Pages: [1] 2   Go Down

Author Topic: Internal USB3 card reader problems  (Read 18808 times)

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Internal USB3 card reader problems
« on: September 26, 2012, 02:07:23 pm »

I've recently built a new system based on the components below. Generally the system built with it has performed to expectations.
However I'm running into a regular problems with the onboard USB3 header.

I have been trying to use it with a front panel Asaka AK-ICR-17 USB card reader and hub. More recently, to see if the Asaka panel is faulty, I bought another USB3 card reader front panel which has displayed the same issue, but has allowed me to tie down the problem more accurately.

When first installed the card readers (same behaviour with both) are seen and after Windows installs the necessary drivers, they work correctly. However when re-booted they are no longer available on the system.
On investigation; the second card reader has an LED power light. When first installed this lights up and remains lit.
When the system is first rebooted the power light remains on until the 'starting Windows' splash screen, then goes off and when at the desktop, the reader isn't seen.
On subsequent re-boots the light no longer comes on at all, even when booting into bios.
There are no indications in device manager that there are any hardware problems.

If the reader header cable is removed, the system re-booted, closed down then the header re-connected the same installation loop of being seen, then being lost on re-start happens again.

I'm unclear if this is a motherboard or OS issue, or a combination of both.
I've tried the hardware vendor, manufacturer and a couple of other computer support forums, but no one seems to be able to help with this. I wonder if anyone has come across this issue before and have any idea of how it can be resolved ?

Thanks in advance.

Gigabyte GA Z77X-D3H (rev 1.0) latest bios F15 running all the latest drivers.
Intel i7 3770k processor (not overclocked)
32gb Ram (Corsair XMS3)
Windows 7 64bit (fully patched and updated)
Logged

John.Murray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 886
    • Images by Murray
Re: Internal USB3 card reader problems
« Reply #1 on: September 27, 2012, 12:14:52 am »

Question - is the only USB 3.0 device plugged into the header this reader? Have you tried any other USB devices?  I've seen shorted USB header cables / devices cause issues.

Try booting without the reader attached, open Device Manager and expand the USB hub - now attach the reader.....
Logged

Steve Weldon

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1479
    • Bangkok Images
Re: Internal USB3 card reader problems
« Reply #2 on: September 27, 2012, 01:12:30 am »

Are you saying when doing a cold boot (power on from off) they're seen and work fine, but when subsequently do a hot boot (restart from Windows or the reset button) they are no longer seen?

If this is the case.. I'd look at the BIOS options and see what 'might' be causing it.   It could very well be a combination of this particular card reader needing a certain BIOS or Windows setting to come up off a hot boot.
Logged
----------------------------------------------
http://www.BangkokImages.com

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #3 on: September 27, 2012, 07:47:28 am »

Question - is the only USB 3.0 device plugged into the header this reader? Have you tried any other USB devices?  I've seen shorted USB header cables / devices cause issues.

Try booting without the reader attached, open Device Manager and expand the USB hub - now attach the reader.....
Thanks for the interest.

The motherboard has 2 USB3 controllers; a VIA one that drives the ports on the MB backplate (which all work fine) and the Intel one that just drives the header on the motherboard.
I've tried both Card readers and they behave the same, but I haven't anything else that could plug into it yet. It seems unlike both would have the same fault.

Device manager shows the same hubs regardless of what's plugged in. When the readers are first connected they show up as Generic storage device USB device under disk drives and also as portable devices (each reader CF/SD/MS etc shows as a different drive)

I'm rather reluctant to risk messing around with direct motherboard connections when the system is fully powered though.

I'm guessing that there's an Intel driver somewhere that's switching things off during the boot process, but don't know where to look further to resolve the issue.
Logged

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #4 on: September 27, 2012, 08:15:32 am »

Thanks for the help Steve,
Are you saying when doing a cold boot (power on from off) they're seen and work fine, but when subsequently do a hot boot (restart from Windows or the reset button) they are no longer seen?
No, everything is being done as complete shut downs and restarts, no hot restarts.
Quote
If this is the case.. I'd look at the BIOS options and see what 'might' be causing it.   It could very well be a combination of this particular card reader needing a certain BIOS or Windows setting to come up off a hot boot.
The motherboard only has one specific bios USB3 option and that's for the VIA hub and it's enabled.

The Intel USB3 hub seems to be influenced by EHCI & xHCI options. Everything seems to be enabled, but I'm reaching the limits of my knowledge trying to work out if things like smart auto or auto are  better options than the performance defaults.

The bios options stay the same regardless of whether the readers work or not, so it's puzzle why it doesn't indicate it's powered up on the second boot up. Maybe Windows is changing the reader's firmware and it resets itself when removed from the system ?
Logged

Steve Weldon

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1479
    • Bangkok Images
Re: Internal USB3 card reader problems
« Reply #5 on: September 27, 2012, 11:22:21 am »

Thanks for the help Steve,No, everything is being done as complete shut downs and restarts, no hot restarts.The motherboard only has one specific bios USB3 option and that's for the VIA hub and it's enabled.


I'm still not understanding.  Are you saying from a cold boot this 'sometimes' work and 'sometimes' doesn't work?   I'm not understanding what conditions make it work for you and what conditions make it not work.

I'm using this Koutech model and have installed maybe 20-25 of these on various MB's including yours.. and have had no issues.   I find with these types of items finding one that works and keeps working is a blessing so I'll keep using it as long as I can.  With USB2.0 internal card readers less than half would work as they should with more than a few motherboards so when I found one that did.. I'd stock the shelves.. :)  It's just not one of those things a major manufacturer makes a lot of money on so we're dealing with the rest.

Is this a fresh Win7 install?  I'm asking because sometimes a software install will write something to the registry and cause issues.

I would try what John suggests, chances or hurting your MB are very slim.. just be careful to not short something out in haste.

Just to make sure.. on mine, when you right click on the drive in the device driver and select properties.  Under the "policies" tab I use "quick removal" because of the nature of the device, and when you right click on the drive in Windows Explorer and select properties and then the tab "poilicies" I have selected "better peformance" on the top and "enable write caching" on the bottom.   These are the choices the device selected during install.. curious if yours are the same of if they make a difference?
Logged
----------------------------------------------
http://www.BangkokImages.com

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #6 on: September 27, 2012, 12:01:26 pm »

I'm still not understanding.  Are you saying from a cold boot this 'sometimes' work and 'sometimes' doesn't work? 
Not intermittent, but absolutely repeatable.
When installed reader is seen and works OK, power light on.
Next re-boot, power light on during POST, goes off when Windows starts up and card reader can't be seen.
On further re-boots, power light never comes on and reader never seen again.
If disconnected and run through a boot cycle back to square one.
Same behaviour with both readers.
Quote
I'm using this Koutech model and have installed maybe 20-25 of these on various MB's including yours.. and have had no issues.
This looks very similar, but mine doesn't have two USB plugs to connect it, both have a motherboard 20 pin plugs.
This issue seems specific to the Intel header on the motherboard, not the rear sockets on the VIA controller.
Quote
Is this a fresh Win7 install? 
The first reader was originally installed within a day or two of commissioning the system. It's just taken a while to define the problem.
Quote
I would try what John suggests, chances or hurting your MB are very slim
Not sure what doing this would tell me that I don't already know. With the OS on a SSD reboots are pretty fast anyway.
Logged

Steve Weldon

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1479
    • Bangkok Images
Re: Internal USB3 card reader problems
« Reply #7 on: September 27, 2012, 01:13:33 pm »

Not intermittent, but absolutely repeatable.
When installed reader is seen and works OK, power light on.
Next re-boot, power light on during POST, goes off when Windows starts up and card reader can't be seen.
On further re-boots, power light never comes on and reader never seen again.
If disconnected and run through a boot cycle back to square one.


What is the time span between reboots.. and does the time span affect this cycle?   It sounds like the power supply isn't sitting long enough to completely drain it's caps and they're acting like a backup battery of sorts.  Pulling the connect interrupts this "battery" allowing the cycle to repeat.  Now.. to figure out what this is telling us.  ;D

Which means the reader is supplying some signal to the motherboard which isn't allowing it to be seen.   This leads me back to a possible BIOS settings and away from a Windows program.  The BIOS and booting sequences are happening before Windows even boots..

It could be an incompatibility with the motherboard.. actually it obviously is.  But why?  BIOS setting, BIOS needs to be updated, or MB design will never be compatible..

I'd start with making sure the BIOS is updated to it's latest version.  If that doesn't fix it then consider a beta version if there is one.

Next would be to contact Gigabyte.. but their replies can take weeks and their form is almost comical in the amount of information they want just to let you ask a question.  Still.. this is the choice.

OR/andor/Finally..  Replace the card reader.  It's $30.. the MB is a lot more.   The one I showed you worked fine for me.


If you had a high end multimeter and the right type of probes we could measure the logic states at the connector and find out which one is being left hot.. but this bit of knowledge wouldn't change your subsequent steps.


Quote
This issue seems specific to the Intel header on the motherboard, not the rear sockets on the VIA controller.The first reader was originally installed within a day or two of commissioning the system. It's just taken a while to define the problem. Not sure what doing this would tell me that I don't already know. With the OS on a SSD reboots are pretty fast anyway.

I'm assuming you've checked for the latest drivers for the USB3 controllers?  If I'm right above this 'shouldn't' make a difference but it's still a good practice.

When I connect these I either go through the back panel like in that link I posted.. or I use the 20 pin adapter (also in the link) to to connect to the MB connector.. and no issues with the Koutech.

If it works right on one USB3 controller and not the other.. this means it's almost for sure a BIOS/motherboard issue WITH that specific reader.   Gigabytle will never know to fix it if it's only the one reader and not all of them.. especially if people don't take the time to fill out their question forms.    You've tried their forums?  Often I'll find the problem and solution in the same thread..

I can't tell you exactly what John in thinking though I have a guess.. but I'm sure he knows.  His hardware logic rarely is off.
Logged
----------------------------------------------
http://www.BangkokImages.com

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #8 on: September 27, 2012, 02:06:07 pm »

Thanks again for your continuing help Steve.
What is the time span between reboots.. and does the time span affect this cycle? 
No, leaving it overnight doesn't change the behaviour.
Quote
It could be an incompatibility with the motherboard.. actually it obviously is.  But why?  BIOS setting, BIOS needs to be updated, or MB design will never be compatible..
I'd start with making sure the BIOS is updated to it's latest version.  If that doesn't fix it then consider a beta version if there is one.
Yes, running the latest stable bios (and all other drivers too).
This problem isn't so severe that I'd risk using a beta bios that might cause wider or more serious system problems.
Quote
Next would be to contact Gigabyte.. but their replies can take weeks
Err yes, the enquiry went to them last week, no reply yet :(
Quote
OR/andor/Finally..  Replace the card reader.
Yes, on the second internal already, but I do have an external USB3 reader to run off the external ports for now.
Quote
If it works right on one USB3 controller and not the other.. this means it's almost for sure a BIOS/motherboard issue WITH that specific reader.   
Having spent another hour with Google, it looks like this issue isn't confined to Gigabyte boards, but also Intel MB users are suffering in the same way.
There's some talk on the Intel support forums of people hacking pin 10 out of the header on the motherboard as it may be causing the problem if improperly terminated. Maybe a less drastic fix might be tried by surgery to the reader's plug instead.

I'm also going to try a header > twin USB3 adaptor and see how that works, although from the Intel forums I don't think it will help much, but they're cheap as chips to try.
Logged

John.Murray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 886
    • Images by Murray
Re: Internal USB3 card reader problems
« Reply #9 on: September 27, 2012, 02:15:21 pm »

Not sure what doing this would tell me that I don't already know. With the OS on a SSD reboots are pretty fast anyway.

If the reader cable has a short, or the reader's current consumption is excessive, you'll see the USB hub go down.....
Logged

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #10 on: September 27, 2012, 03:12:34 pm »

Thanks John,
If the reader cable has a short, or the reader's current consumption is excessive, you'll see the USB hub go down.....
The hub still appears in Device manager even if the reader is not being seen by Windows.

But they both work fine when first connected until the first reboot.
Surely if there was a cable fault or excessive current consumption you'd never be able to use it at all ? Plus it is unlikely two different units would have the same cable fault*.

* Although in reference to the Intel problems mentioned above; it's possible both units may have earthed pin 10 unnecessarily or some-such.
Logged

John.Murray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 886
    • Images by Murray
Re: Internal USB3 card reader problems
« Reply #11 on: September 27, 2012, 05:04:16 pm »

could be the reader's current consumption is at or near the limit.  i've never had an issue with this one:

 http://www.silverstonetek.com/product.php?pid=315&area=en

you might also check for a firmware update on your chosen reader?
Logged

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #12 on: September 27, 2012, 05:55:25 pm »

i've never had an issue with this one:
 http://www.silverstonetek.com/product.php?pid=315&area=en
Thanks again.
I might punt for that if all else fails.
Quote
you might also check for a firmware update on your chosen reader?
No firmware updates available for the Asaka reader. For a cheap ebay clone ? who knows ? I doubt it.
Logged

Steve Weldon

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1479
    • Bangkok Images
Re: Internal USB3 card reader problems
« Reply #13 on: September 27, 2012, 08:53:54 pm »

Thanks again for your continuing help Steve. No, leaving it overnight doesn't change the behaviour.Yes, running the latest stable bios (and all other drivers too).
This problem isn't so severe that I'd risk using a beta bios that might cause wider or more serious system problems.Err yes, the enquiry went to them last week, no reply yet :(Yes, on the second internal already, but I do have an external USB3 reader to run off the external ports for now.Having spent another hour with Google, it looks like this issue isn't confined to Gigabyte boards, but also Intel MB users are suffering in the same way.
There's some talk on the Intel support forums of people hacking pin 10 out of the header on the motherboard as it may be causing the problem if improperly terminated. Maybe a less drastic fix might be tried by surgery to the reader's plug instead.

I'm also going to try a header > twin USB3 adaptor and see how that works, although from the Intel forums I don't think it will help much, but they're cheap as chips to try.

Well.. this type of problem isn't that unusual and is why I'll really push my clients towards a motherboard/components I have personally tested and know work.  Every time there's a new generation motherboard released we go through this in one way or the other and the best we can hope for is that the chipset is popular enough for the manufacturers to clean up most of the issues.  And while I love gigabyte and asus, it's why for professional clients I like going with Intel.. they validate their hardware much better on average.

I'd agree you should do any connector/wiring changes on the card reader side.  If you have the right tool you can actually push out pin 10 from the connector.  But cutting it a few inches up knowing you can solder/heatshrink them back together is far preferable to breaking a pin off the MB.. 

When I said "try another card reader" I'm sure you knew I meant another brand?  At this point I sincerely doubt we're taking about a 'defective' anything.. just a small glitch they'll probably fix in the next firmware upgrade.

Gigabyte is pretty solid with their beta firmware.. which I know sounds forgiving considering how many versions they "fix" on average per MB.. but I've never seen them put something up that will cause a crashed system or anything where you couldn't flash it back.

Logged
----------------------------------------------
http://www.BangkokImages.com

John.Murray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 886
    • Images by Murray
Re: Internal USB3 card reader problems
« Reply #14 on: September 28, 2012, 03:14:27 am »

The pin 10 issues you allude to are indeed related to overcurrent protection:

http://download.intel.com/technology/usb/USB_3.0_Internal_Connector_and_Cable_Specification.pdf

I've seen vendors mis-interpret "pulldown" (ie: set to "0" or "1") as "short" instead of using a resistor, all in the effort to reduce cost.  Some of Rose***'s serial RS-232 implementions immediately come to mind (yeah - I still get to mess around with HP pen plotters and various Gerber devices)

I'd be hesitant to disable it, we really want the O/S to be aware of power issues.  Give the Silverstone or Steve's recommended a try...
« Last Edit: September 28, 2012, 03:20:39 am by John.Murray »
Logged

kaelaria

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2223
    • http://www.bgpictures.com
Re: Internal USB3 card reader problems
« Reply #15 on: September 28, 2012, 08:03:30 am »

I agree with the above - simply get a different reader.
Logged

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #16 on: September 28, 2012, 04:29:51 pm »

Give the Silverstone .....a try...
OK, I've ordered one and will report back. Delivery will take a few days though as it seems to be out of stock at most places.

Logged

PierreVandevenne

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 512
    • http://www.datarescue.com/life
Re: Internal USB3 card reader problems
« Reply #17 on: October 01, 2012, 08:23:00 am »

Did you connect both the internal cable and the external one? They both need to be connected. This being said I bought a AK-ICR-10U3 a couple of years ago and it is/was(sitting in a drawer now) frequently unreliable on Asus, Gigabyte and Intel boards.
Logged

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Internal USB3 card reader problems
« Reply #18 on: October 02, 2012, 01:29:47 am »

Did you connect both the internal cable and the external one? They both need to be connected.
Only the Asaka has multiple connectors and, yes, they were all connected.
I also don't think you're correct in this assertion in needing to connect all the leads; you only need to do that if you require all the facilities.
Logged

PierreVandevenne

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 512
    • http://www.datarescue.com/life
Re: Internal USB3 card reader problems
« Reply #19 on: October 02, 2012, 06:53:01 am »

Probably not needed for the card reader part, but on my older model, it simply shuts down, not unlike what you describe, when the external cable isn't connected eventhough it has a separate power cable.  My older version is a somewhat puzzling piece of hardware as it has power through a standard molex, through the sata power, and both usb connectors.

Correction: just arrived at the office and fetched it from garbage drawer and no molex on that one. Still shuts down completely when external USB cable removed.
« Last Edit: October 02, 2012, 07:41:43 am by PierreVandevenne »
Logged
Pages: [1] 2   Go Up