1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Power supply Qs/Confusion: Running duals & rail draw... (grab a drink, long read)

Discussion in 'Cases, Modding & Electronics' started by Formula350, Feb 6, 2010.

  1. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    There's a fair amount of backstory to this, to help anyone who is up to the task of aiding me, if you want to skip it all though, third paragraph is where to start :p

    So I'm running basically the same rig as my sig except I got a DFI LanParty UT NF4-DR Expert that was "bad" (would BSOD when trying to load or install an OS, which I also encountered upon getting it), which miraculously started working shortly after the rest of my PC did, long story short I pulled my PC out of storage and it wouldn't even post no matter what I did and then all of a sudden one day it worked like nothing had occured. Anyways I'd been previously toying with my Gigabyte board with trying to squeeze out more performance than previously, with little luck, and noticed a new BIOS was out and tried it. Got me thinking about trying to flash an older BIOS to that DFI board, and so I started to put some spare parts in it, enough to get it running. Well at first I was having problems getting it just to power on, but finally did, and was monkeying around in the BIOS before I went about flashing it (sadly no in-BIOS option to flash like GB had) when I thought it'd be interesting to run the built in version of MemTest and see if that caused any crashes. Well it didn't, and plugging in a spare HDD that had an OS on lead to a BSOD-free boot! :eek: So now I'm running this mobo :)

    Since I knew two things about the components I have, being that the CPU can do 2.8ghz and the memory can stably do 450-480mhz (which I'm going to post on after this, as I've encountered a huge oddity) so with that knowledge I figured I'd see if this board offered any other hidden potential of the two since it gives a ton more voltage and tuning options... Well the first hiccup I got, which I didn't with the GB board, was crashing when bumping up the HTT (FSB if you will, even though yes it's not) randomly leads to a crash well below the max of either CPU or Mem, first one encountered was around 225mhz. Thought possibly was due to not dropping the HT from 5x to 4x, so I did and still happened. Well fine, I chucked in 220mhz in the BIOS and booted into windows where I continued on overclocking. The reason I do it in windows is A) I've had MUCH success doing it this way in the past, B) so much damn easier than moving it up a few MHZ via BIOS and testing, reboot, adjust, test. I use primarily A64Tweaker, SuperPi and Sandra, with SysTool tossed to change the QFS option (dunno wth it is lol) and nTune which is the only app I've found that lets you change the HT multiplier in Windows :p Anyways, the limit I've seemed to reach on the DFI board was a very sad 260mhz, far shy of what I expected given the ability to crank up the voltages on many more things than the GB allows which was able to hit 280mhz. That in itself is an oddity as apparently the HTT speed isn't really pushing anything like the FSB did, and one could theoretically run it at 500mhz with CPU multiplier of 4x and memory set at whatever divider required for 200mhz (DDR400) and it should run the same as if it was @ 200mhz 10x.... That left me wondering if it was my old and trusty Antec PSU finally showing it's 6+yr age lol That and it's only 430w which 410w being on the used rails AND that'd be it's "perfect circumstances" rating. So I wired up 300w I had laying around to take some load off....

    I know there are 'proper' ways of doing dual PSU rigging, and a fair amount of them at that, but I am on the "cheap" right now, where $5 for a relay or $10-15 in other parts is just not feasible. So I what I did was just run the PWR_OK and PS_ON (green and brown, typically) into the PWR_OK and PS_ON of the Antec, which lets both turn on at the same time. I then pinned a few of the 300W wires in place of the Antec's, which were the last 4 of a 24pin ATX (as my Antec is only a 20pin which the GB board supported), the ATX12V and the EXTRA power header that uses a floppy molex. The Antec then supplies the 20pin ATX, the 2 HDDs, 1 CDRom and floppy. Video card is picked up by the 300W. This is where I get a bit confused: technically the 700W (600W max, to be fair) should be able to supply this rig fine, which it does, but the 5v rail voltage is STILL 4.65-4.7v. If I hot-wire just the 300W on with the computer off, while still plugged in, I get a reading of 5.02v which also has the computer in partial-on. By that, I mean there is power virtually everywhere, the Antec is basically also on (either the 300w is charging it making it seem on due to fans, or it actually is due to some weird phenomenon when voltage is sensed it auto kicks on), but the system is not running and the little DDR power LED (DFI thing) is also not on. If I leave the 300W hotwired and press the power button, the voltage stays over 5, the DDR light turns on BUT the system doesn't start. Once I remove the PWR_OK from PS_ON, the voltage drops to the ~4.7v and the system boots normally.

    I just can't figure out what is up with the voltage drop. I tested this setup with the GB board a few weeks ago with the system running Intel Burn Test and GPUTool rendering, with the video card @ 635/830 and CPU/MEM @ 2.4 and 440, and I'm pretty damn certain it was reading over 5v... I've even bridged 2 molexs together between the Antec and 300W, in an attempt to equalize any load draw, but nothing seems to do the trick :\

    Anyone know why I'm experiencing these voltages? I haven't tested the 3.3v rail, but BIOS is showing 3.3v on it, and 12v is showing 12.02v with a 12.12v reading off the Antec and a 11.92v reading off the 300W (as it's the one powering the additional ATX12V connector). I'm at a loss.

    Sorry for rambling, just wanted to get out all the info I could, but I can understand if that might have lead some bits to being tricky to understand heh Be that the case, assuming anyone read all this, I'll be happy to clarify anything :)

    Thanks :toast:
  2. sonnytiger

    sonnytiger New Member

    Joined:
    Dec 30, 2009
    Messages:
    204 (0.12/day)
    Thanks Received:
    10
    Location:
    Lower Sackville, NS
    Not enought Amps in the 12v Rail? Or even the 5v Rail if its old enough, pairing a new powersupply with old hardware can be bad as old hardware typically uses the 5v rail more, might be something to look into.
  3. xrealm20

    xrealm20 New Member

    Joined:
    Jan 29, 2010
    Messages:
    467 (0.29/day)
    Thanks Received:
    129
    Location:
    Houston
    Check your 5v output on the antec using a multimeter if you have one available. Your motherboard could be reading the voltage incorrectly. And besides, 4.7v on the 5v rail is just BARELY within specficiation. So if your motherboard is reading the voltage too low, say by .1 or .2 volts, it would be perfectly in spec. Also that would mean that when it's reading 5.1 on the 300w power supply, its over spec and that's more dangerous than under spec.
  4. angelkiller

    angelkiller

    Joined:
    May 16, 2008
    Messages:
    1,258 (0.56/day)
    Thanks Received:
    225
    Location:
    North Carolina
    Ok. I missed the need for using two PSUs. Either PSU by itself is powerful enough for your system. All you're running is an Athlon 64 and a X1950 Pro. That can't be more than 150W DC on load. I don't think there's a need for connecting the two PSUs together as it unnecessarily adds complexity

    It appears to me that one PSU simply can't give a stable 5v. The PSU that gives power to the graphics card won't have it's 5v rail used at all. I'm a bit confused, but have you measured the 5v rail from both PSUs when they're running? Get a measurement when each PSU has nothing connected to it (literally nothing), and also when each PSU is connected to your system (at idle).

    The difference between the no-load reading and the with-load reading will tell you how bad of a problem you have. A small difference means that the PSU gives a stable voltage reading, but it's just low and a big difference means that the PSU can't give a stable voltage because the load is too high. That's bad because the load you used (your idle system) is very small)

    Agreed, all voltage measurements should be taken with a multimeter.

    I thought the ATX spec was like +/- a certain number. And I thought that getting a voltage higher than what you want is equally bad as having a lower voltage. I could easily be wrong though.
    Last edited: Feb 12, 2010
  5. Clement

    Clement New Member

    Joined:
    Feb 11, 2010
    Messages:
    192 (0.12/day)
    Thanks Received:
    26
    Location:
    SouthEast Blue Mountain, Pa
    As a side not, Are your two PSU's grounded together?

    What were the results of using a multimeter?
    Last edited: Feb 12, 2010
  6. xrealm20

    xrealm20 New Member

    Joined:
    Jan 29, 2010
    Messages:
    467 (0.29/day)
    Thanks Received:
    129
    Location:
    Houston
    Higher or lower voltage is bad news. And for the ATX 5v spec runs +/- 5% or .25 volts -- so 4.75 to 5.25
  7. xrealm20

    xrealm20 New Member

    Joined:
    Jan 29, 2010
    Messages:
    467 (0.29/day)
    Thanks Received:
    129
    Location:
    Houston
    Another question -- what Antec PSU do you have and what are the specs on the other "300 watt" PSU?
  8. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    OK, I can understand people missing that I did take readings with the multimeter, given how gobbled together that post was lol I've never actually ran the PC off just the 300W, as an FYI.


    The hardware IS newer than the PSU, but only by about a year. I initially got the Antec in 2004 or 2005 when I upgraded my AthlonMP system to a newer mobo (and CPUs) that required a minimum of 32A on the 5V rail.
    That being said, here is the + ratings.
    5v - 36A
    12v - 20A
    3.3v - 28V
    5v SB - 2A


    The 4.7v reading is what the sensors read, and I've never taken them as being correct anyways, so all my readings that I actaully go by are via multimeter. That reading would be (at least in dual psu mode) 4.83v actual with the board reading 4.7v.

    Antec with only one HDD (the HDD is sitting under the tray the mobo is mounted to on the bench, too much trouble to disconnect):
    5v = 4.93v
    12v = 12.12v

    Same with a synthetic load (12v automotive brake light bulb, not many watts) on each rail:
    5v = 4.73v (12v reading of 12.13v)
    12v = Thinks I'm shorting it and shuts the PSU down.

    With 1HDD, 1 blower fan, 1 CDRom with synthetic load:
    5v = 4.83v

    I figure it reads more with the higher load because it's sensing more so it'll provide more current, but that's just a guess...

    The only reason I thought I might need a second PSU is due to A) the age of the PSU, B) my PC initially not working (would power on but no POST or even error beeps) and C) after the computer finally working again for whatever reason, I was having issues getting back to the same OC speed as before I packed it away.

    Yes, that was the main thing I did. I thought it foolish to run 2 PSUs together and not have them running a common ground. That was also why I suspected the voltage being lower because they didn't share the -12v and -5v connection as well, but after joining them together, turned out there was no voltage difference.


    That must be a revised rating, last I remember was 10%, which is sadly quite a bit heh. This PSU isn't new enough for 2.0+ specs though, if that 5% is for that.


    It's the TruPower (TruePower?) 430W. The 300W is some junk off-brand I got in a PC I was given.
    5v - 20A
    12v - 8A
    3.3v - 15A
    5vSB - 0.72A

    No load readings:
    5.38v
    11.44v

    Lightbulb load on just 5v rail:
    5.15v
    12.27v

    On just the 12v rail:
    5.43v
    10.94v

    Now, in my experiments with trying to buck voltages using a PSU as a bench-top power source I stumble across the oddity that running the PSU in "230v" input mode resulted in different readings for the 5v and 12v rails, and yes the input voltage is still 115v. With that said, I've not ran the computer in 230v mode off the Antec, but I have ran the 300w in 230v with it supplementing the Antec. Here are the 300w readings just by itself in 230v mode:
    No Load:
    5.25v
    12.04v

    Lightbulb load on just 5v rail:
    5.15v
    12.27v

    On just the 12v rail:
    5.25v
    11.77v

    I suspect it results (at least on this crappy PSU that could have corners cut) in better efficiency due to the rating of the PSU input being 6A @60Hz for 115v and 3A @50Hz for 230v. Yet I'm unsure as to the long term effects on the PSU by running it under load on 230v setting. For kicks I tested the Antec the same way, but the 5v under load drops to 4.53v, so it's seeming like a proper PSU shouldn't result in the same readings heh

    I would've replied sooner BTW, but I just got the email saying there was one reply :ohwell:
  9. xrealm20

    xrealm20 New Member

    Joined:
    Jan 29, 2010
    Messages:
    467 (0.29/day)
    Thanks Received:
    129
    Location:
    Houston
    First off, Make sure you put the PSU's back to 110 Volt mode. I've burned up a PSU by having the voltage set incorrectly - Also, when the PSU failed, it caused damaged to the connected components.

    ok -- so after that being said --

    With just the Antec 430 PSU are you able to get the system to boot and run regardless of the voltage?
    How many hard drives / optical drives do you have connected to the psu?

    Are you trying to power the system in your specs?
  10. Clement

    Clement New Member

    Joined:
    Feb 11, 2010
    Messages:
    192 (0.12/day)
    Thanks Received:
    26
    Location:
    SouthEast Blue Mountain, Pa
    Have you tried running the Antec to power just the Motherboard (both the atx and 4 pin cpu plug) & Video, and the 300w to power everything else? (Fans, HDD, etc.)
    Can you run the dual psu setup without overclocking? (ALL default clocks) Does it pass all stability tests?
    You said this board originally had issues part of that could be in the power circuit on the board, or elsewhere that is showing up when you overclock.
    There could be a common component that is causing issues or cannot handle the fluctuations in power.
    You may* also be seeing such drops because both psu's are trying to compensate, and interfering with each other.
    If you are using the x1950 in your specs, I know from experiences when I ran mine with dual psu's (mine has two power connectors on the back of it) I had to use the same power supply to power both plugs. Mine is an AGP version, but either way if you have two power connectors on it, make sure only one psu is powering it, do not split it up.

    I strongly recommend that you add up all the power needs directly off of your hardware and from research. Add up exactly how much power you need and then compare that to what is available. If you do this, please post the results here. If you could tell us your exact hardware I'm sure we can all help find the values to assist you. Values from your HDD's are usually very easy to get right off the drive (make sure to provide 12 and 5v mA requirements). Then we could deduce if your Antec should be able to power this system on its own.

    What other components do you have available to swap in and out for testing? (Audio, Video, etc.)
    Last edited: Feb 17, 2010
  11. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    I do have them both set to 110v when I'm actually using the computer, I just posted that bit of info because I found it interesting :)

    Going to answer the last Q first. The system in my specs is what I'm powering with the exception of it being a DFI LanParty UT NF4-D Expert; however, I was able to run the Gigabyte with just the Antec powering everything. I don't know if I've tried running just the Antec to power the DFI or not, but I can try. At the moment I'm only powering 1 HDD and 1 CDRom. 1 CPU fan (60x60x20mm), 1 thin CPU fan (60x60x10) and one 40x40x20 fan.


    I haven't run just the Antec PSU yet (everything is unplugged ATM) but with the Gigabyte and jus tthe Antec I was able to run it overclocked and stock speeds with it being stable.

    Since I only recently got the DFI board I'm only going off what I was told by the guy who gave me it, but he was getting BSOD when trying to install or boot into an OS. After asking a couple Qs he did mention it happening after he updated the BIOS. My original intent then was to flash an older one, because I got the same issues when I first got it. Then recently I tried the board again and something I did (dunno what lol) ended up fixing the board and it runs fine now :confused: :laugh:

    My x1950 only has 1 external power plug on it and is fine generally. My reason for going to dual PSU started when I was running the DFI board with just the Antec and it's 5v was dropping down too low. The odd part about the 5v on this is that if there is a decent draw on the 12v rail it drops the voltage on the 5v rail. So when I was powering the video card with the Antec, I noticed once the 5v was getting down to 4.7v (actual via multimeter) that when I would a video card stress test, the card would stop responding. Then Win7 would restart the driver and everything was fine, since that would stop the stress test and remove any extra load. The 6pin power adapter for it also is just 1 12v and 2 ground, I suspect mostly due to it being PCIe which can supply more voltage. It CAN handle 3 12v and 3 ground, but my adapter plug doesn't have those wired up.

    As for the power requirements of some things, I'm not sure exactly where I'd get that info for the motherboard. The memory I run at anywhere from 2.65 to 2.88v. Video card memory is like 1.8-2.2v (fluctuates) and Core is 1.425v. The card's Vreg reports idle draw of about 10A and load ranges from 13-30A, if that's accurate I can't say as the software I use to read that reports the vCore at 0.92v :\ I'll check on the HDD, but it's a Samsung Spinpoint 2.5" IDE 40GB, so from looking at the other HDDs I have which are older and 3.5" I don't see it drawing over 1A on the 12v and 600mV on the 5v. All the fans are 12v and under 700mV with the CPU fan running at 1/3 it's max speed. Everything else is onboard and the DFI has a daughtercard for sound which I didn't get so there isn't any power draw from that. The IRD, FireWire, additional GigaBit, all SATA and COM/LPT ports are all disabled in the BIOS since I'm not using them. So all the readings for voltages I've reported are with that in mind.

    As for spare parts, not much. Just that Gigabyte board in my specs, a Venice single core 2ghz CPU (I'd like not to remove the CPU because my heatsink suctions onto it and literally tears it out of the CPU socket:cry:). Only have old video cards to use as spares, but I have a PCI TNT2 Vanta. Also have 2 3Dfx cards (well 5, but the other 3 are commercial cards and not exactly plugnplay lol) which Win7 doesn't have drivers for so they can't even boot with vga video heh Only have 1 spare stick of memory which may or may not be bad as I've never been able to verify :\ I used it to initially get my Gigabyte system working, but I couldn't get the DFI to work with it, so who knows. And not that it matters, but I'm not sure where my DigiFire 7.1 sound card is :p Speaking of addition components, I'm not using any add-on cards except the x1950. While my main system generally has 5 HDDs and 1 is connected via a PCI RAID/JBoD card, it's not installed nor are any of those 5 HDDs.

    :respect:
  12. xrealm20

    xrealm20 New Member

    Joined:
    Jan 29, 2010
    Messages:
    467 (0.29/day)
    Thanks Received:
    129
    Location:
    Houston
    try using just the antec and let me know what the voltages are --
  13. Clement

    Clement New Member

    Joined:
    Feb 11, 2010
    Messages:
    192 (0.12/day)
    Thanks Received:
    26
    Location:
    SouthEast Blue Mountain, Pa
    I would venture a guess that he had a new setting set to Auto (or not set correctly - or it was reset after the update) in the bios that needs to be set manually (Probably Memory timings/settings.) I can't confirm this though, just an educated guess.

    You could try resetting the BIOS completely and starting from scratch, don't know if you tried that already. I know you said it works now, but be sure you have everything set properly.

    First at default frequencies and voltages (all), then overclocked, please.
    Last edited: Feb 17, 2010
  14. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    It's possible that was the case. I've reset the CMOS via jumper (had to a few times) and via "Set Defaults" option. I've also gone through the BIOS completely and changed or looked at everything there is. So I've become pretty familiar with it and the options that I prefer to be changed if I do have to reset. I still haven't done a BIOS flash though and it's running which ever one he flash. I do have one that I edited with MODBIN to unhide some settings (did that with my Gigabyte to bring forward a few more items) but I haven't bothered yet to flash it.

    I'll see what I can do about getting those readings for you. I do know that at default it's still under 5v and overclocked it drops it down about 100mV. Another thing is that the voltage is where should be when the PSU is cold, and as it starts to warm up, the voltages settle to their 'usual' readings around 4.8v. Been reading some reviews on HardwareSecrets on PSUs and I'm coming to wonder that might be due to a thermistor which regulates the power as heat output increases OR the (I think this is what I had looked at last night) the main Rectifying bridge, which are generally not cooled. Which when they heat up more their output efficiency takes a significant dive. So I was thinking about trying to cool either of those components if so. Also this unit has "silent" fans and I'm tempted to redirect their power from on the PSU PCB, to a full 12v source, to see if that helps anything. Seeing as the whole system has been well worth the money I put into it, if I do happen to toast anything, I won't feel bad lol While I'd rather NOT, to sell them to help with upgrades, I did get the DFI free :) Speaking of which... if anyone is interested in either my Gigabyte or DFI, I'll take offers ;) Contrary to what may come across in this thread, the DFI is stable for me. All lock ups are caused by me, tweaking the memory timings/speed or trying to find the max CPU speed :p Planned on keeping either of them but since I'd be more comfortable with a new PSU when I finally upgrade, I'll need to offset that extra cost heh

    Anywho I'll post back with my findings. Oh and I ran the UL code on this PSU to see who exactly made this, and I was happy to find out it was at least a CWT rebrand. Was hoping for a Seasonic, but that was wishful thinking :laugh: The internals are quite quality in appearance, not the cheap ass "finger" heatsinks you see in cheap supplies. I'll take pics though.
  15. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    Will edit this with some pics. Found the (what I figure based on pictures from multiple different reviews on Hardware Secrets) rectifier or PWM, or whatever :\ YAY IGNORANCE! :lol: Modified one of my old heatsinks I was using on a 9700 Pro back in the day, basically you run of the mill Radio Shack VRM heatsink. The pics will show it and it's close to a transformer, so I had to trim a fin down and tape that fin because I'm not sure if the coil windings are 'hot' or not and didn't want it accidentally shorting out on it. So that is now cooled, but I don't see a thermistor, and if what I see IS it then it's too close in proximity for me to cool it or move it away slightly. Hopefully the increased CFM from giving the fans more juice will help out more. So far though the EMI filter looks decent, except it has a transformer instead of a ferrite coil, but that may be correct either way haha There are many more X Capacitor down by the main Caps and I do see a large ferrite coil as well, so that might be it but downstream more. Come to think of it, that transformer is probably for the switching of 115v or 220v. Going to run wiring for the fans, then close it back up and see how it does.

    EDIT: OK so I'm not sure if the 2 main caps have slightly leaked, or if it was just the original glue/epoxy/silicone to help keep them in place, but the one pic will show it. I prodded at it with a screw driver and it was crusty, and I think I saw some in another spot too not on a cap, but not sure. I also think I may have located the thermistors. Hardware Secrets mentioned they are typically flat-n-round and olive in color.

    ARG! Pics are too large, need to crop them more >_< Stand by...

    EDIT2: Alright there they be. The one pic shows which ones I think may be the thermistors. I'm thinking it's more the top (bigger) one of those 2.

    Attached Files:

    Last edited: Feb 17, 2010
  16. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    I did try before I placed that heatsink, but it seems the DFI board may have bit the dust due to user error :\ Since I had wasn't sure if the DFI board was one that needed 24pin (GB board didn't) I had made the extra 4 from wires in the 20pin which were doubled up (apparently Antec was ready to release 24pin variants) and so I have to push them in myself. Well I accidentally plugged the 5v into the Ground and vice versa, so when I hit the power button it only powered up for a second. Get's a little foggy after that as it was quite late, and I'm not sure if I hit the power button a few more times or not. Reason being is the DFI sometimes has power-on hiccups where it won't turn on after the first few presses of the button. There will be just a ms of power which will cause the fans to spin, so power does course through the board. Well, after it didn't turn on at all I looked through the wiring and found that goof and switched them, nothing after that. That was when I removed the 300w. Now NOTHING happens when I push the button. The board is getting power as the LED lights up, but that's it. I'm going to try jumping the PSU while it's plugged in after I eat. I've already tested the PSU and it works fine, still only provides the same amount of voltage while under the light-bulb load :\ If I can't get the DFI to power back on, I'll swap in the Gigabyte and get those readings for you. Cross your fingers for the DFI! :p
  17. Clement

    Clement New Member

    Joined:
    Feb 11, 2010
    Messages:
    192 (0.12/day)
    Thanks Received:
    26
    Location:
    SouthEast Blue Mountain, Pa
    Be sure to let the DFI sit without any power for a good while. Take out the cmos battery also, and let all the power drain out before you try again.

    Sadly, you may have fried a vital part of it. Lets hope not!
    Formula350 says thanks.
  18. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    Will do, thanks. I hope it isn't too heh
  19. Formula350

    Formula350 New Member

    Joined:
    Mar 13, 2007
    Messages:
    864 (0.32/day)
    Thanks Received:
    49
    Location:
    Eastern Tennessee
    Well I'm not going to cannibalize all the parts off the board yet, but seems like it is highly probable now the DFI is dead. Let the board sit w/o power or a battery for almost 2 full days. I haven't tried powering it with 2 PSUs, but that's a pain in my ass right now to set back up heh I'll tackle that a bit later on. I'd like to put the CPU out of the DFI in the Gigabyte, but I hate pulling the heatsink off knowing the CPU will come with it :( But if I don't then I know I won't have diagnosed the thing properly. So that's why I'm holding off on it's confirmed status of being dead. I'll leave that verdict for after I get a new PSU, a proper 24pin. I'd be exstatic though for it to actually be the CPU that is somehow causing this, but I'm pretty sure it can't be that. Can't rule it out of course, but seems unlikely for there to be an internal short in it, since I think if it was dead the board will still power on, even if it didn't make it to POST.

    We shall see... In the mean time though I guess I can still carry out my tests with the single-core in the gigabyte and see if the voltage readings are the same as the DFI. So I'll probably do that tomorrow, seeing as it's 11:17pm and I've not gotten crap for sleep recently heh

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page