DS90UB964-Q1: DS90UB964 rear view camera analysis

Hi Raymond,

Thanks for the information.

1. Can you please send a register dump of the issue 964 and the normal 964 so I can compare them?
2. I noticed that you included a schematic, but it is too small for me to read closely. Can you send the 964 schematic again so I can check if there is anything wrong?

Regards,

Cindy

dear cindyt

1. there is register dump of the issue 964 and the normal 964 in original file which free back from customer engineer

2.attachment is about  964 SCH,please check，thanks

UB964.pdf

Hello, 

Since today is a US public holiday, we will continue support on your request on Monday. Thank you for your patience. 

Regards, 

Logan

Hi Cindy,

If we only set 0x1F=0x03 for 400Mbps, what is the problem on this? Some abnormal display issue?

We found for 964 chips with 400Mbps setting, most of chips work good, but for 2 out of them, the CSI rate will increase from 400Mbps to 1.2xGbps, cause black screen issue. Can you share why the issue will occur on particular 964 chips? 

Johnny

Hi Johnny,

We recommend setting the extra timing registers for 400Mbps even if it seems to work for other chips because this is how we have validated it to comply with the MIPI specifications. The timing parameters are required in order to use 400Mbps, so not following these steps could cause CSI-2 video timing issues at that rate. All I can share is that datasheet requirements must be followed in order to get intended performance across devices.

Regards,

Cindy

Hi cindy

attachment is about Rear view camera black screen analysis  and 964 SCH from customer xiaopeng,update lastest test condition，please help to check，thanks

Rear view camera black screen analysis(2).docx

0601.UB964.pdf

Hi Raymond,

Here are my comments on the schematic: 

• For the PDB pin in the schematic, there should be a 10kOhm pull-up resistor to 1.8V and a >10uF capacitor to ground (see Typical Application diagram in 964 datasheet). This is important in order to ensure that the power-up sequence is followed. If you are relying on the external RC circuit to drive PDB, you will need to fix this in the schematic.
• There is no ferrite bead recommended in the datasheet for the VDDIO pin. I see the document says there is no abnormal power supply, but please confirm that VDDIO is within spec since this pin does not exactly follow the datasheet. 

I see for the power-up sequence waveform you just show the power up of 1.1V and 1.8V. Can you also probe VDDIO and PDB and make sure that the entire power-up sequence is followed in Section 10.2?

Comparing the good.txt and bad.txt register dump, the only difference is with the FPD-Link packet frequency in registers 0x4F and 0x50. These registers read back the effective FPD3_PCLK frequency which depends on the operating mode (see note 2 in UB964 datasheet on page 10). Refer to this E2E for more information on how it is calculated. Can you confirm that the PCLK on the serializer is stable and meeting datasheet requirements? 

Regards,

Cindy

1.  PDB on board is controlled by software, the time of raising PDB is far behind the time of power supply stabilization (more than 200ms) , which meets the requirement of manual completely

2. VDDIO series magnetic beads are good for filtering high frequency harmonics, to improve the quality of power supply, the actual test did not meet the requirements of the manual. The whole circuit design is evaluated and approved by TI.

3. Make sure that the FTP link is OK, that ABA cross-validation has been done, and that the configuration is OK.

any advice？thanks in advance

Hi Raymond,

Thank you for the clarification on the schematic.

I noticed in the register configs that you have continuous clock mode turned off in register 0x33. For debugging purposes, can you try enabling continuous clock mode and then measure the CSI clock output on the scope to see if it is still drifting in frequency? 

For the register dump of the issue unit (bad.txt), what do you mean by "the failed machine reads at the time of replay?" Were these registers read when the CSI clock reached the high frequency and the screen went black?

Are these units tested over temperature, or were these measurements taken at room temp?  

Regards,

Cindy

Closing this thread now that it is being continued on another E2E thread: https://e2e.ti.com/support/interface-group/interface/f/interface-forum/1226011/ds90ub964-q1-ds90ub964-rear-view-camera-analysis-continue-2