Ah. My mistake. I thought the power switch was connected to the PWRON line and grounded the line directly when the switch was momentarily closed, and that the long-press-to-switch-off was performed by the PSU rather than ACPI/BIOS driving PWRON low.
I've learned something!
Unlikely. that shouldn't see it only sometimes fail to work.
Wouldnt produce that symptom.
Most likely imo.
IME the most common cause of age-related PC problems is failing/failed electrolytic capacitors in the power supply or on the motherboard. The latter can usually be spotted, as they develop a bulge in the end of the can. If they're really bad they can actually split open, or push the bottom out of the can.
and if these capacitors are involed in the supply of +5VSB (+5 V Stand By) then thats enough to stop the PSU from starting....
Its usually a 22uF across an IC's Vss/+ve and Vdd/GND pins.
Is that still a thing?
I can remember replacing capacitors in a TV about 10 years ago, but around that time I remember motherboards started advertising that they were using better quality components, i.e. the problem with poor quality caps was recognised and addressed. I haven't had anything fail for quite a while.
Yep.
Plenty of others have with power supplys and motherboards.
Power supply, cable to switch. Connectors are the main suspect here as long as wiggling the button does not actually make it work in which case it is the switch. Brian
Most modern PCs do not have mains anywhere near the motherboard. what they do is have a point in the already running pSU where a momentary switch is connected to turn on and operate the various modes of the shut down. Those switches tend to be membranes and or the sort you find on remote controls, and do fail, particularly on laptops. Brian
Well, I don't think shoes make the slightest difference. wrist bands and a resistor to chassis is all I use, or one hand on the metal. I tend not to work on a pc with the mains connected in case a freak fault is present. I have no wish to be fried.
Reseat all connectors while you are in there if they allow this, some are heat glued together and normally do not have issues. Brian
But might cause corrosion and does not make for a very comfy working position. One other thing. Only try one thing at a time, then plug in and test. That way you may get an inkling where the fault was as poor connections are notoriously fickle and you might find the fault returns and if you do several things, then you are back to square one. I had a machine that did not boot and weirdly it turned out to be the reset button which had shorted internally, not allowing the machine to boot. I merely unwired it. It was an old machine and I only use it as a way to get old data from its drive over my network anyway. It would have been easy enough to have rigged up another switch. Brian
Yes but normally when this is the cause, you get a twitch from the fans inside before it shuts down. I don't suppose he knows if the 5v on the usb ports is there when the pc is off. In mine it is, proving to me that at least part of these is always on. If this supply goes or alters when you try to turn on, then its probably going to be the psu or some weird dry joint somewhere. Brian
Oh and to the original poster, please let us know what it turns out to be, for others use. also what motherboard or make of puter if relevant. Brian
That means there is no +5VSB.
Check:
ATX mains supply connector snug in wall.
Mains connector fuse is good.
(ATX supply typically has a five amp fuse or so, inside the chassis. This never blows. It's a slowblo fuse, and it is likely a protection item related to the large capacitor inside the ATX PSU exploding. Again, this never happens.)
Switch off ATX supply at the back, wait 60 seconds, switch on again. Did the blue light come on ?
Remove all USB cabling, except keyboard and mouse. This is, as someone already described, an attempt to remove a +5VSB overload from some USB peripheral.
The supply could have a bad +5VSB block, but this isn't a typical failure.
Replacing the supply with a spare, replacing the mains cable with a spare, are some things you can try.
Computer front panel switches break. Some of them crush easily, the plunger stays depressed. You can remove the pair of wires from the 2x4 header area in the corner of the motherboard, and use a separate switch with twisted pair 1x2 connector on the end. Then test that the switch works.
In some cases, users have swapped the power and reset
1x2 connectors - if the fans start but the PC stays in reset, that tells you the reset switch is working (because it turned on the power), while the power switch is not working (it kept the reset asserted).You can also "fake" the usage of a switch, using a screwdriver to short the two pins on the 2x4 front panel header. But this requires some amount of skill (like, magnifying glass plus a bright light).
In any case, good observation. We're getting somewhere. Not necessarily a bad motherboard now. Problem could be elsewhere. Something a bit easier to fix at home (front power switch issue, back power switch issue, [weak] PSU issue, mains cord issue, mains plug/fuse issue, outlet issue, breaker in panel). Things will go much better, once the blue light is made to come back. Remove loads from +5VSB, unplug Apple iPad and Apple iPhone.
Paul
He's got no blue light. It can't "twitch" until the +5VSB comes back on.
His front power switch, has a blue ring light around the switch. This is powered by +5VSB and gives a +5VSB status indicator.
Since the ring won't light, he has no +5VSB. PSU is most suspect, but could be a lack of mains. Unplug Apple iPad and Apple iPhone, then retest. Task now, is to get that ring LED to light up.
Paul
I have a mechanical problem with my ON button in that it sometimes jams in at an angle and that causes all manner of wierd issues. Sometimes the BIOS reloads itself from its copy and strange errors appear on the boot screen.
Powering off at the mains and prying out the cosmetic button to allow the actual electrical hardware button to reset fixes it.
Or just flick the wall switch to OFF :-)
I always choose SHUT-DOWN from the start button and wait until it has powered itself off then switch off at the wall.
It started working reasonably okay again. I am now satisfied it cannot be the switch (which works at all other times). I am testing an alternative theory at the moment. I just wonder if it could be software. The computer was set to shut down after a certain period (after I previously returned to default settings) and I wonder if this could be causing the problem. I have changes shut-down to 'Never'. It starts up after holding the button for long enough.
I have disabled the optical drive to simplify the start-up.
Incidentally, is it possible to get into the BIOS with a Bluetooth keyboard?
There are dongle receivers for keyboards, which present themselves as USB HID (Human Interface Device) and the BIOS can see and use those. Such a thing does not rely on a Bluetooth protocol stack being in the BIOS.
A device which is merely Bluetooth, implies the laptop has a Bluetooth receiver already inside the laptop, and it has some kind of profile suited to a keyboard. I don't know exactly what that would be. Things like phones seem to accept Bluetooth (foldup) keyboards. So a smartphone, is the closest thing to a candidate for a "pure" Bluetooth keyboard. Tablet PCs also seem to work with BT keyboards.
But on a platform which isn't particularly adept at Bluetooth, buying a keyboard with the receiver provided in the package as a nano-USB receiver, that should work at BIOS level. The receiver on this one, could go into a USB2 slot.
UEFI is capable of a great many things, but it cannot particularly have drivers for everything under the sun. It the equipment has Wifi+BT module as part of the hardware build, then it's possible the UEFI will have some support for it.
In the old days, an example would be owning a mobo with a USB 1.1 chipset, then plugging in a PCI USB2 card to augment the I/O. And discovering that the BIOS completely ignored the USB2 card. And only the OS and OS drivers could operate it. That's because the BIOS was not expecting to find a USB2 card. Other things the older BIOS did not like, included SATA storage cards (some BIOS did not know what SATA was), and there are BIOS that cannot read a DVD drive but can read a CD drive. The issue there, is the "hard disk emulation" has a fixed size, the CD is treated as an "800MB hard drive", but there was no "4.7GB hard drive" emulation to make a DVD work for booting.
An example of modern bozo behavior, is not knowing what an NVMe drive is.
Devices which have a "config eeprom", those can have a loadable driver. Many things are possible with those, but many engineering organizations are not interested in "doing the right thing". For example, in theory, a carrier card with an NVMe on it, could have had a config eeprom and supported booting. So far, the only card to do so, is a Highpoint NVMe RAID card (RAID 0, RAID1, presumably JBOD), and you could potentially stick that in a 2015 PC and boot from it. It would likely only work with UEFI, as it's too much to expect it to work with just anything. Highpoint deserves a medal for bravery, for doing the work, and accepting endless tech support phone calls :-)
Paul
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.