Changing voltages.

[Spork Schivago]

Re: Changing voltages.

Will do. Gonna take the kitten out to play for a bit and then when we get back in, I'll open the PSU up and see what's going on.

Another way we can test if it's the PSU I guess would be for me to use another PSU. If the BIOS hardware monitor still shows the value jumping around, we could probably assume the PSU is good, right? I'll let you know what the inside of the PSU looks like when I get back.

[ngvuanh]

Re: Changing voltages.

A computer running slow doesn't mean it has voltages problem. There are a lot more factors causing a computer slowing down.
Electronically, all semi components have marginal working voltages, so if all supply voltages stay within range, it should be fine.
Let get another hard drive, do a fresh Windows install and compare to current one to see if it make any difference.

[Spork Schivago]

Re: Changing voltages.

I took the PSU apart. The caps physically look good. It's gonna be a pain to get to the board so I can remove them to test them. At first, I thought it had some Rubicon caps. But at closer examination, they're Micon. It's weird because the font on them is very similar to a Rubicon.

The PSU is made by a company I've never heard of, Okia. It's 450 watt.

[Spork Schivago]

Re: Changing voltages.

A computer running slow doesn't mean it has voltages problem. There are a lot more factors causing a computer slowing down.
Electronically, all semi components have marginal working voltages, so if all supply voltages stay within range, it should be fine.
Let get another hard drive, do a fresh Windows install and compare to current one to see if it make any difference.

I understand that a computer running slow doesn't mean voltage problems. And why would I replace the hard drive if it isn't bad? There's plenty of programs out there to read the SMART data and everything. I had just noticed the voltage jumping around in the BIOS (along with the fan speed) and had questioned if that was normal, before I went through and replaced the thermal paste, backed up the data, formatted and did a clean install.

Thanks.

[stj]

Re: Changing voltages.

take some psu foto's

[ChaosLegionnaire]

Re: Changing voltages.

uh oh, junk/unknown brand caps in a junk/unknown brand psu? will have to assume the psu is faulty by default due to that reason. replace with a known good working japcap psu or post photos of the psu like stj said to see if its worth recapping the psu. caps can go bad without bloating especially junk/unknown brand caps! so dont assume the caps are fine unless that can be proven with an esr meter.

[Spork Schivago]

Re: Changing voltages.

I hooked up my 20-pin PSU. Definitely a difference. The 20-pin is an el-cheapo as well, but Vcore bounces around 1.208v, 1.216v, 1.224v. After a good 30 seconds or so, 3.30V changes. 3.392v, 3.408v. Very seldom, 12V changes. 12.144, 12.196V, 12.249V. Most of the voltages seem stable now. 3.30, 5 and 12 aren't going nuts like before. Fan speed is still jumping. With the 24-pin one though, they changed much, much faster.

It's like this now, Vcore changes multiple times real quick like. After so many times, 3.30V changes. After 3.30V changes so many times, 5.00V changes and 12V changes. Before, I could watch them all change all at once. Now I have to wait a good while for something besides Vcore to change.

Do you guys think I should assume bad PSU at this point in time and just have him replace it with a better one?

[Spork Schivago]

Re: Changing voltages.

uh oh, junk/unknown brand caps in a junk/unknown brand psu? will have to assume the psu is faulty by default due to that reason. replace with a known good working japcap psu or post photos of the psu like stj said to see if its worth recapping the psu. caps can go bad without bloating especially junk/unknown brand caps! so dont assume the caps are fine unless that can be proven with an esr meter.

Yeah, I agree that they can go bad without bloating. I just don't think it's worth replacing them. The amount of money it'd probably cost, the time involved in replacing them vs just buying a reputable PSU. You were posting around the time I was so I think you answered my question.

Any brands you recommend? I bought one of those EVGA PSU's not too long ago for a customer because it was cheap. This customer doesn't have a lot of money...

[Spork Schivago]

Re: Changing voltages.

A lot of *con caps I guess! The big ones are Canicon! The little ones, Micon. Both seem to use a font similar to Rubicon. I guess people like having the word con in their capacitors name!

[Spork Schivago]

Re: Changing voltages.

One of the ceramics look a little toasty. It's between the heatsink / a 1/4 resistor (or some other big type resistor) and some of those little transformers. It's dark in colour.

[stj]

Re: Changing voltages.

that's a POS, i doubt it could out more than 250w safely!

if vcore is jumping, you need to turn to the caps on the mobo around the cpu next.

[Spork Schivago]

Re: Changing voltages.

that's a POS, i doubt it could out more than 250w safely!

if vcore is jumping, you need to turn to the caps on the mobo around the cpu next.

Okay. And for that, I need to wait until my new iron comes in the mail. My 40 watt just isn't high enough to heat up the solder. I can always get one side, but the other side, it's always a no go. The board gets real hot so I'm pretty sure the plane is acting like a heat sink and a higher wattage iron will help.

Do you have any good suggestions on a PSU Stj? Or anyone for that matter?

[stj]

Re: Changing voltages.

something made by SeaSonic.

lets have some motherboard foto's

[Spork Schivago]

Re: Changing voltages.

So I'll probably suggest this one for him:

http://www.newegg.com/Product/Produc...-074-_-Product

It's a SeaSonic S12II 430B 430 watt ATX12V V2.3/EPS12V 80 PLUS BRONZE Certified Active PFC Power Supply.

I wish some of the modular systems were a bit cheaper. He doesn't need anything fancy. He uses onboard video, no gaming. Just e-mail, weather, maybe on-line banking. He's retired now but after that whole investors screwing a lot of American's recession thing, his 401K took a huge dive and went down to almost nothing.

[stj]

Re: Changing voltages.

http://www.hardwaresecrets.com/seaso...supply-review/

review looks good.

[Spork Schivago]

Re: Changing voltages.

Good to go. The board uses the nicer ferrite chokes, not the iron ones. So that's a good sign. I haven't counted how many phases there are yet but I figure I'll concentrate (once I get my new iron) on the voltage regulator circuit. I'll check the caps first. They all look physically good, that was the first thing I did, a physical inspection. So I'll pull them, test the ESR, and if I can't find a leaky one, I'll post some pics and come back here...

Unless you just wanted to see the pics for the heck of it Stj. I appreciate the help with diagnosing the PSU. I'm almost always used to having either a completely dead PSU or one where one of the voltages is completely dead. I'm not used to finding ones that don't supply steady voltage. I'll be sure to keep a lookout for this in future and I'll be sure to test the voltages with a DMM next time I work on a PC.

Thanks!

[stj]

Re: Changing voltages.

i want to see the type and placement of the caps on the board

[Spork Schivago]

Re: Changing voltages.

i want to see the type and placement of the caps on the board

Okay, I will post pictures later today.

I want to make sure I understand all of this now. The hardware monitor shows the vCore voltage, the 3.3v, 5v and 12v. Because the power supply supplies the 3.3v, 5v and 12v, the hardware monitor is reporting what the power supply is supplying, right?

The vCore voltage changes on what CPU is being used so motherboards use a voltage regulator circuit to change the voltage for the CPU, the RAM, etc. Right?

The voltage regulator circuit gets it's power from the PSU though. Is there a chance that once I get a _good_ power supply in there, the vCore will straighten out? If the voltage regulator circuit isn't getting stable voltages, I would think that would affect the various voltages it produces (ie, vCore).

[Spork Schivago]

Re: Changing voltages.

These aren't the best pictures because the board is still in the case. When I get my iron, I will pull the board and post better pictures for you, unless you want me to take the board out now and take better pictures. These pics will at least show you how the caps are laid out. They're mostly (if not all of them) Evercon and look like the last pic posted (that's not from my board, I googled them).

[stj]

Re: Changing voltages.

vcore comes from a switching psu on the mobo, it originates from the 12v from the psu.
it should be stable, an unstable vcore = an unstable cpu.

it looks the the 3phase switcher only has 3 caps - cheap bastards!
can you make out the markings?