TFP410: UXGA blinking

Hello,
Can you elaborate more on what do you mean by "blink", do you have some captures?
Can you get a scope capture of the pixel clock when the issue happens?
What is the temperature in which the issue happens?
Have you tested other frequencies/resolutions? These devices have different PLLs for frequency ranges, can you do a frequency sweep from what you are using an below to see where the issue stops?

Regards

Hello,

> Can you elaborate more on what do you mean by "blink", do you have some captures?

It depends on the monitor.

Some monitors are completely blacked out for a few seconds, and some monitors have distorted several frames of images.

Original image

Blackout image

Distorted image

> Can you get a scope capture of the pixel clock when the issue happens?

Is pixel clock mean "PIN 57 IDCK +"? Or do you mean "PIN 21, 22 TXC + -"?

There is no trigger to scope capture the pixel clock at the moment the issue .

What kind of abnormality do you think is?

It is possible to focus on that point and measure it.

> What is the temperature in which the issue happens?

There are two "TFP 410" on one board,

One is 25 ° C to 40 ° C and the other is 0 ° C to 15 ° C.

It is the same trend on another board.

DVI connector -> TFP 401 -> TFP 410 -> DVI connector
                                                -> TFP 410 -> DVI connector

> Have you tested other frequencies/resolutions? These devices have different PLLs for frequency ranges, can you do a frequency sweep from what you are using an below to see where the issue stops?

1080p(148.500MHz) -> OK

WUXGA(158.250MHz) -> OK

UXGA(162.000MHz) -> NG

Are there other resolutions to check?

Regards

> Can you get a scope capture of the pixel clock when the issue happens?
Quantum Data 980 was used to measure pixel clock.
Measurement is done three times and three images are attached.

The measurement temperature is the top of the chip case.

The value of TMDS CLK has changed.

Case1_1: before issue

Case1_2: issue

Case1_3: after issue

Case2_1: before issue

Case2_2: issue

Case2_3: after issue

Case3_1: before issue

Case3_2: issue

Case3_3: after issue

Regards

I am still expecting your response regarding this issue.

Hello,
This could be a marginal parameters issue.
What is the failure rate?
Can you swap a TFP device from a passing to a failing board to see if the failure follows the part?
What are the lot codes of the failing devices?
Please confirm the following assumptions:
1) The issue only happens with a pixel clock of 162MHz.
2) When using a pixel clock of 162MHz and controlling the temperature to not pass certain limit, the issue is not present. If so, what is that temperature limit?

Regards

Hello,

>What is the failure rate?

The failure rate is about 95%.

>Can you swap a TFP device from a passing to a failing board to see if the failure follows the part?

Failure will be handed over.

>What are the lot codes of the failing devices?

6938716TW6 1644+5
6962965TW6 1644+5

>Please confirm the following assumptions:
>1) The issue only happens with a pixel clock of 162MHz.

Yes.

>2) When using a pixel clock of 162MHz and controlling the temperature to not pass certain limit, the issue is not present. If so, what is that temperature limit?

What is temperature limit?

Hello,

I've just realized you have a TFP401 and a TFP410, we need to isolate the failing device, can you share your schematic? you can send it to elias.villegas@ti.com

It looks like one of the TFPs is losing sync.

We need to look at the voltage rails and the clocks when the issue occurs, for both the TFP401 and TFP410.

We also need to look at the SCDT signal in the TFP401, is it going low when the issue occurs?

Have you tried with a different source? Is it DVI or HDMI?

We are tracking down the test results for the lot codes you provided.

I didn't understand your answer for the A-B-A swap test, is the failure following the part or is it following the board?

Regards