USB 3.0 ports stopping working

[j7n]

I have made the following observation. I have two computers with the Panther Point chipset that have some blue USB 3.0 ports. They have no drivers and work in USB 2.0 mode for booting and in Win 2003. (I only found a driver recently.) If I pull out a USB drive without using Safely Remove Hardware, the port stops working until the machine is powered down. A reboot is not enough. It feels like there is something like a resettable fuse. They are normally grounded to the case. One set is on the front_panel cable and another is on the i/o shield. Initially I thought that the USB 3.0 ports wouldn't work at all because of no drivers, which would suck because of only 2 or 4 ports remaining, until observed the pattern. If I use Safely Remove, there is no problem.

I was told by another person that he has not encountered this. But he is the kind of person who would not admit that any item he owns has any kind of flaw.

Have you seen this behavior?

[Tripredacus]

Not working like Windows not detecting a new device or not working electrically? If you redo this test but put a mouse instead, it should light up even if Windows doesn't recognize it. 

I have no concept of "Panther Point" chipset, it would be easier for you to put the board model in your post. Perhaps there are BIOS settings that are related to USB that can be helpful.

[j7n]

They don't work in Windows and can't be used to boot the computer in BIOS. When plug in a mouse, the mouse lights up, but Windows doesn't see it. The LED on a storage device initially comes on and goes out immediately.

The motherboards are: GA-B75M-D3H (American Megatrends EFI BIOS) and DH77KC (Intel Visual BIOS / grey AMI).

I fear that I may have damaged the B75M now. The Kingston DataTraveller 111 now doesn't connect to the USB 3.0 port at all even when the computer has been just turned on. It works in the USB 2 ports; but it can also disable them. (Mouse is powered on but not seen by computer.) It still works in DH77KC. The speed is increased to 70 MB/s from 32 MB/s in USB 3.0 mode. This DataTraveller might be a cancer and somehow shorting the port when unplugged.

In the BIOS I have the following settings:

XHCI Pre-Boot Driver: Enabled

XHCI Mode: Smart Auto

HS Port {1..4} Switchable: Enabled

XHCI Streams: Enabled

[D.Draker]

DH77KC This board is over 12 years old. It's actually fantastic it worked that long, the capacitors are dead, most likely.

[D.Draker]

21 hours ago, j7n said:

GA-B75M-D3H

This one is even older! And Gigabyte is known for using junk capacitors.

[Klemper]

I once had a desktop with a strange glitch. The brightness was increasing by itself.

Firstly, I replaced the GPU, secondly, the PSU. Nothing helped until I replaced the mobo.

Gigabyte on G41 chipset. 4 years old at that time.

[j7n]

That comment was not constructive. I put together this system new and love it. It boots fast, and the BIOS is nice looking, and navigable with the keyboard like in old days. I've been a fan of the Gigabyte brand and bought into their ultra durable marketing (even though other boards soon had the same parts). It was my mom's PC and wasn't driven hard.

It may have been "broken" years ago. I put XP on it because if familiarity, and accepted that USB didn't work.

All boards have a bunch of extra unused USB headers. But the brackets never come with anything and must be ordered by mail.

Edited December 4, 2024 by j7n

[Tripredacus]

I've had various USB issues in the past with a DG45FC board that were similar to what you describe. I don't know if the ports were what type, all I know is that some devices would not work on some ports or other times (for drives) would take a long time for Windows to detect them, like up to an hour. In the end I do not know where the issue was with that board, as I eventually had to replace the board because the video card wasn't up to snuff anymore and ended up making a new OS install. 

However, I did manage to limp along by occasionally using USBDeview to remove all of the USB devices on the system and rebooting. Also doing it in Safe Mode if needed and it seemed to help although as I said I don't know what the actual issue was. I suspect it was hardware related.

https://www.nirsoft.net/utils/usb_devices_view.html

And a note for using this tool, if you are using a USB keyboard and delete all the keyboards, obviously you are going to have a bad time. 

[D.Draker]

I had an OEM board made by Acer. In 2007 or so. LGA775. It came with an Acer ready-made PC. All USB ports suddenly died. Later I read those were affected by static electricity charges.

[D.Draker]

14 hours ago, j7n said:

It was my mom's PC and wasn't driven hard.

It doesn't matter, the caps have a limited shelf life. But most don't notice because they upgrade every 1-4 years.

Probably, you were too young to remember the early 2000, when caps on board lived for 1-2 years at best.

[mjd79]

Nowadays (and even those 12 years ago) standard electrolytic capacitors are not used on most motherboards, not like as on older boards. The new ones practically never lose capacity.

Edited December 4, 2024 by mjd79

[D.Draker]

The failed 14 y.o. GA-B75M-D3H has the trivial capacitors (electrolytes). Nothing fancy.

https://image1280.macovi.de/images/product_images/1280/794307_1__8428640-1.jpg

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[Klemper]

23 hours ago, mjd79 said:

Nowadays (and even those 12 years ago) standard electrolytic capacitors are not used on most motherboards,

I think it's their marketing strategies. They still are. Ceramics are expensive. 

[j7n]

Polymer capacitors were added on most boards around the CPU. Gigabyte and soon others added them all over the board. But the Ultra Durable™ was presented as a big deal. Pentium 4, one of the word CPUs of all time, on Socket 478 boards often could be seen cooking old style capacitors. But that is not the topic.

There was another issue with cheap computer cases where the shield of the connector was not connected to the metal case because the whole front was plastic. It has a ground in one of those 4 wires and the a fifth that is the shield. That caused the PC to lock up when a device was plugged in (not removed like in my case) because static was discharged into the connector.

[D.Draker]

11 hours ago, j7n said:

There was another issue with cheap computer cases where the shield of the connector was not connected to the metal case because the whole front was plastic. It has a ground in one of those 4 wires and the a fifth that is the shield. That caused the PC to lock up when a device was plugged in (not removed like in my case) because static was discharged into the connector.

That may be it! I also don't like when Flash drives are all metal. One can still easily discharge all the hell onto the USB, be it with, or without the grounded case.

I usually ground myself with a piece of wire, when working with electronic parts (maintenance and the such).

I also touch something grounded before inserting flash drives.