TFP401A: Garbage Display or Distorted Image appeared during Boot up (2-3secs)

Hi,

When do you drive TFP401A PD# and PDO# signal high? Do you drive them high before you are driving the HPD high?

What is the overall system power up sequence?

Thanks

David 

Hi,

As per checking 3 signal drive as the same time, refer the attached HPD Signal circuit, then the PDO and PD Signal put up to 3.3V, the Power up sequence is PD and PDO are pull high when power up but HPD is pull low when you power up

Thanks. Please advise.

Hi,

The SCDT output, which indicates link-disabled or link-inactive, can be tied directly to the PDO# input to disable the output drivers when the link is inactive or when the cable is disconnected. Is there a way you can modify the existing design to connect the SDCT output to the PDO# input?

Thanks

David

Hi David,

For SCDT we design to see status of IC not connect to PDO and our design is not using HDMI cable to connect, the connection are on PCB.

it is possible to add below person in this loop/discussion to have smooth discussion with our D&D Person In charge

Nopporn Kaewkad  Email: k.nopporn@electrofont.com.sg

Thanks 

We will try to make 1 sample to modify the circuit to connect the SDCT output to the PDO# input, to simulate if this issue will improve.

HI David,

We tried modifying the circuit from the NG units to connect SDCT output to the PDO Input, and the result was OK. But still, we want to understand more details in this issue see below inquiry

in our current design (SDCT output not connected to PDO input) we only see this problem in 30% of the units (and followed IC if do SWAP ABA), why other ICs (70%) didn't show this symptom in our current design?

Thanks

Marlon

Marlon

Do we have a way to probe the HDMI input between a working and a non-working board? 

With PDO always enabled by pulling it up to 3.3V, the TFP401A will send out the data at its output as soon as it sees data at its input even if the data may not be valid, and this is why you see garbage display.

But with SDCT tied to PDO, the TFP401/401A monitors the state of DE to determine link activity. In this case, the link is only considered to be active when DE is actively switching.

Thanks

David 

Hi David

Just to clarify, do you mean that are we able to probe and check the HDMI input signal for the good and NG units for comparison? using oscilloscope?

Thanks
Marlon

Marlon

Yes, I want to see if this is possible to differentiate between the good and the bad behavior.

Thanks

David

Hi David,

Noted, we will provide this information to you, give us some time to do the checking.

Regards

Marlon

Hi David,

Refer to the measurement result using the oscilloscope. Based on the data, both OK and NG units have the same input signal, so we suspect that something is different between the IC of the good and NG units. We also confirmed that the swapping (ABA) the unit with NG display always follows the IC. Kindly advise if we are able to proceed with FAE.

Regards
Marlon

Hi Marlon,

It seems that the waveforms going into the good unit and NG unit are almost the same. If we see that the issue does follow the NG part, the I recommend commencing the QEM process with your local CQE & FAE.