated coils and three TO-252 MOSFETs. (There is also a row of three MOSFETs directly above, but ignore them for the sake of this post, they're a slightly different model - AOD452 - and not relevant).

    Now as you can clearly see, directly above each of the three TO-252 MOSFETs (Their model is AOD472) is an accompanying pad which is obviously bare. I'm assuming the pad is connected electrically in parallel with their assumed counterpart because there is a clear trace connecting both gate pads.

    Now, these particular FETs are no weaklings, they're pretty highly rated and I could probably get by just fine leaving things the way they are, but I wanted to populate them since I was filling empty capacitor spots anyway and thought I'd see if I could source some AOD472. Lo and behold, this part isn't listed on A&O's website nor is it shown to be stocked anywhere (Though it is listed on Digikey, but the datasheet provided curiously shows it as a DPAK part eventhough it is clearly a TO-252 part).

    So I naturally bought some of them from Aliexpress, but eventhough I am principally opposed to buying semiconductors from China, I assumed that since this part was so obscure, the Chinese wouldn't go to the lengths to counterfeit it. I even bought from a seller who showed an evidently used part for the item picture (Used is a lot better than fake). Unfortunately they sent me obvious fakes.

    So I have several questions:

    1) Why doesn't A&O have this listed?
    2) Was this subcontracted out? If so, how can I find out who actually made it?
    3) If I can't find an exact equivalent, should I try to match it as closely as possible with an equivalently spec'd part or would the slight variations negate any benefit of adding them in the first place?
  • ReverendJones
    Resurgam
    • May 2010
    • 49

    #2
    Re: Matching VRM MOSFETS on a motherboard

    Even though Alpha & Omega doesn't list the part, they do have a data sheet online: http://aosmd.com/pdfs/datasheet/AOI472A.pdf
    Perhaps a special order part for Foxconn?

    I see that there are some listed on ebay: http://www.ebay.com/itm/251242738068
    Machina improba! Vel mihi ede potum vel mihi redde nummos meos!

    Comment

    • mockingbird
      Badcaps Legend
      • Dec 2008
      • 5484
      • -

      #3
      Re: Matching VRM MOSFETS on a motherboard

      The 472A differs from the 472 handling only a maximum of 46A @ 30v as opposed to 55A.

      Comment

      • severach
        Badcaps Legend
        • Aug 2007
        • 1055
        • USA

        #4
        Re: Matching VRM MOSFETS on a motherboard

        Alpha & Omega AOD472 are common on Dell boards. I have so many I don't even collect them any more unless it's an easy solder board.
        sig files are for morons

        Comment

        • mockingbird
          Badcaps Legend
          • Dec 2008
          • 5484
          • -

          #5
          Re: Matching VRM MOSFETS on a motherboard

          Interested in selling a few to me?

          Comment

          • momaka
            master hoarder
            • May 2008
            • 12175
            • Bulgaria

            #6
            Re: Matching VRM MOSFETS on a motherboard

            I wouldn't bother unless those MOSFETs run extremely hot or you just want to do it as an experiment.

            Comment

            Related Topics

            Collapse

            • BlackScreen_22
              Harman Kardon sub TS11 (ds15 amplifer) blows fuse and mosfets
              by BlackScreen_22
              Hi, I'll start by saying that I'm not an expert and I've reached a point where I don't know what to try/do anymore...
              Starting from the beginning: I opened this subwoofer for the first time with a very strong smell of burnt electrical components, I realize that on the amplifier board there is a hole between V+ OUT and GND about the size of a finger, the damaged components were the shorted mosfets, two damaged zener diodes and two capacitors. I bought everything, I fixed the hole by cutting the pcb and putting a piece of a perforated board.
              I had no problems connecting mosfets and capacitors,...
              08-27-2024, 12:46 PM
            • kotel studios
              Intel D865PERL - no Vcore after replacing MOSFETs and MOSFET drivers
              by kotel studios
              Hi,

              Finally got time to fix this board up.
              At first it didn't give any POST codes. Turns out it was an unsupported Celeron D cpu. After dropping in an Celery 2.6gHz the CPU came out of reset with clock signal, but still no post codes. For a while it did work when I applied pressure to the ADP3168 chip, but on next reboot the MOSFETs started to burn. Then I left it as is for months.

              Now I came back prepared. Replaced the bad MOSFETs and the ADPxx18k drivers. Now I don't get any MOSFETs blowing up issues anymore. But sadly, the Vcore is missing. The whole VRM isn't...
              04-22-2025, 05:08 AM
            • flerre
              Gainward Phantom GTX 980 - mosfets w/ integrated driver NCP5369
              by flerre
              Hi!

              Managed to get a Gainward Phantom GTX 980 in an auction for a bin with components.
              No video out so I started fault finding and found a blown fuse for 3 mosfets. Started checking for shorts and found what I think is a short through all of those 3 mosfets; 0.1 ohms between "VIN" input and "VSWH" output. All mosfets in this gpu are the same ones (NCP5369) with integrated drivers and when I tried measuring the other ones through these ports they were somewhere in the kilo-ohms range.

              So, could I safely assume all of these 3 mosfets are...
              03-13-2022, 12:14 PM
            • NanoTesting
              Lenovo Ideapad 15IIL05, turns on and off with charger BUT works by injecting 20v after main mosfets
              by NanoTesting
              I got a 15-IIL05 that failed to turn on BUT IT DID TRY TO TURN ON and then turned off after 30+- seconds, i disassembled it and found a screw loose inside the case.
              With charger it turns on, fan spins but gives no image.
              Tested main mosfets no short, tested 3.3v,5v,DRAM, CPU etc and they were present.
              Reflashed BIOS, same problem.
              Replaced both main mosfets.
              Checked main mosfets again and i noticed that they closed once the laptop shut down, so i decided to bypass them and used a bench PSU to check how much power the laptop was drawing while trying to turn on, 20v@3A,...
              10-30-2024, 10:48 PM
            • debruina
              Pioneer VSX-859RDS with all mosfets gone
              by debruina
              Hi,

              I am working on a pioneer VSX-859. It started with only one problem but i ended up with a lot more. Probably caused by me

              Anyway it started with a DSP NG message on the display. That one was easy and was caused by a bad 5v rail. Replaced the capacitors and the message was gone. Everything seems to be working except for a blob sound every time i switched from source. Eventually i tracked that down to a leaking battery on the analog board. Repaired some via's and lines and everything worked ok. That same circuit also caused a not working DCdetect protection. ...
              08-29-2023, 11:54 AM
            • Loading...
            • No more items.
            Working...