0
\$\begingroup\$

I've got a hardware platform (setup as host) which it's got a USB2.0/3.0 Micro type B female port.The system is definitely capable to detect and interface USB3 devices (can see in the embedded OS the USB3 and USB2 HUBS).

But in order to plug standard USB3 devices with a "normal" Type A connector I had to use this adapter.

The problem I've got is that all USB3 devices I'm plugging in are detected as USB2 ones instead. I tried different devices as cameras and memory sticks with same results.

I need to understand how an UBS3 device is detected over an USB2 one, I'm trying to dive into the spec and I'm no able to find a clear answer easily....

My initial guess is that this adapter is changing something in the physical layer that makes the USB3 device be detected as USB2...

  1. Would it be possible that the adapter is causing this problem?
  2. What can change at the physical layer that an USB3 device is forced to be detected as USB2?

Any ideas/clues will be appreciated.

\$\endgroup\$
  • 1
    \$\begingroup\$ There is something with USB OTG and pin 4. It mentions on the linked page the adapter is unsuitable for OTG. I suggest you search there. \$\endgroup\$ – Jeroen3 May 13 at 9:35
  • \$\begingroup\$ Thanks, yes I already seen that, but I found not connection with the fact the devices are detected as usb2...a usb3 memory stick for instance doesn't support OTG is a slave device.. \$\endgroup\$ – joe May 13 at 10:20
  • 1
    \$\begingroup\$ USB OTG is a host feature. \$\endgroup\$ – Jeroen3 May 13 at 10:36
  • \$\begingroup\$ Yes, that's right, the adapter doesn't support OTG but a memory stick is not OTG so it should be detected as usb3 device by the host but is detected as usb2 device instead. \$\endgroup\$ – joe May 13 at 10:40
3
\$\begingroup\$

Micro-AB USB 3.0 ports are usually OTG ports, and require proper ID pin assertion to work in USB host mode. And since your devices do operate at least as USB 2.0 devices, it means that the ID pin is wired correctly in this "adapter", ID = ground. So everything should be all-right.

However, many so-called "adapters" (in $1 range) are made with brutal violation of necessary impedance matching requirements for USB 3.0 signaling rate. Many are made with loose wires soldered between the u-AB plug and Type-A receptacle and then molded into the form. I have seen horrible impedance deviations in this kind of dongles, up to 150-200 Ω in differential impedance, so even USB 2.0 devices would behave flaky.

For USB 3.0 the SuperSpeed link detection comes first. The link gets the termination detected and proceeds with link training. But the transmission quality requirement for SS rate is 10-fold tighter than for HS, and there is no surprise that some adapters fail USB3 link training with horrible number of unrecoverable errors, the USB3 interface gives up, and the port/device reverts to USB2 mode.

My experience with adapters was always a gamble, you never know if the device is good or not. You never know. There are good adapters on the market, but you need to electrically characterize them using high-priced equipment (Time Domain Reflectometer, TDR). Or you can investigate the quality of adapter by using longer cables/extenders, and find the limit. In you case the adapter is very bad, you need to procure some other brand/shape, or request USB-IF certification data on the device.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.