ISOUSB211: USB2.0 Compliance Failure

Hi Shitij,

Thanks for reaching out and providing detailed inputs. Few questions and inputs below:

1. Can you please share a rough block diagram of the test system  -  this will help us see the total number of devices attached and sequence of connection.
2. Can you remove the hub in upstream and check the eye diagram only with ISOUSB211
3. This can also be an issue of wrong triggering because the flat horizontal zero line in middle of eye does not make any sense to me.
   1. RIGOL software suite is something we have never done hands on so not sure which settings to play with.
4. Can you check if the device and device on EVM has X marked initials?
   1. Need to be sure that you have latest released devices of ISOUSB211.

Regards
Varun

Hi Varun,

1. This is the block diagram of the system. There are 3x usbc ports connected through the isolator chip and 1x USB-A port connected firstly to the hub. The eye diagram taken at the USB A connector is good, there is no end of packet or sync bit count failure. All ports are unconnected except the one under test while running compliance. The usb hub and isolator chips share the same PCB. It is first powered on and then connected to the host PC running HSETT

2. The third eye diagram in the original post was taken without the hub. That test case had the eval board directly connected to the HOST PC running xHSETT.

3. That could be the case, however, we are only seeing this happen with the ISOUSB211 chip. Measuring the eye-diagram directly at the USB port of the HOST PC and the USB HUB does not show any issue. They all pass and look like the second eye diagram shown in the original post.

4. The device marking is ISOUSB211. There is a another string: 26ACP3T and G4. The eval board is ISOUSB211DPEVM and has INT133A marked on it.

Thanks,
Shitij

Hi Shitij,

Thanks for the details. Let's take this forward based on your input 2 which is stated as below: 

Shitij Avlani said:

The third eye diagram in the original post was taken without the hub. That test case had the eval board directly connected to the HOST PC running xHSETT.

So this statement says that ISOUSB211 EVM when tested standalone with PC and USB-C test fixture, the eye diagram fails in scope.

Question based on you above observation:

1. How did you connect USB A port on Downstream side of EVM to your USB-C Test Fixture
   1. Can you share picture of you test fixture?
2. The test fixture used, is it as per USB2.0?
   
   1. Details of test fixture will help more here.
3. HOST PC port to which Upstream side of EVM (USB-B port) is connected - Can you verify in the PC if the port is USB2.0 or USB3.0.

ISOUSB211 Passes USB2.0 compliance eye diagram test with USB2.0 Host as well USB2.0 HS compliant test fixture. This is explicitly mentioned in the datasheet under Figure 6-1 and respective result can be seen in Figure 5-3.

Regards
Varun

Hi Varun,

1. The output of the eval board was connected with a USB-C to USB-A passive converter as shown in the picture below.

2. The test fixture is shown below. It only breaks out the DP and SN lines to an SMA connector

3. The eval board is connected to the PC through a USB 2.0 type-A to type-B cable. The USB port on the host PC is USB3.0 capable but that should not affect compliance. I have tested the host PC USB port directly with the same type-A to type-C passive converter and test fixture; and it passes compliance without any issue.

Thanks,

Shitij

Hi Shitij,

Thanks for the details. Let me ask around to some USB experts in my team get back to you by Monday, 25th August.

May I know if Mendaera Inc, US has NDA signed with TI or not?

Regards
Varun

Hi Varun,

Thanks for that! We have not signed an NDA with TI yet. What are the steps/process for it? Is there a point of contact?

Thanks,

Shitij

Hi Shitij,

Wanted to follow up on this issue. Did it get resolved at your end?

If not, can we communicate over your registered email?

Regards
Varun