DS90CR216A: Black lines on screen

Hi Brian,

Many times the receiver is not the cause of the problem, but where the problem is observed. In order to help debug the issue, information is required throughout the data path:

What is the input device PN to the LVDS receiver?
Do you have jitter characteristics of the clock source? This would be clock input to the receiver.
What clock source are you using? Have you tried to change to a clean clock source? What are the results?
What are the characteristics of the cable(s)? Do you know the attenuation / loss of the cables?
Is the layout optimized for 100 ohm differential impedance?
Is the board optimized for high frequency in PCB layout? Customer is recommended to follow the notes on page 23 in the datasheet.

Thanks,
Dennis

Hi Brian,

Are the lines horizontal or vertical? I would expect with bit errors that the black lines are horizontal.

Can you replace another part to see if two black lines go away?

Dennis

Hi Dennis,

Thanks for your reply,

The two strips are about 1" thick and are separated by 1/2" space located in the center of the screen. The strips are vertical.

We have tried 40 different screens and PCB that contain the LVDS receiver, Host controller and cables.

We see some correlation from the different length of cable (3M 14526-EZ5B-xxx-02C) used to connect the LVDS transmitter (IMX6 processor) and the DS90CR... A 0.5 M cable almost always fails and the screen displays the lines. A 1M cable fails on an intermittent rate. A 2M cable almost never fails

We have checked and believe that the transmitted signal is as required

The cables have an impedance of 100 ± 10Ω

We have copied the layout as closely as possible with the routing optimized for impedance and length matching

Brian

Hi Brian,

If there are vertical lines, it implies that there are errors at the same place as each horizontal line displays. I am wondering if there may be some kind of receiver skew margin, which is very difficult to determine when the DS90CR216A is used in the entire system. There could be other contributions to this issue elsewhere.

Since you see the performance is better at 2m cable instead of 0.5m or 1m, can you please probe across the 100-ohm differential termination on the DS90CR216A input? It would be good if you get a capture of one data channel and the clock channel. This will tell us if somehow there is noticeable degradation when the cable lengths change.

You may email me the eye diagrams at dennis.wu@ti.com

Dennis

Hi Dennis,

Have you ever seen anything like this issue (black stripes on screen) previously.

Brian