USB 3.0 port stopped working - Fixed by unplugging everything

[writhziden]

People may recall that I had some issues with my motherboard for the first few months or so that I owned it : Gigabyte LGA 2011 DDR3 2133 Intel X79 SATA 6Gb/s USB 3.0 ATX Motherboard GA-X79-UP4 : https://www.sysnative.com/forums/hardware/6042-cpu-psu-fan-spin-second-stop-new-build.html

I have had intermittent issues since then, mostly with peripherals getting into odd power states. I had a TV tuner card that would cause all my HID devices to show warnings in Device Manager (I believe the warning was that they could not start?), and I would have to do a hard shutdown or reset to get my system back up and running. Updating my BIOS to the latest version seemed to provide a bit more stability. Clean installing Windows fixed that issue 100%, which suggests it was probably driver related with possibly some issue with BIOS stability.

Last night, I was using my system, and I kept hearing the 'Device Disconnected' notification sound. This morning, I turned on my system to find that one of my USB 3.0 ports was not working properly. It seemed fine through Device Manager, but devices were not able to interface with Windows. The devices received power, but Windows was not registering them being connected. There was not even a message that an unknown device was detected.

Additionally, my internal Blu-Ray drive stopped reading properly during the abovementioned USB 3.0 port issue. Windows became unresponsive because Explorer was trying to access the drive. I could not eject the drive or access it (Access Denied message when trying to eject or open). I think this may all be related.


I waited for my system to finish some tasks it was doing, and then I restarted to reset the BIOS to optimized settings. That didn't work, so I disconnected all peripherals from the back of the system and then unplugged the power. I left the PSU in the on state and pressed the power button to close the circuit and drain any residual power. I then plugged everything back in and reconnected the power. Now the USB 3.0 port is working, but I'm really starting to wonder if I should continue to trust this motherboard given all its odd power behavior since I first bought it. The system is four years old, so replacing parts is something I am considering.

Thoughts?

 

[Digerati]

then unplugged the power. I left the PSU in the on state and pressed the power button to close the circuit and drain any residual power.

There's a misconception here. With ATX power supplies, when you unplugged the power, the PSU was in a totally "off" state. And pressing the power button at this point does nothing. This is just a momentary "signal" circuit connected only the +5Vsb standby circuit - totally isolated from the main +12V, +5V and +3.3V circuits. There are no residual voltages to drain. If there are any throughout the computer and power supply, they drained within a couple seconds of you unplugging the power from the wall.

In the good old days, before ATX, there was the AT Form Factor standard. And with AT power supplies, there was a wiring harness that ran directly from the power supply in back to the case's front panel power switch. And holding that power button down did drain residual voltages in the PSU. But ATX supplies are totally different and the front switch is just a remote switch.

With ATX, when the power supply is simply plugged in (and if equipped, the master power switch on the back of the PSU is set to 1 or On), +5Vsb standby voltage is distributed to many points on the motherboard, including USB ports. This allows many things, including the ability to wake a sleeping computer by pressing a key or wiggling the mouse (or by pressing the front panel power button).

I have no idea why, but it is not uncommon for USB ports, or the devices connected to them, to lock up, stop functioning, or to disappear from the BIOS and OS. Simply rebooting does nothing because that +5Vsb voltage remains present keeping corrupt settings set (at least that's my theory!). So doing a "cold" reboot (that is unplugging from the wall) removes that +5Vsb from all points and ports, forcing the BIOS and to reset everything anew and boogie on!

So what cleared your issue was probably the fact you unplugged from the wall. Pressing and holding the power button just makes your finger tired!

I have to do this (a cold reboot) frequently after I do something like connect my phone to down load pictures, or download my BP readings from my BP monitor, connect my drive docking station, or even a thumb drive. Even if I "Safely remove the USB device", I will start hearing those Device connect and disconnect sounds. Rebooting does not help. A cold boot does every time. And I have seen this on dozens of computers, with XP, Vista, 7, 8 and now 10. It is very frustrating to me they (the USBIF) have not sorted this problem out. My guess is childish politics is preventing the members from compromising and coming to a consensus. :banghead: :censored2:

I don't like USB at all. It is a very unreliable interface and each new version is supposed to fix the problems with the last, but don't.

Clean installing Windows fixed that issue 100%, which suggests it was probably driver related with possibly some issue with BIOS stability.

I doubt it is anything to do with BIOS stability. That is rare and normally only seen on brand new boards, and it is quickly fixed with a BIOS update. More likely it is with the USB interface drivers for the motherboard, or the USB device drivers for the connected devices. And fresh install reset all.

Try swapping in different and known good quality power supply before you start replacing other parts just to make sure you are supplying good clean power. And if not running off a "good" UPS with AVR (automatic voltage regulator) I highly recommend do. A surge and spike protector is little more than fancy and expensive extension cord. It is the AVR feature of the UPS that is the mainstay and primary reason to use a "good" UPS. Battery backup power during a full power outage is just a side bonus.

 

[writhziden]

then unplugged the power. I left the PSU in the on state and pressed the power button to close the circuit and drain any residual power.

There's a misconception here. With ATX power supplies, when you unplugged the power, the PSU was in a totally "off" state. And pressing the power button at this point does nothing.

Yep, it just turns some systems on for a brief second. My current system did not turn on, and I did not hold down the power button as you suggested; I merely pressed it for a couple seconds to see if it would turn on like my previous desktop system did. I figured that meant it did something to clear power from capacitors. Is that brief surge of power that used to occur on my previous desktop PC really nothing?

Try swapping in different and known good quality power supply before you start replacing other parts just to make sure you are supplying good clean power. And if not running off a "good" UPS with AVR (automatic voltage regulator) I highly recommend do. A surge and spike protector is little more than fancy and expensive extension cord. It is the AVR feature of the UPS that is the mainstay and primary reason to use a "good" UPS. Battery backup power during a full power outage is just a side bonus.

I guess I will have to buy a new PSU. This one is a quality PSU, but it is also four years old, so I doubt it will last much past the five year warranty. Buying a new PSU now is not too crazy.

I will also get a good UPS for the system. Do you have any recommendations since you seem to be emphasizing the word "good" in some way?

 

[Digerati]

I figured that meant it did something to clear power from capacitors. Is that brief surge of power that used to occur on my previous desktop PC really nothing?

There are no storage capacitors in that circuit to drain. A brief surge of power is probably the PSU trying to turn on, the BIOS then sees something is wrong and immediately shuts down, or there are voltages missing preventing the rest of the computer to start up. So the PSU may start to put out +12V for example causing fans and drive motors to spin for a second, but if the +5V or +3.3V is missing or out of tolerance, the rest of the system cannot function.

Do you have any recommendations since you seem to be emphasizing the word "good" in some way?

Like power supplies, there are cheap UPS and there are "good" UPS - even within the same brand.

I like APC but Cyberpower and Tripp Lite also make "good" UPS. Unfortunately, like many things, the better features and better quality often only come in the bigger models. So while a 700VA UPS may be plenty to support your system, the better quality features don't normally show up until you get to 1000VA and up. But at least with the bigger UPS, you can support your monitor and all your network gear too.

 

[writhziden]

I figured that meant it did something to clear power from capacitors. Is that brief surge of power that used to occur on my previous desktop PC really nothing?

There are no storage capacitors in that circuit to drain. A brief surge of power is probably the PSU trying to turn on, the BIOS then sees something is wrong and immediately shuts down, or there are voltages missing preventing the rest of the computer to start up. So the PSU may start to put out +12V for example causing fans and drive motors to spin for a second, but if the +5V or +3.3V is missing or out of tolerance, the rest of the system cannot function.

If there are no capacitors being drained, where does the power come from when the power turns on and the system is unplugged? :huh: I am not an electrical engineer by any stretch, but I did take enough circuits courses to know something has to be storing power in the system for that to happen, and capacitors would be the likely culprit since that is their function.

I also never said which capacitors I was trying to drain. I was merely attempting to remove any power from the system that might keep components active. After that initial burst of power, pressing the power button ceased to do anything, so there would seem to be less chance of components still being active.

Like power supplies, there are cheap UPS and there are "good" UPS - even within the same brand.

I like APC but Cyberpower and Tripp Lite also make "good" UPS. Unfortunately, like many things, the better features and better quality often only come in the bigger models. So while a 700VA UPS may be plenty to support your system, the better quality features don't normally show up until you get to 1000VA and up. But at least with the bigger UPS, you can support your monitor and all your network gear too.

Have you ever had a surge get through the UPS unit and damage equipment? Reading reviews on the APC brand UPS units made me a little uncomfortable buying from them due to some of their policies with damaged equipment. Let me know if you have had a better experience with APC; I take reviews with a grain of salt since user error is pretty common with electronics. Tripp Lite seems to have more favorable reviews with equipment replacement/insurance but poorer reviews with user friendliness of their UPS units.

 

[Digerati]

If there are no capacitors being drained, where does the power come from when the power turns on and the system is unplugged

You lost me there. Are you saying if you unplug the computer from the wall and press the power button, the computer starts to boot? If that is happening, there is a fault somewhere in the power supply.

You are absolutely correct that it would be capacitors storing that power, but the only power running through that front panel power switch is the +5Vsb voltage. That +5Vsb is removed (almost instantly) when you unplug the power supply from the wall (or flip the master power switch on the back of your PSU - if it has a master power switch). And that +5Vsb has a small current capability that is isolated (or is supposed to be as per the ATX Form Factor Standard) from the 4 main high current rails; +12VDC, +5VDC, +3.3VDC and -12VDC (though for most computers, the -12VDC is not used). When the power button is pressed to signal the PSU to turn off, those 4 main rails are supposed to turn off and have zero potential with respect to ground - meaning there should be no voltages being delivered to any other components as there are bleeder resistors in the PSU just for that purpose - assuming the PSU is working properly. And as mentioned earlier, that circuit the front panel button is connected to is a "momentary" circuit. Once the short is detected when you first push the button, the circuit ignores further input until the next short arrives (you press the button again).

We are talking about a desktop PC, right? And not a notebook?

Have you ever had a surge get through the UPS unit and damage equipment?

The only time I have heard of a surge getting through a UPS is when the transformer on the pole outside a client's house took a direct hit from lightning. That strike, besides starting the pole on fire, took out electronics in all 3 houses on that transformer. But short of a direct lightning strike from an angry Mother Nature, I have never seen or heard of an UPS failing to protect the connected equipment. Even if the batteries are failing and no longer capable of supporting the connected equipment during a power outage, they can still absorb massive surges - much better than a surge and spike protector.

I have seen UPSs of several brands fail - but in the process, they simply cut power to connected devices.

I have used APC UPSs for over 25 years and have never had to make a claim - and I live in Tornado Alley. In fact, I still have my very first APC UPS, a Smart UPS 900 that I got way back in 1994 and it still works great. I have it on my garage door opener! Other than having to replace the batteries a few times over the years (normal maintenance), it keeps humming along.

I currently have on this computer an APC BackUPS XS 1500. It is supporting this computer with a 550W PSU plus my cable modem, wireless router, 4 port switch, home phone plus two 24" LCD monitors. I've had this UPS for about 4 years and never a problem. If I immediately power down completely (using the master power switch) my computer and monitors when the power goes out, this UPS will keep my network alive for at least 6 hours so I still have access via my wireless devices.

I actually have a "Geek Squad" brand UPS on my home entertainment equipment supporting my 55" OLED TV, DVR, Blu-Ray player and AV receiver. It is made by Tripp Lite and it has worked flawlessly since I bought it too.

You may see "marketing hype" about pure sine wave outputs. That's all it is, hype. Any 1/2 way decent AC/DC power supply can handle the "stepped sine wave" or "stepped approximation" waveform just fine. They have for the last 25 years with no problems so there is no reason to believe the much more reliable and robust power supplies of today can't either. They are much more capable at handling power line anomalies than PSUs of yesteryear.

The ATX Form Factor standard requires all ATX power supplies to "hold" voltages for a mere 17ms (milliseconds) during abnormal power events. We (as humans) are not able to detect a flicker in the lights of just 17ms. But a "good" UPS can react easily within that time frame. Most have a cut-over or transfer time of 10ms or less.

And that's not just computer PSUs, but the power supplies for our modems, routers, switches, monitors and more. The only time you really need a "pure" sine wave output UPS is when used to support highly "sensitive" health monitoring and life support equipment as found in hospital intensive care units.

All computer and networking devices use DC voltages. It is actually easier to make a clean DC voltage converted from a stepped approximation sine wave than it is from a pure (round tops) sine wave because less filtering is required.

Do not forget that 99% of the time, our UPS is not "on-line". That is, it is not acting as a battery backup, except during a total power outage. It is just "in-line" acting as a sophisticated "automatic voltage regulator" (AVR). It is really for the AVR that we need a "good" UPS. Backup power during a full power outage is just the icing on the cake. But even then, power supplies do just fine with stepped approximation outputs.

The only real complaint I have with APC is they tend to use their reputation to charge more for their UPS. But they get away with it because they do make reliable products. I know many techs who won't by any brand but APC.

There is one downside to using a UPS. Most use SLA (sealed lead-acid) batteries and they have to be replaced about every 3 - 5 years. The UPS makers would like you to buy their expensive branded replacements, but there are sites that sell suitable replacements that are just as good, and often the exact same as the original, just a different brand label stuck to the side, but cheaper. I buy my UPS batteries from Apex, BatteryWholesale, Battery Mart or Batteryplex, depending on who has the best price that day, as they change almost daily. Be sure to factor in shipping. Depending on the deal of the day, shipping can be significant, or free!

 

[writhziden]

If there are no capacitors being drained, where does the power come from when the power turns on and the system is unplugged

You lost me there. Are you saying if you unplug the computer from the wall and press the power button, the computer starts to boot? If that is happening, there is a fault somewhere in the power supply.

I do not believe it is a power supply issue. I have seen this on HP and Gateway systems. My HP system that I linked to in a previous post used the same power supply I am currently using in my newer system, and this current system I built behaved differently, i.e. did not turn on at all, when not connected to my outlet. Could it be something with how these OEM motherboards were designed to receive power from the PSU? I had a shop test my PSU the other day, and it's still supplying clean power even at nearly 5 years old (I had forgotten I bought it the previous summer for my old HP system and it is actually older than I said in post #3). The PSU appears to be working perfectly for the time being.

We are talking about a desktop PC, right? And not a notebook?

Yes, the HP and Gateway systems I powered up without being plugged into the outlet were both Desktop systems. Both were made around 2008 when Vista was the primary OS with an upgrade to Windows 7 being free.

The only time I have heard of a surge getting through a UPS is when the transformer on the pole outside a client's house took a direct hit from lightning. That strike, besides starting the pole on fire, took out electronics in all 3 houses on that transformer. But short of a direct lightning strike from an angry Mother Nature, I have never seen or heard of an UPS failing to protect the connected equipment. Even if the batteries are failing and no longer capable of supporting the connected equipment during a power outage, they can still absorb massive surges - much better than a surge and spike protector.

There was a review I read where a guy's Tivo system was destroyed in a planned power outage using an APC system. APC tested the system and said it worked to its specifications, so they did not honor the insurance for the Tivo. One possibility is that the Tivo was plugged into the surge protection portion of the UPS instead of the Battery/Surge supplies, but I would still think APC would have found a fault in that line if that were the case. It is an odd review based on your feedback.

 

[Digerati]

Could it be something with how these OEM motherboards were designed to receive power from the PSU?

It should not have anything to do with a board being OEM or regular retail. But newer boards are designed with more "standby" features to speed boots and support various "wake on..." commands. For example, in standby mode, RAM will go into a low voltage state so data can be retained and will not need to be read in from the drive(s).

There was a review I read where a guy's Tivo system was destroyed in a planned power outage using an APC system. APC tested the system and said it worked to its specifications, so they did not honor the insurance for the Tivo.

I don't know what to say there. I would be interested to know the full home theater setup. For example, were the data/video inputs protected by a surge and spike protector too? Was the TV protected? Were the wall outlets properly "Earthed"? And I would like to get my hands on the Tivo device to troubleshoot and determine what exactly was wrong with it. It would not be hard to determine if the Tivo's power circuits took the hit (which would point to the UPS) or if the video side took the hit (pointing to something else).

I don't think APC would be untruthful about this. I mean what's a Tivo cost? Most are under $200. And all electronics will fail - eventually. It could have just been an untimely and unlucky coincidence the Tivo device decided to fail at that time.

I think you should go with what you are comfortable with. If you are uneasy with APC, don't get one. I'm just saying I've had great experience with APC - though I have never had to deal with their tech support. If you search on "best UPS" I think you will find APC, CyberPower and Tripp Lite mentioned most.

 

[writhziden]

Could it be something with how these OEM motherboards were designed to receive power from the PSU?

It should not have anything to do with a board being OEM or regular retail. But newer boards are designed with more "standby" features to speed boots and support various "wake on..." commands. For example, in standby mode, RAM will go into a low voltage state so data can be retained and will not need to be read in from the drive(s).

I did not mean to imply that it was due to them being OEM but more due to the boards an OEM tends to choose. Your hypothesis that it was more likely related to age is probably correct. I seem to recall my old Pentium 4 MSI motherboard exhibiting the same behavior where it would turn on briefly even after unplugging the PSU from the outlet.

There was a review I read where a guy's Tivo system was destroyed in a planned power outage using an APC system. APC tested the system and said it worked to its specifications, so they did not honor the insurance for the Tivo.

I don't know what to say there. I would be interested to know the full home theater setup. For example, were the data/video inputs protected by a surge and spike protector too? Was the TV protected? Were the wall outlets properly "Earthed"? And I would like to get my hands on the Tivo device to troubleshoot and determine what exactly was wrong with it. It would not be hard to determine if the Tivo's power circuits took the hit (which would point to the UPS) or if the video side took the hit (pointing to something else).

I don't think APC would be untruthful about this. I mean what's a Tivo cost? Most are under $200. And all electronics will fail - eventually. It could have just been an untimely and unlucky coincidence the Tivo device decided to fail at that time.

I think you should go with what you are comfortable with. If you are uneasy with APC, don't get one. I'm just saying I've had great experience with APC - though I have never had to deal with their tech support. If you search on "best UPS" I think you will find APC, CyberPower and Tripp Lite mentioned most.

It is not so much being uneasy with APC as being uneasy with buying anything over $100 without doing a bit of research into the investment. I want to make sure what I buy is high quality, safe for my electronics, and reasonably covered in the event of failure.

As far as the review, you can read it in the following link if you're interested in more details. The user does claim to have plugged his Tivo into the battery supply : Amazon.com: L. Jackson's review of APC Back-UPS 550VA UPS Battery Backup & Su...

I do note that it is one of the less expensive UPS units, so that may be a factor.

 

[Digerati]

as being uneasy with buying anything over $100 without doing a bit of research into the investment. I want to make sure what I buy is high quality, safe for my electronics, and reasonably covered in the event of failure.

I call that being wise. :)

Another advantage to not buying the cheap models is the better ones typically come with an LCD status panel. These are not really necessary but are nice. In fact, just the other day, my grandson and I came home from the park and all my UPSs were beeping and had shut down. But all the house lights were still on so it was not a power outage. A quick push of the UPS status button showed that the incoming line voltage was 146VAC!!! I panicked a little bit and stuck my multimeter into a wall outlet and sure enough 146.2VAC. I called the power company, said I was an electronics technician and told them to get someone out here right away or they were going to be paying to replace the air conditioner compressors, refrigerators, and freezers for everyone in this neighborhood running off that transformer. I guess I sounded convincing because within 30 minutes they had someone here. Fortunately, the transformer had taps so they could change the output, which is what they did. We are now at 122VAC which is just fine.

As for that Amazon review, it seems odd to me that of all the connected devices, only the Tivo device fried. He did say the Tivo was connected to the battery side so it should have been totally protected from any destructive anomalies - at least on the power side.

I do believe APC's product replacement policy is pretty common. I also note that 86% of the reviewers of that UPS gave it 4 or 5 stars, only 7% gave it just one.

 

[Tekno Venus]

then unplugged the power. I left the PSU in the on state and pressed the power button to close the circuit and drain any residual power.

There's a misconception here. With ATX power supplies, when you unplugged the power, the PSU was in a totally "off" state. And pressing the power button at this point does nothing.

Yep, it just turns some systems on for a brief second. My current system did not turn on, and I did not hold down the power button as you suggested; I merely pressed it for a couple seconds to see if it would turn on like my previous desktop system did. I figured that meant it did something to clear power from capacitors. Is that brief surge of power that used to occur on my previous desktop PC really nothing?

[/QUOTE]
Anecdotal evidence I know, but my machine occasionally does the same. If I switch it off and unplug it from the wall then press the power button, sometimes there's enough power for it to start to spin the fans. Not every time. But there's always enough power for the lighting on my motherboard to remain on for 5-10 seconds after unplugging it from the wall. Very high quality PSU too: Corsair SF450 450W Review

 

[Digerati]

I can see 5 - 10 seconds, but longer would indicate a problem or at least poor design.

But I may be causing some confusion here. There are no storage capacitors in the circuit for the front panel power switch. That's just a fact. So pressing "and holding" down that button does nothing more than simply pressing and releasing the button.

But pressing that button still puts a short (dumps the circuit to ground) on the circuit. IF the bleeder resistors in the power supply are faulty or the wrong size to begin with, the charge in the filter caps may not discharge as quickly as it should (which should be within a few seconds). In that case, pressing the button will still signal the PSU to start. But operating at 50-60Hz, they quickly discharge through their output side and never have a chance to charge again.

Note without a bleeder resistor a filter/storage capacitor can hold a charge for several days or longer. This is one reason there is usually a sticker on power supplies that says, "No user serviceable parts inside." Not to mention, anything that plugs into the wall can kill. :r1:

For the record, what I do when I want to "cold boot" a computer, I unplug it from the wall then count 10 to 15 seconds - 30 at most. That is plenty long to allow any residual voltages to decay to safe or inert levels. Any more is wasting my time.

"In theory" if your PSU has a master power switch, simply switching it to off should totally discharge any residual voltages much faster - assuming the power cord is in good working condition and the wall outlet is properly grounded to Earth ground. But my training has taught me to never trust the power cable for grounding - especially one that cost $.25 to make.

As for that Corsair - yep! That is a nice one. Sadly, not all Corsairs are these days. But your JG review made me think. I don't recall any PSU review from any site measuring output decay rates. So I just read through the ATX Form Factor standard for PSUs and I see nothing there either. I find that odd. They specify the required "hold-up time" (which is just 17ms as mentioned earlier) but not the decay rate. That is most likely the reason review sites don't measure for it.

BTW, JG does not test for hold up time. This is unfortunate as some otherwise excellent PSUs fail in that area. A short hold-up time is often the cause of sudden "unexplained" computer reboots. But I see on other sites that Corsair can hold up output for more than 20ms - which is excellent. :)