Changing voltages.

[ChaosLegionnaire]

Re: Changing voltages.

So can I always fall back on dividing the number of switching mosfets by two?

nope. some vrms are the "double forward switching" type or the "double output switching" type in that they have 1 mosfet per phase doing the input and 2 mosfets per phase doing the output. which means there are 3 mosfets per phase. counting the number of inductors is the way to go. just try to spot the smaller inductor by the atx12v connector and not count it! only count the same sized inductors!

[Spork Schivago]

Re: Changing voltages.

nope. some vrms are the "double forward switching" type or the "double output switching" type in that they have 1 mosfet per phase doing the input and 2 mosfets per phase doing the output. which means there are 3 mosfets per phase. counting the number of inductors is the way to go. just try to spot the smaller inductor by the atx12v connector and not count it! only count the same sized inductors!

Gotcha, thank you!

[Spork Schivago]

Re: Changing voltages.

The FSB was replaced by QPI - QuickPath Interconnect on the 1366 socket.
The memory controller was moved to the CPU.

So the CPU is directly connected to the RAM then, right? Isn't this what makes the 1366 and the 2011 sockets extremely fast? No bottleneck there anymore?

I'm still waiting on the new power supply to arrive. When it does show up, I'll let you guys know if it fixes all the jumping voltages or if vCore is still bouncing around. On the 6th, my new iron is supposed to be here. Once that comes, I can start pulling caps if I need to. I'm thinking the PSU will arrive before the iron.

[Spork Schivago]

Re: Changing voltages.

I received the new power supply today. Vcore and 5.00v are jumping a lot. 3.30 jumps every so often. 12.00v jumps very seldom but does jump. Fan speed is still changing just about as fast as vCore. I'm going to start looking at the caps now because my new 120 watt iron came as well.

[stj]

Re: Changing voltages.

try the new psu on another computer - and use the scope.
then you have a reference to compare against.

[keeney123]

Re: Changing voltages.

I received the new power supply today. Vcore and 5.00v are jumping a lot. 3.30 jumps every so often. 12.00v jumps very seldom but does jump. Fan speed is still changing just about as fast as vCore. I'm going to start looking at the caps now because my new 120 watt iron came as well.

Have you tried disconnection things like the hard drive,ram and see if it still does it.Try disconnection they separately.

[Spork Schivago]

Re: Changing voltages.

Have you tried disconnection things like the hard drive,ram and see if it still does it.Try disconnection they separately.

Kinda. I've disconnected the hard drive but not the RAM. I don't really see how the PC can POST without the RAM. Unless, of course, you mean just leaving one stick in and trying it. I don't think it'll be necessary though.

I just got back from my friends death gathering stuff. I pulled the heat sink off and found the little bigger! One of those polymer caps right next to the CPU. I believe this isn't the best design. The heat from the heatsink will probably fry the caps. There's five caps total, idental. But four of them are in a row, the fifth one is up off by itself. I'm going to replace all of them.

[Spork Schivago]

Re: Changing voltages.

I believe these are 4v but are the 68uF or 680uF? I'm going to pull a working one and check the capacitance. I do have some nicer 6.3v 680uF polymer caps floating around here somewheres. If the ones on the board are 680uF, I'll use the nicer ones I have. They're Nichicon RL80J681MDN1KX. Even though they have a bit higher voltage rating, I think that'd be fine. So long as the ones on the board are 680uF and not 68uF. These Nichicon ones say 681. If those cheaper caps follow the same naming convention, I'd think the ones on the board are 68uF.

[stj]

Re: Changing voltages.

hmm - cant ID them
http://capacitor.web.fc2.com/solidcapacitor.html

whatever they are, i'v never seen one like that before - tempted to ask to see one opened up - incase they arent poly's

[Spork Schivago]

Re: Changing voltages.

hmm - cant ID them
http://capacitor.web.fc2.com/solidcapacitor.html

whatever they are, i'v never seen one like that before - tempted to ask to see one opened up - incase they arent poly's

Yeah, I couldn't find them either but I was looking at 680uF. The size is very hard to find at 4v 680uF. The only ones I could find that really matched the size (either roughly or exactly) where the Illinois Capacitor 687ULR4R0MFF.

[stj]

Re: Changing voltages.

well you didnt say the size

[Spork Schivago]

Re: Changing voltages.

DIGITALTECH kinda fits, wrong colour though. The caps say:
89
680
4J.

I can try opening the bad one up tomorrow if you want. Just tell me the best way to do it. I'll pull a good one tomorrow and test it with my Peak Atlas ESR meter to figure out the capacitance for sure.

I figure they're probably real cheap.

[Spork Schivago]

Re: Changing voltages.

well you didnt say the size

Whoops. Lead spacing (between the two terminals) is 3.5mm. The rest of the numbers might not be exact. I only have the one removed and it's bottom is blasted out a bit. Height (of just the metal can) seems to be 9mm, width seems to be 8mm. I think it's very hard to get 680uF, 4v in something that small. Makes me think that 680 is actually maybe 68uF. I thought for the VRM circuit though, these caps would be closer to 680uF. When I pull a good one tomorrow, I'll post the confirmed capacitance.

[Spork Schivago]

Re: Changing voltages.

Okay, I pulled one real quick. I just couldn't wait. 0.02Ω ESR, 1380μF!!! That can't be right! 1380μF in this little package?? How can that be? If they're failing, could they actually register higher than what they're marked as? I can pull a couple others and test them if you'd like.

[stj]

Re: Changing voltages.

lol - if your meter could show it i bet leakage is huge!

[ChaosLegionnaire]

Re: Changing voltages.

lol no way they are 68uF. that is too low capacitance for cpu vrm output use. they usually use 560-820uF polys for the output.

also, i dont understand the part about the size being hard to find. there is an oscon sepc 560uF 4v 8x9mm part as well as lots of nichicon fpcaps available in the same size as well? were u thinking those were wet electrolytics caps on the vrm output? well, u need to watch those kzgs on the vrm input as well. those need to go as well!

[Spork Schivago]

Re: Changing voltages.

lol no way they are 68uF. that is too low capacitance for cpu vrm output use. they usually use 560-820uF polys for the output.

Thank you for explaining this. I wondered if the VRM circuit used caps around that range. I was looking around the net (over the years) and they always seemed to fall somewheres around there. I wasn't sure though for certain. I know the boards I have around here are around there for the VRM stuff.

also, i dont understand the part about the size being hard to find. there is an oscon sepc 560uF 4v 8x9mm part as well as lots of nichicon fpcaps available in the same size as well? were u thinking those were wet electrolytics caps on the vrm output? well, u need to watch those kzgs on the vrm input as well. those need to go as well!

What I meant about the size being hard to find, I cannot find very many 680uF 4v capacitors that are around the same physical size. The Illinois ones seem to fit. I could find the 560uF just fine, however, I'd like to stick with the 680uF if that's what these are. I've read many posts how people were switching out 680uF for 560uF OS-CON SEPC's with success.

I do have some 6.3v 680uF Nichicon's that I'll probably try using first. If I didn't have them, I'd probably give the 560uF OS-CON's a shot. I heard they're supposed to be really good caps.

So, what do you think about my ESR meter showing such a high capacitance? I'll start pulling the other ones and checking their capacitance / ESR.

[Spork Schivago]

Re: Changing voltages.

I heard there's some Sanyo caps out there that people have been using as well and they're also supposed to be real good polymers...

[Spork Schivago]

Re: Changing voltages.

lol - if your meter could show it i bet leakage is huge!

My meter can detect leakage but not show it. It'll say Leaking / In-Circuit. I now always remove the caps to test them so I can ignore the In-Circuit part. For some reason though, with the cap I checked here, it didn't say anything about leaking...

[stj]

Re: Changing voltages.

i have seen caps where the uF value shoots up, it is acompanied by very high leakage.