PC WON'T POST, please help

[Aerpelding]

All,

I wasn't sure where to post this, but i'm hoping someone can help me out.
I walked into my office today, pressed the power button on my PC and it won't post. The last time i used it was yesterday evening. When i shut it down it said "update then shut down" and i clicked it. Then left the house and went to dinner and thought nothing of it.
When i pressed the power button on the case, it turned on, then turned off, then turned back on, then turned off, then turned back on and stayed on however it won't post. No beep, no start up screen, nothing.
Also, i should note that the CPU PWR light is illuminated orange the entire time. I also have no additional LEDs lighting up.
I tested the power supply with the CoolMax PS-228 LCD power supply tester and everything showed up what seemed to be normal. However, i did test it a few times and it would flash "HH" and sound an alarm on the PG voltage reading. But it wouldn't do this all the time, only a few times and the rest it would read somewhere near 400ms.

My hardware is as follows:
Asus ROG STRIX Z270E
Core i5 7600K
16GB Corsair Vengeance LPX DDR4 3000
Corsair RMx 750X PSU
Asus ROX STRIX 1070
The full build is in my signature..

I've tried to boot with the GPU disconnected and removed, i've tried to boot with no ram sticks, 1 ram stick, and both ram sticks. I've checked the cmos battery and it's reading 3.15v with a voltmeter. I've left the CMOS battery removed to clear the cmos. I haven't tried to clear the cmos yet though via a jumper yet as i can't find a jumper. I've unplugged all SATA cables, and still nothing. No post.

I have a feeling it's the PSU due to the way it was acting when i first turned it on however the PSU tester seemed to show no issue? Could the PSU have taken out my motherboard and/or cpu?
or could it simply just be a bad power supply?

Please let me know your thoughts. I'm hoping it's just the PSU and i'm leaning that way due to no error LED's on the motherboard as well as its behavior when i tried to turn it on. I just recently built this PC only two years ago so i'm definitely hoping it's not the motherboard or CPU, yet.

Thanks!
Aaron

 

[wingman99]

After all the steps you did to try and resolve the problem, the motherboard seems to be the problem since it won't POST and the PSU tested OK.

 

[Aerpelding]

what is the "PG" measurement on the PSU tester? You're not worried about the "HH" high voltage and alarm?

- - - Auto-Merged Double Post - - -

Also, i forgot to throw in that i did have the system overclocked. Could this be a sign the CPU is toast also? Or would it still POST and show a CPU failure?

 

[wingman99]

The way you describe the POST failure it seems to be a BIOS chip problem since it boot loops and then the CPU PWR light stays on. Maybe the high voltage damaged the PC it would be good to replace the PSU. It is rare to have a stock processor damaged.

 

[Aerpelding]

it only power cycled the first time i hit the power button. But i get what you're saying.
Luckily i have warranty on both items so i may just replace the PSU just to be safe.

Thanks for the help! With no LED lights to help guide me, other than the CPU PWR i was getting frustrated.

 

[trents]

The way you describe the POST failure it seems to be a BIOS chip problem since it boot loops and then the CPU PWR light stays on. Maybe the high voltage damaged the PC it would be good to replace the PSU. It is rare to have a stock processor damaged.

OP stats in post #3 that he did have the system overclocked. FastFocuse, how much core voltage did you use for the overclock. Tell us more about the the overclock. Still, I would bet it's either the PSU or the motherboard. I would start with the PSU.

 

[Bill_Bright]

A couple things.

I've checked the cmos battery and it's reading 3.15v with a voltmeter. I've left the CMOS battery removed to clear the cmos. I haven't tried to clear the cmos yet

You cannot conclusively test any battery with a multimeter unless the battery is still in circuit and/or under a realistic load. They can easily test good, but as soon as any load is put on them, the output voltage may drop below acceptable levels.

If me, I would just replace the battery since they are not expensive. Also when you remove the CMOS battery, you remove the "holding" voltage from the CMOS memory device. That automatically causes the device to lose any data it was saving. Shorting the two CMOS "Reset" pins on the motherboard does the exact same thing - it removes the voltage from that CMOS memory device by shorting the voltage to ground. That resets (or clears) the CMOS data. And it is also important to understand, with ATX compliant PCs (not talking about notebooks), there are no voltage storage devices (like capacitors) in that circuit (other than the CMOS battery). This means all voltage drops below usable levels almost instantly with just one or two clock cycles. No bleeding of the circuit is needed because there are no components in that circuit the need bleeding - that's by design! So if you see someone suggest you press and hold the power button for 2 minutes, understand all you are doing is making your finger tired. This also means leaving the battery out of circuit for minutes, hours or even overnight is just wasting your time too. 15 seconds is more than enough. And of course removing the battery and shorting the pins is just redundant. One or the other is fine.

Understand it is that easy because motherboard engineers and designers made it that way on purpose by intentionally choosing to use CMOS memory devices in that circuit. If they had wanted it to be harder to reset, they could have easily chosen a different technology memory device, like an EEPROM.

It must also be noted those PSU testers are NOT conclusive either. They are nice, I keep one in my bag for house calls as they can tell you if a PSU is dead, missing voltages or out of tolerance - but only at that tiny load level (typically just 10Ω) presented by the testers "dummy load". They cannot test using a variety of realistic loads, nor can they test for ripple and other anomalies that affect computer stability. This is done by a qualified technician using an oscilloscope or power supply analyzer - sophisticated (and expensive!) electronic test equipment requiring special training to operate, and a basic knowledge of electronics theory to understand the results. Therefore, conclusively testing a power supply is done in properly equipped electronics repair facilities. That means for most users, swapping in a known good PSU is the only way (but an effective way!) to verify a PSU is good (or bad).

Note according to your tester's manual (I have the same tester though it is "branded" by FrozenCPU), "HH" means one or more of your voltages was out of the allowed ±5% tolerance - in this case, it was too high. P.G. is the "power good" indicator. You seeing that means the timing was either too fast or too slow (faster than 100ms or slower than 900ms). You should not be seeing any of those error codes.

So IMO, you need to swap in another PSU and see if your problems continue.

And BTW, I hope you took the necessary ESD precautions when removing the RAM or the CMOS battery - that is, unplugging the PSU from the wall and touching bare metal of the case interior BEFORE reaching in or touching the RAM.

 

[trents]

Some excellent guidance there, Bill. I have often felt just testing voltage when not under load does not really tell all about the health of a PSU or a battery.

 

[wingman99]

I would unplug the PC press the power button then short the CMOS jumper with a screw driver for a minute.

 

[Bill_Bright]

I have often felt just testing voltage when not under load does not really tell all about the health of a PSU or a battery.

That's true. In fact, it applies to virtually any power source - even electric motors and car engines. They must be tested under realistic loads for the tests to be truly valid.

I would unplug the PC press the power button then short the CMOS jumper with a screw driver for a minute.

Sorry, but I guess you didn't notice what I said above.

Like the CMOS circuit, there are NO voltage storage devices in the power button's circuit. So once you unplug the PSU, the +5Vsb standby power is no longer being produced by the PSU and all output voltages quickly drop to 0V by the bleeder circuits in the PSU there for that purpose.

So nothing from the motherboard has to be bled off. And as noted above, shorting the two pins causes the CMOS device to almost instantly (within 1 or 2 clock cycles) dump all stored data - and that's by design! So shorting the pins for a minute is just wasting about 55 seconds of your time.

I note again, motherboard engineers and designers long ago specifically chose CMOS memory devices for this function because that is a unique characteristic of those type memory devices. So they use CMOS memory devices simply to make resetting (dumping) the CMOS data so quick and simple for normal users. CMOS memory devices (or circuits that emulate those devices) are still in use today because it is still the best memory device technology for that purpose. If they wanted it to be difficult or take a long time to totally reset that data, they would use a different technology memory device. And again, even that 1 minute for the battery is way overkill.

Again, this is for ATX compliant PCs - not notebooks.

You can easily verify this. Read your Z370-HD3 motherboard manual on page 19 and note it says (my bold underline added),

You may clear the CMOS values by removing the battery:
1. Turn off your computer and unplug the power cord.
2. Gently remove the battery from the battery holder and wait for one minute. (Or use a metal
object like a screwdriver to touch the positive and negative terminals of the battery holder,
making them short for 5 seconds.)
3. Replace the battery.
4. Plug in the power cord and restart your computer.

Either way, there's nothing in your manual (or any manual I've seen) about unplugging the computer and then pressing the power button - since that does nothing. I can only assume that myth came about from decades ago with the very old "AT" Form Factor standard used on the original IBM PC. The "AT" power supply had a large wiring harness that ran directly from the PSU to the front panel power button. Pressing (and holding) that power button was said to help bleed off residual voltages in the PSU that might still be present. But frankly even that was false - at least with AT PSUs in proper working order.

Regardless, this all changed anyway with the "ATX" Form Factor standard in the mid 90s. It got rid of the wiring harness and went with the "remote" power button used to this day that uses +5Vsb through those two motherboard pins to "signal" the system to power on or off. Note too that is a "momentary" circuit. Once the button is pressed, that short is "registered" in BIOS firmware and all input from then on is ignored until the button is pressed again. The except is when the BIOS "Soft-Off by PWR-BTTN" is set for "Delay 4 Sec". Then you must press and hold the button for 4 seconds to turn off the system. Less than 4 seconds will put the computer into sleep mode.

 

[trents]

This is not related to CMOS necessarily but in my experience with ATX computers after unplugging the PSU from the wall the motherboard (or he PSU) often still has enough power stored for a few seconds to make the fans spin momentarily if I press the power button.

Bill, so you're saying that the conventional wisdom of leaving the CMOS battery out overnight for a "deeper reset" is a wives tale, correct?

 

[wingman99]

That's true. In fact, it applies to virtually any power source - even electric motors and car engines. They must be tested under realistic loads for the tests to be truly valid.

Sorry, but I guess you didn't notice what I said above.

Like the CMOS circuit, there are NO voltage storage devices in the power button's circuit. So once you unplug the PSU, the +5Vsb standby power is no longer being produced by the PSU and all output voltages quickly drop to 0V by the bleeder circuits in the PSU there for that purpose.

So nothing from the motherboard has to be bled off. And as noted above, shorting the two pins causes the CMOS device to almost instantly (within 1 or 2 clock cycles) dump all stored data - and that's by design! So shorting the pins for a minute is just wasting about 55 seconds of your time.

I note again, motherboard engineers and designers long ago specifically chose CMOS memory devices for this function because that is a unique characteristic of those type memory devices. So they use CMOS memory devices simply to make resetting (dumping) the CMOS data so quick and simple for normal users. CMOS memory devices (or circuits that emulate those devices) are still in use today because it is still the best memory device technology for that purpose. If they wanted it to be difficult or take a long time to totally reset that data, they would use a different technology memory device. And again, even that 1 minute for the battery is way overkill.

Again, this is for ATX compliant PCs - not notebooks.

You can easily verify this. Read your Z370-HD3 motherboard manual on page 19 and note it says (my bold underline added),


Either way, there's nothing in your manual (or any manual I've seen) about unplugging the computer and then pressing the power button - since that does nothing. I can only assume that myth came about from decades ago with the very old "AT" Form Factor standard used on the original IBM PC. The "AT" power supply had a large wiring harness that ran directly from the PSU to the front panel power button. Pressing (and holding) that power button was said to help bleed off residual voltages in the PSU that might still be present. But frankly even that was false - at least with AT PSUs in proper working order.

Regardless, this all changed anyway with the "ATX" Form Factor standard in the mid 90s. It got rid of the wiring harness and went with the "remote" power button used to this day that uses +5Vsb through those two motherboard pins to "signal" the system to power on or off. Note too that is a "momentary" circuit. Once the button is pressed, that short is "registered" in BIOS firmware and all input from then on is ignored until the button is pressed again. The except is when the BIOS "Soft-Off by PWR-BTTN" is set for "Delay 4 Sec". Then you must press and hold the button for 4 seconds to turn off the system. Less than 4 seconds will put the computer into sleep mode.

I have had the PSU off and pressed the the power button and the power light on the case went on for a second.

 

[Bill_Bright]

Bill, so you're saying that the conventional wisdom of leaving the CMOS battery out overnight for a "deeper reset" is a wives tale, correct?

For sure.

for a second.

A second. Hardly a minute.

 

[Aerpelding]

Sorry for the delay guys, i have been out of town and haven't been online much. I've setup an RMA for both the PSU and motherboard. So, we'll see what happens from here.
Unfortunately it doesn't look like Asus makes the Z270E Gaming any longer so if it needs replaced it looks like they'll be replacing with a comparable unit.

OP stats in post #3 that he did have the system overclocked. FastFocuse, how much core voltage did you use for the overclock. Tell us more about the the overclock. Still, I would bet it's either the PSU or the motherboard. I would start with the PSU.

core voltage was 1.35 or 1.4v i believe. (It's been a while, and i honestly cannot remember exactly.) I know intel states no more than 1.5v, and i never went above 1.43v during initial overclocking as i know most people say never go to above 1.45v with the 7600k.

I would unplug the PC press the power button then short the CMOS jumper with a screw driver for a minute.

I tried this, and unfortunately still the same result. I troubleshooted every way i could think of and nothing changed.

Thanks everyone for all your advice and help. i'll keep you updated.

 

[Bill_Bright]

Unfortunately it doesn't look like Asus makes the Z270E Gaming any longer so if it needs replaced it looks like they'll be replacing with a comparable unit.

No longer in production does not mean they don't have any "refurbished" boards in stock just for this purpose. Note too these big board makers typically can just plug the board into a mockup analyzer that often is able to isolate the faulty circuit or even component and, if economically feasible, they may repair it. For this reason (mostly out of curiosity) I make sure I record the serial numbers before sending RMA parts in to see if they send me the same unit back or if I get a different one.

Don't forget to remove your RAM and CPU before returning.

 

[trents]

Yes, I agree with Bill. Record the serial number of the board before shipping it out. I have had the frustrating experience of RMAing a board and having it returned to me in the same condition it was sent.

 

[Bill_Bright]

I have had the frustrating experience of RMAing a board and having it returned to me in the same condition it was sent.

Usually with a comment of "no trouble found". In their defense, that is often the case. But not with me. I thoroughly test parts to make sure I have isolated the problem to that part. But I'm not normal!

 

[trents]

Usually with a comment of "no trouble found". In their defense, that is often the case. But not with me. I thoroughly test parts to make sure I have isolated the problem to that part. But I'm not normal!

Actually, most of the time there was no comment at all, even when the board was replaced, so I was left guessing.

 

[Bill_Bright]

most of the time there was no comment at all...so I was left guessing.

Yeah. Figures. Makes you wonder if they are just lazy, or just don't want us to know what they did.

 

[trents]

Yeah. Figures. Makes you wonder if they are just lazy, or just don't want us to know what they did.

Exactly! I think not many of us who participate in this forum would give any of the motherboard manufacturers high marks for customer service when it comes RMA time. In contrast to that are companies like EVGA and Corsair who I have found to be very customer-centered when customers have problems with their products.