Possible Emergency? CPU at 1.475V rather than 1.35V after un-overclocking

[johnny333]

Mine will do that if I us the on-board over-clocker. I can change everything back manual or just use the jumper. What do you BIOS voltages say. I have used a voltmeter on a PC in over 10+ years they don't work to test a power supply because they do put a load on it ether power supply test is the only way to go.
Marry Christmas Johnny333 65+++

 

[Bobnova]

The issue is accuracy. A motherboard voltmeter and how those numbers are reported by programs are both completely unknown. A motherboard voltmeter has a history of inaccuracy. Obviously at least one of two programs are also inaccurate. Which one? Multiple unknowns exist. All questions get answered simply by one minute with a multimeter. Which assumes the OP still wants an answer.

Measuring 3v3/5v/5vsb/12v/-12v isn't a bad idea, but it tells you absolutely nothing about the vcore voltage.
Vcore is generated off the 12v rail with active feedback, you can change the 12v and watch the vcore stay exactly where it started. I have, it did.
If the 12v rail is really low than many things will not work, but the vcore will stay the same up until the controller for the vcore regulation setup decides that VIN isn't high enough and shuts down.

 

[Tech Tweaker]

Mine will do that if I us the on-board over-clocker. I can change everything back manual or just use the jumper. What do you BIOS voltages say. I have used a voltmeter on a PC in over 10+ years they don't work to test a power supply because they do put a load on it ether power supply test is the only way to go.

Well, you can fairly easily plug in some multimeter test probes into the power supply wires (12v, 5v, or 3.3V) while the system is running, then put a load on the power supply (prime 95, IBT, etc) and see what the voltages are.

 

[johnny333]

Well, you can fairly easily plug in some multimeter test probes into the power supply wires (12v, 5v, or 3.3V) while the system is running, then put a load on the power supply (prime 95, IBT, etc) and see what the voltages are.

If system will run. But if you try shorting the I think orange to a black ground and using a hard drive or two and fans too. It can still read good but isn't. For $20.00 are so you can't beat the right tool for the job. Mine has paid for itself many times over and a lot easier too. It my 30+ years I have been bitten more than once by a power supply they can give all kind of weird and intermittent problems .
Merry Christmas Johnny3
33 65+++

 

[Bobnova]

Orange is 3.3v, if you short that to ground you're going to get sparks and a shutdown at the least, fire and a dead PSU at the most.

A $20 PSU tester won't tell you anything that a paperclip and a $7 multimeter won't, other than the time to turn the PWR_GOOD signal on.

 

[RGone]

Orange is 3.3v, if you short that to ground you're going to get sparks and a shutdown at the least, fire and a dead PSU at the most.

A $20 PSU tester won't tell you anything that a paperclip and a $7 multimeter won't, other than the time to turn the PWR_GOOD signal on.

Yep green to ground for PWR/ON signal.

 

[westom]

Orange is 3.3v, if you short that to ground you're going to get sparks and a shutdown at the least, fire and a dead PSU at the most.

Shorting 3.3 volts will not create a fire. Will not damage the PSU. Could be done even to any voltage on the original IBM PC without damage. Is defined as non-destructive in ATX Standards. Intel, in defining power supplies for the orginal Pentiums, even said how thick that shorting wire must be. Because shorting out any PSU voltage even is a standard test. A normal and non-destructively test for any computer's power supply - not just a PC supply.

Any supply damaged by a shorted output is probably defective when purchased.

Nobody said anything about a power supply tester. The two are obviously completely different. Defined was a multimeter that, in one minute, says more than a paperclip test. A multimeter provides numbers. Neither a paper clip nor tester do that. Numbers demonstrate a meter's significant benefits.

Even a defective supply might pass the paper clip test. Or be declared good by a power supply tester. What good are tests that declare good a defective PSU?

Apparently the OP is not concerned. However, others can learn why a meter is so quick, inexpensive, powerful, informative, and useful. A meter even makes possible additional assistance from those who better know this stuff. Who know shorting PSU outputs does not damage a properly constructed PSU. Who know a paper clip can only identify some PSU failures and never define a PSU as good. And who know the difference between a power supply tester and a multimeter.

 

[adams1029]

A wild stab in the dark here. But are ou using a gigabyte modo? I had this issue with both my 1090t and my 4100. F7 after a over clock will set it too 1.475 every time for me until I did it a second time and sometimes would even have to boot fail safe. Happen to me on
970-ud3
990-ud3 boards.

 

[Seebs]

Measuring 3v3/5v/5vsb/12v/-12v isn't a bad idea, but it tells you absolutely nothing about the vcore voltage.
Vcore is generated off the 12v rail with active feedback, you can change the 12v and watch the vcore stay exactly where it started. I have, it did.
If the 12v rail is really low than many things will not work, but the vcore will stay the same up until the controller for the vcore regulation setup decides that VIN isn't high enough and shuts down.

Amen brotha!!! Say it like you mean it!!! <--- Says Seebs in his best "Sothern Preacher" impersonation.

 

[Bobnova]

Shorting 3.3 volts will not create a fire. Will not damage the PSU. Could be done even to any voltage on the original IBM PC without damage. Is defined as non-destructive in ATX Standards. Intel, in defining power supplies for the orginal Pentiums, even said how thick that shorting wire must be. Because shorting out any PSU voltage even is a standard test. A normal and non-destructively test for any computer's power supply - not just a PC supply.

Assuming that the PSU has functional OCP/SCP, yes. That can be a large assumption, and not one I am willing to make. There is a reason it's a test. A test is to see if something fails the test. If every single PSU passes the test there is no reason to have the test.

Any supply damaged by a shorted output is probably defective when purchased.

Or poorly designed, or you didn't get a good solid short and instead had a sustained arc that melted something. Or your paperclip was too thin and turned into a heating element that burnt your fingers. Been there, done that.

Nobody said anything about a power supply tester. The two are obviously completely different. Defined was a multimeter that, in one minute, says more than a paperclip test. A multimeter provides numbers. Neither a paper clip nor tester do that. Numbers demonstrate a meter's significant benefits.

Johnny implied one. I stated it. A basic PSU tester jumpstarts the PSU and takes voltage readings. Just like a paperclip + multimeter.

Even a defective supply might pass the paper clip test. Or be declared good by a power supply tester. What good are tests that declare good a defective PSU?

Significantly more useful than testing the 12v rail when you're wondering if the CPU vcore is too high. Also useful as most (not all) dead PSUs will throw bad numbers on the paperclip test too.

Apparently the OP is not concerned. However, others can learn why a meter is so quick, inexpensive, powerful, informative, and useful. A meter even makes possible additional assistance from those who better know this stuff. Who know shorting PSU outputs does not damage a properly constructed PSU. Who know a paper clip can only identify some PSU failures and never define a PSU as good. And who know the difference between a power supply tester and a multimeter.

If you're attempting insults, you're failing. If you're attempting to be helpful... You're failing.
I think everybody involved in this thread knows all of the above, it's not exactly rocket surgery.

None of your post has anything to say about why you think that rail voltages that do not cause any instability are going to cause a vcore issue Could you please explain a bit on that front?

 

[westom]

None of your post has anything to say about why you think that rail voltages that do not cause any instability are going to cause a vcore issue

I never said that. The OP has numbers from two programs that contradict. Is program A correct? Is program B correct? Is the motherboard voltmeter providing wrong voltages? Are both programs in error? All three unknowns exist. All three get answered by measuring with a digital volt meter. Then all three unknowns are eliminated or solved.

As others noted, 12 volts could be defective (OPs 12 volts may be marginal). Still Vcore, that is powered from 12 volts, could be perfectly ideal. A defect on any one causes computer failures or future instabilites. All those voltages must be accurate.

All three OP's unknowns are quickly eliminated. Other potential problems are identified even before a crash happens. And all that only takes one minute with a meter.

So, which program was reporting good numbers? Can anyone say without numbers from a multimeter? Apparently not.

 

[TeknoBug]

The core turbo option would usually push the voltage up, I've seen it where I set 1.4 but with CIE, turbo and CnQ enabled it kicks up to 1.475v. Right now I have all disabled and set to 1.450v but CPU-Z says 1.472v, doesn't bother me as long as it doesn't go over 1.5v which it did when I manually set it to 1.475v previously.

 

[Bobnova]

Measuring CPU voltages is difficult since how each adjacent CPU power supply works can be unique for each motherboard manufacturer. One way is to measure all power supply pins until finding a voltage is closest to what software reports.

CPU voltage does not come from the power supply. You can measure every PSU pin and you won't find the CPU voltage there.

Generally, if other voltages are slightly high or low, then that CPU power supply voltage should also measure slightly high or low proportionally

Not true.

I was unaware that any CPU changes voltages on the fly. Generally, a CPU voltage was preset (by the CPU). And remained at that voltage.

Not for years.

 

[RGone]

OP used program that gave CPU requested voltage from mobo.

OP used a "different" program that gave CPU supplied voltages.

OP did not know that he was getting two correct answers for two different voltages.

OP did not want to do a full diagnostics on his stuff. Just be told which voltage was which from each voltage read-out software. Job Done.

 

[visiblegorilla]

Shorting 3.3 volts will not create a fire. Will not damage the PSU. Could be done even to any voltage on the original IBM PC without damage. Is defined as non-destructive in ATX Standards. Intel, in defining power supplies for the orginal Pentiums, even said how thick that shorting wire must be. Because shorting out any PSU voltage even is a standard test. A normal and non-destructively test for any computer's power supply - not just a PC supply.

Any supply damaged by a shorted output is probably defective when purchased.

Nobody said anything about a power supply tester. The two are obviously completely different. Defined was a multimeter that, in one minute, says more than a paperclip test. A multimeter provides numbers. Neither a paper clip nor tester do that. Numbers demonstrate a meter's significant benefits.

Even a defective supply might pass the paper clip test. Or be declared good by a power supply tester. What good are tests that declare good a defective PSU?

Apparently the OP is not concerned. However, others can learn why a meter is so quick, inexpensive, powerful, informative, and useful. A meter even makes possible additional assistance from those who better know this stuff. Who know shorting PSU outputs does not damage a properly constructed PSU. Who know a paper clip can only identify some PSU failures and never define a PSU as good. And who know the difference between a power supply tester and a multimeter.

This guy! I think I just like the way he writes. His timing is almost Shakespearean, iambic pentameter and all that. "But soft! What spark through yonder wiring arcs? 'Tis the 3V, and pap'rclip is the short!"

 

[RGone]

This guy! I think I just like the way he writes. His timing is almost Shakespearean, iambic pentameter and all that. "But soft! What spark through yonder wiring arcs? 'Tis the 3V, and pap'rclip is the short!"

I did not want to get this thread started again man, but U Killen Me. Iambic Pentameter!

The tantalizing sound to get us ready. The short or long paper clip to hand. The ground to a power supply output. The sparks to come. The release to joy.

Deeyam am losing it man.

 

[johnny333]

On my mother board turbo will up the voltage. I use CIE because my PC doesn't run wide open all the time an it will turn the voltage and multiplier down too. To me MB monitor is good enough to check the voltages. This old man would say you need to be shore you have a quality power supply with enough watts. It been 20+ years but if you really want to see what is going on with your power supply you need a oscilloscope(big bucks) get a 24 pin extension cable put some test point in the extension and go at it. It will show you how pure your voltage is and look for spikes. But to me that would be more trouble than it is worth .
Have fun and Merry Christmas Johnny 65+++

 

[ivanlabrie]

A friend's m5a88-v evo does the same thing, it overshoots voltage frequently and I couldn't pinpoint what the issue was...I think Turbo was disabled, same as CnQ but he had a basic LLC setting enable (no different llc levels, just on or off)

And, yeah, Westom does sound kinda poetic in his lunacy lol