USB 2.0 Full-speed Device Enumeration With Certain Cables – Intel® 600 Series

[HT77]

Hi!

Does anyone has experience with the following errata number 017 (USB 2.0 Full-speed Device Enumeration With Certain Cables) on Intel 600 series chipset?

Here is the link to Intel homepage:

https://edc.intel.com/content/www/us/en/design/ipla/software-development-platforms/client/platforms/alder-lake-desktop/intel-600-series-chipset-family-platform-controller-hub-pch-specification-upd/004/005/errata-details/ 

We use a USB 2.0 full speed device (banknote cash recycler module) with a 4,6m long cable, shorter is not possible.

Did anyone found a workaround, because in the errata sheet there is no workaround and also no fix.

We found one workaround but that is not really practicable for series production. If we use a USB hub device (in our case from Transcend) we will get a communication with our USB device and there is no USB descriptor failure. If we disconnect the hub, USB descripter failure occurs again.

We will be very thankfull for all comments on this topic.

[Diego_INTEL]

Hello @HT77,

Thank you for contacting Intel Embedded Community.

I have verified if there are any update regarding this issue and there is none at the moment. At least it has been identified but we must have to wait if there may be any solution soon. My apologies for the inconvenience.

Also, that is an interesting workaround, but of course, those hubs may be expensive and using one with every device would be costly. 

Best regards,

@Diego_INTEL 

[HT77]

Hello Diego!

Do you know if there is a fix planned in future. If not, can you check it?

I have also seen the 700 series chipset has the same bug/problem.

Best regards

Harald 

[Diego_INTEL]

Hello Harald @HT77,

It is exactly as you have mentioned, the sighting has been reported for both 600 and 700 series chipsets.

The latest report about this sighting is from November, 2022, and still a fix/workaround has not been available. It is under investigation as there are more people actively asking about this issue too.

I was looking for curiosity about what may be doing the difference from the hub and maybe is a retimer that is helping with the signal, but I'm yet not sure either, is just a comment from my part.

https://www.edn.com/why-is-a-retimer-required-for-high-speed-data-channels/

Best regards,

@Diego_INTEL