DS90UB926Q-Q1: Internal Test Pattern Generation

Amy Luo

Expert 6895 points

Part Number: DS90UB926Q-Q1

Hi team,

One of our customer's issues, I'm forwarding it below, could you please provide some troubleshooting suggestions

Deserializer: DS90UB926Q-Q1

Display: Rgb888 800*480 5-inch,

No serializer and LVDS input, Only 926 screen connected

Using the 926 internal pattern generator to light up the screen. I find that sometimes the display is normal after I power up, sometimes the display is abnormal like the picture below. Black lines are displayed.

The top(white) image is missing part of the left side with black lines on the top. The bottom(green) image is complete on the left, but there are black lines on the top.

When power-up display is normal, the picture stays for a little longer, it will automatically appears abnormal like the white image, or green image, or splash screen

Please check if there is a configuration error, or if there is anything else that needs to be configured, or DE HS VS needs to be modified on hardware. Currently these three pins are series of 100 Ohm resistors connected to the display. The 926 is configured as follows

(0x01, 0x04)
(0x02, 0xFA)
(0x03, 0xF0)
(0x04, 0xFE)
(0x05, 0x2E)
(0x06, 0x00)
(0x24, 0x08)
(0x29, 0x60)
(0x2A, 0x00)
(0x2C, 0x00)
(0x41, 0x03)
(0x44, 0x60)
(0x56, 0x08)

(0x65, 0x04)
(0x64, 0x11)

Best Regards,

Amy Luo

over 2 years ago

0 Fadi Abdulhameed over 2 years ago

TI__Mastermind 29295 points

Hey Amy,

Thanks for the request. A few things to clarify:

How many ICs with this issue are you seeing?
Is this a new application?
What's the failure rate?
Have you done ABA swap, does the issue follow this suspect IC?
Can you provide your schematic?
What PCLK freq. and resolution are you running here?
Does this issue only happen at power-up or does it power-up good and goes to bad ?
On a good run, does the display stay good or does it also fail after some time?
Can you provide a register dump of a good run vs. a bad run?

Regards,
Fadi A.

0 Amy Luo over 2 years ago in reply to Fadi Abdulhameed

TI__Expert 6895 points

Hi Fadi,

Thank you for your reply. The following is customer feedback:

1. At present, we have tried 3 ICs , all of them are the same phenomenon.

2. Yes, it's a new application.

4. The display has also been replaced and this is the same behavior, thus eliminating the possibility of a bad display.

6.The PCLK is around 25.8MHz and the resolution is 800*480.

7.Both of these situations occur, sometimes at power up or some time after power up.

0 Fadi Abdulhameed over 2 years ago in reply to Amy Luo

TI__Mastermind 29295 points

Hi Amy,

Thanks for the additional information. I will review this and get back to you within a week.

Regards,
Fadi A.

0 Fadi Abdulhameed over 2 years ago in reply to Fadi Abdulhameed

TI__Mastermind 29295 points

Hey Amy,

Can you measure the PCLK at the 926 output?

Regards,
Fadi A.

0 Amy Luo over 2 years ago in reply to Fadi Abdulhameed

TI__Expert 6895 points

Hi Fadi,

The customer feedback that the PCLK is a 25.64 MHz waveform.

0 Fadi Abdulhameed over 2 years ago in reply to Amy Luo

TI__Mastermind 29295 points

Hey Amy,

I meant can you measure the PCLK output on the scope to confirm that the signal frequency coming out of the 926 matches the expected Pclk freq. of 25.64 MHz?

Regards,
Fadi A.

0 Amy Luo over 2 years ago in reply to Fadi Abdulhameed

TI__Expert 6895 points

Yes, the PCLK output signal frequency is 25.64 MHz.

0 Fadi Abdulhameed over 2 years ago in reply to Amy Luo

TI__Mastermind 29295 points

Hi Amy,

I'm currently out of office. I will review and provide next steps when I get back to office by Tuesday 11/15.

Regards,
Fadi A.

0 Fadi Abdulhameed over 2 years ago in reply to Fadi Abdulhameed

TI__Mastermind 29295 points

Hi Amy,

It seems like a timing issue, you can try probing HS/VS and DE signals with a scope or logic analyzer to see if the timing of these outputs is as expected and no jitter issue, etc.

I can't see the values on your schematic, but make sure Pins 55, 57, 58, and 4 require 4.7uF caps each - see datasheet typical application below.

Before you go into isolation, can you remove the HS/VS/DE 100 Ohm resistors that are in series with these pins. I don't see that as part of our recommendation in the datasheet.

Make sure your power-up sequence matches the datasheet recommendation.

C13 should be >10 uF to ensure proper PDB delay