Motherboard power phasing?

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Maxxarcade
    Badcaps Veteran
    • Jul 2006
    • 973

    #1

    Motherboard power phasing?

    I read a lot here about the number of phases on VRM circuits for motherboards. I'd like to know more about the differences between the different VRM circuit designs between different boards. How do I tell how many phases a particular board has? Don't some boards have multiple sets of MOSFETs but only one phase?

    I understand what the reason for multiple phases is (distributed load) but like I said, how do I know how many phases are actually on a given board? I should know this because I repair boards for a living, but I guess it's not something I've ever needed to know. But now I'm just curious, and it's always good to understand as much as possible about the things you work on. What I'd really like to see is some schematics of typical circuits.

    Also, Anyone know a good cheap source for 40mm x 10mm fans for these teeny tiny Enhance brand power supplies that I'm working on? Today I had to completely recap one and replace the seized up fan after only ~1900 hours on the machine. Darn Merit touchscreen games with crappy cooling... Doesn't help that all the caps are Teapo and Jamicon either
  • gg1978
    Badcaps Veteran
    • Dec 2004
    • 431
    • USA

    #2
    Re: Motherboard power phasing?

    Well, the easiest way is to look for a set of toroid inductors (most common) that each have a set of FETs by them.. Some boards have 2 FETs per phase, some have three, which typically have two as bottom FETs, and one as the top FET, or even four FET's, with two top FETs and two bottom FETs.. There is also typically a single inductor at the input of the VRM itself, next to the input caps, that serves as an input filter for the VRM, in conjunction with the VRM input caps.. Then you have your FET/VRM phases, then the output caps, which are typically right by the CPU, and the high frequency filter caps, which are typically under the CPU in the "hole" in the center of the socket. If the board has no hole, then they'll probably be on the other side of the board.

    Most common number of phases on consumer level boards is typically three phase.. Enthusiast boards can have more, ranging from 3 to 8 phase, with phases of over 6 being less common and more likely to be found on high end boards.



    so the way the power gets to your CPU is this:


    + 5V or +12V ---> VRM input inductor---> VRM input caps -> multiple FET/inductor phases ----> VRM output caps..


    I've attached a quick sketch i did in PAint.. T is for Top fet, B is for bottom FET.
    Attached Files
    Last edited by gg1978; 10-05-2006, 07:07 PM.

    Comment

    • gg1978
      Badcaps Veteran
      • Dec 2004
      • 431
      • USA

      #3
      Re: Motherboard power phasing?

      The way it works is that the top FETs pull the inductor node up to Vcc, while the bottom FETs pull the inductor node down to ground.. The duty cycle of the FET's is controlled by a chip that connects to the gate of each FET, either directly or though an intermediate FET driver chip. The controller chip gets feedback from Vcore itself to determine the duty cycle of the FETs. For example, as CPU load inceases, the CPU sleeps less, so it pulls more current, which will require a higher duty cycle in order to keep it going.

      Here's a link to another post i've made with a picture of a Socket A VRM section.

      You can see the ATX +12V connector, then the input inductor, the Input caps (the 16V Rubycons), then of the three VRM phases, you only get a good look at two, as the third is behind the tall "naked" Nichicon HD's and Sanyo Oscons that are the VRM output caps.

      https://www.badcaps.net/forum/attach...achmentid=3572
      Last edited by gg1978; 10-05-2006, 07:19 PM.

      Comment

      • Maxxarcade
        Badcaps Veteran
        • Jul 2006
        • 973

        #4
        Re: Motherboard power phasing?

        I see, it's basically just like the circuit in a common switching power supply.

        Was your pic showing a 3 phase or 1 phase circuit? Looks like one phase with 3 sets of FETs to me. Would higher end multi-phase VRM's have more than one output coming from the driver chip, each to it's own set of FETs?

        Comment

        • Maxxarcade
          Badcaps Veteran
          • Jul 2006
          • 973

          #5
          Re: Motherboard power phasing?

          Err my mistake, the one input line in your pic is the +v input, not the driver line from the PWM IC.

          Comment

          • gg1978
            Badcaps Veteran
            • Dec 2004
            • 431
            • USA

            #6
            Re: Motherboard power phasing?

            Originally posted by Maxxarcade
            Err my mistake, the one input line in your pic is the +v input, not the driver line from the PWM IC.

            Yup, i didn't include the PWM IC..

            On a side note, it's also possible for a board to have more than one input inductor.. Or at least my Gigabyte S939 board does anyway..

            Comment

            • Maxxarcade
              Badcaps Veteran
              • Jul 2006
              • 973

              #7
              Re: Motherboard power phasing?

              I have two Gigabyte GA-K8U-939 boards, one of which I'm typing on now. I was hesitant of the ULI chipset, but it turned out to be great, after I updated the chipset drivers to fix a SATA problem. I wanted to keep my AGP vid card, otherwise I'd have bought an Nforce based board.

              How are the caps on newer Gigabyte boards? I never thought to look at mine before I installed it to see what was on it.

              Also I'm curious, when did switching VRM's start appearing on motherboards? I think up until the later Super Socket 7 boards, linear VRM's were used. I remember how hot the VRM got in my old Shuttle HOT-557 board when I put a K6-3 450 in it

              Comment

              • schoeds
                New Member
                • Oct 2006
                • 6

                #8
                Re: Motherboard power phasing?

                Are these "R47" components VRM inductors and if so would this be considered an 8 phase? If this is correct how are they different from the green and copper coils of the above example?
                Attached Files

                Comment

                • gg1978
                  Badcaps Veteran
                  • Dec 2004
                  • 431
                  • USA

                  #9
                  Re: Motherboard power phasing?

                  Yeah, i've heard good stuff about the Uli chipset, which is why i guess Nvidia bought them.. I've got a pair of the K8NS Ultra 939's, with the before mentioned 6 phase VRM.. One has Sanyo WG for the VRM input caps, and Nichicon HM for output, while the other has Panasonic for input and Nichicon HM for output.. One of them has been going for over a year now, so hopefully it's Nichicons aren't from that bad batch they had..


                  As for how long switching VRM's have been on boards, i stole some Sanyo Oscons from a 1996 vintage Intel VS440FX i blew up by putting a PPro 200 with 1M of cache on it, and overclocking it to 233.. It doesn't have the top/bottom FET architecture, with just a single FET and a snubber diode. As for the type of VRM mentioned above, i've got an Intel AL440LX with a single phase VRM, with 4 FETs total, two top and two bottom..

                  As for SS7, yeah it was the later ones that started to get switching VRM's, with the older ones being linear ones that ran hot..

                  Comment

                  • gg1978
                    Badcaps Veteran
                    • Dec 2004
                    • 431
                    • USA

                    #10
                    Re: Motherboard power phasing?

                    Originally posted by schoeds
                    Are these "R47" components VRM inductors and if so would this be considered an 8 phase? If this is correct how are they different from the green and copper coils of the above example?

                    Looks like 8 phase to me.. Can't see the FETs very well as they're under the cooling fins.. The input inductor is the unmarked vertical one by the 8 pin +12V connector. As for the R47 marking, i'm guessing that they're 4.7uH inductors..

                    Comment

                    • Maxxarcade
                      Badcaps Veteran
                      • Jul 2006
                      • 973

                      #11
                      Re: Motherboard power phasing?

                      I've got an HP Netserver E40 with the 1MB PPro 200... I tried running it for a NAS using NASLite+, it ran great when it worked but kept crashing with a Kernel panic every 3 days or so. Too bad cause I really wanted to use it, it's built like a tank.

                      I tried to OC it to 233 when I got it, but the BIOS just says there is a DIP switch set wrong and won't let the system continue past that point.

                      I was also surprised how well such an old system runs Windows XP. Loading is a bit slow with the 5400RPM hard drives, but not bad for GUI speed. But it has 320MB of SIMMS in it

                      Comment

                      • gg1978
                        Badcaps Veteran
                        • Dec 2004
                        • 431
                        • USA

                        #12
                        Re: Motherboard power phasing?

                        Hehe, awesome.. I would have thought they would have disabled that setting in hardware.. I had a PPro 200/256K with 256M of RAM running XP RC1 back in '01.. It ran respectably well, given it had a 7200RPM WD Expert HD in it.. Of course said HDD was only running in PIO Mode 4 (16.6M/second)..



                        Originally posted by Maxxarcade
                        I've got an HP Netserver E40 with the 1MB PPro 200... I tried running it for a NAS using NASLite+, it ran great when it worked but kept crashing with a Kernel panic every 3 days or so. Too bad cause I really wanted to use it, it's built like a tank.

                        I tried to OC it to 233 when I got it, but the BIOS just says there is a DIP switch set wrong and won't let the system continue past that point.

                        I was also surprised how well such an old system runs Windows XP. Loading is a bit slow with the 5400RPM hard drives, but not bad for GUI speed. But it has 320MB of SIMMS in it

                        Comment

                        • Maxxarcade
                          Badcaps Veteran
                          • Jul 2006
                          • 973

                          #13
                          Re: Motherboard power phasing?

                          Ya the IDE controller is crap (PIO4) on that HP too. But this one came stock with an Adaptec SCSI card. Too bad I can't afford to put a pair of big fat server grade SCSI drives in there. It would make a nice server for how old it is.

                          And I really love NASLite... Lets you use huge drives in old computers since it accesses the drives directly and not through BIOS. I had a pair of 320GB Western Digital drives in the old HP, running on the stock onboard IDE. NASLite-2 looks even better with many more features, but I haven't got around to upgrading yet. My NASLite 1.5 server is currently a PIII 550E system with 512MB ram and the above mentioned 320GB drives. So I have 640GB, enough space to back up all my computers, and also store most of my DVD collection, uncompressed, for on-demand viewing in my home theater via the LAN. No wear and tear on my DVD's or DVD player

                          Comment

                          Related Topics

                          Collapse

                          • momaka
                            Corsair CX750M [Model 75-002019] - burned output toroid inductor
                            by momaka
                            Looks like I may need a little help from the PSU experts (or anyone really! ) I have a Corsair CX750M (Model 75-002019) that I picked up for free about 4 years ago. This is the PSU:



                            So here’s what’s strange about this one: it appears to work normally (normal output voltages) and any PC is stable with it. However, after a while (typically 10-20 minutes, but time can vary depending on the load), there is a sweetish smell of burned magnet wire insulation emanating from the PSU. Upon opening the PSU after this happened (many times), I was able to confirm...
                            09-19-2021, 08:44 PM
                          • Bartoloni
                            Lenovo S145 -15iwl vcore inductor warm.. looking with oscilloscope there is a spike.
                            by Bartoloni
                            Hi, i have a Lenovo S145 that after 20 seconds shutdown and restart (it's working correctly on thse 20 seconds)
                            everything seems ok, but on Vcore inductor ( this is an i3 with just 1 switching power supply for CPU)

                            there is an output to teh CPU with a spike.. seems that inductor can't handle the input...
                            replaced the capacitor (330) and the inductor (R15) without any change
                            any idea? very thank you!

                            input of the inductor seems as usual... only strange thing is that the upper/lower voltages seems to be 21V (main power line is 19.3) ... but not sure this...
                            12-09-2024, 12:42 PM
                          • badcap111
                            How to read inductor specs?
                            by badcap111
                            I have a broken samsung galaxy s10 with missing inductor. I want to replace the missing inductor and I have found some vendors. But I don't understand how to read the specifications of the inductor. The specs are:

                            INDUCTOR-SMD;470NH,20%,2520,T0.8,0.029OH

                            I understand that the inductor is 470 nH with 20% margin, and the size is 2520 or 1008 (imperial measure), but what do T0.8,0.0290OH mean? I can't find any explanation on the web.
                            09-12-2023, 01:25 PM
                          • silver60618
                            Blown Inductor on WD My Passport
                            by silver60618
                            Hi Guys, my WD My Passport drive powers on, then powers off. Ive taken out the motherboard and identified a blown Inductor. Im trying to recover my data only

                            Reading between leads show nothing. If I removed this Inductor, what would be a suitable replacement? am I correct that its a ferrite Inductor is to just filter noise?
                            Ive seen people just bridge them, any harm in replacing it with a differnt value Inductor?

                            Lastly am I correct in thinking there is no polarity on the smd component?

                            Thanks!...
                            04-05-2024, 09:54 AM
                          • Gadjet
                            PS5 Slim 3 sec Blue Light Issue
                            by Gadjet
                            I have a system I'm repairing that came in for overheating, which also had a busted USB port in the back.
                            Reapplied new liquid metal to the system and got the BLOD.
                            I then found a bad Fuse on F3501 and replaced it.
                            My client also found out that the system had fallen, and that's how the USB port was broken.

                            I found a short on the cap in the photo, but the cap isn't shorted off the board. The cap isn't shorted on the board if I remove the black chip next to it (I think it's a diode, not sure), or if I remove the inductor.

                            UART readings say: Error Code:...
                            05-11-2025, 12:26 PM
                          • Loading...
                          • No more items.
                          Working...