DS90UH949-Q1EVM: HDMI not detected

Hi Madan,

I would like to confirm the configuration you are running. From the screenshot provided, it looks like the link is detecting a UH 926 deserializer, could you check and confirm the link that is established between the SER and the DES?

Reading from these registers, "0x005F" returns "0x00", which indicates that the HDMI receiver is not detecting a valid signal where it should be returning the frequency. 

Would you be able to confirm these two:

1. The link established between the SER and DES shows no fluctuations or loss of lock, and that the received DES is confirmed to be 948 in ALP?

2. Ensure that the following Power-Up sequence is followed from the datasheet recommendations: VDD18, then VTERM, then VDD11, activate PDB then apply HDMI input.

Best,

Miguel

Hi Miguel,

confirming below:

1. The link established between the SER and DES shows no fluctuations or loss of lock, and the received DES is confirmed to be 948 in ALP?

 - DES is 948 which is over the display, but ALP not detecting it.

2. Ensure that the following Power-Up sequence is followed from the datasheet recommendations: VDD18, then VTERM, then VDD11, activate PDB then apply HDMI input.

- Yes, we followed the recommended sequences, but still HDMI was not detected.

Also, tried below:

1. Force assert to HMDI 

2. Cross-check the jumper and switches.

3. ALP re-install, and cross-check whether the library present or not

ALP is getting stuck many times, it is the latest software release downloaded from TI website.

Let me know how we can proceed.

Thanks,

Madan

Hi Madan,

Thank you for the updates.

I would like to request additional information regarding the schematics and layout between the 949A and the 948 device, focusing on evaluating the link established between the Serializer and Deserializer. This may be a prerequisite for the HDMI to register on the 949A. Ultimately, we would like to have lock register between the devices.

If you prefer to share the schematic privately, you can email me at m-bolante@ti.com or accept my friend request on E2E to send over a private message.

Best,

Miguel

Hi Miguel,

Thank you for the reply.

From the schematics side, we looking to get from the hardware team.

There are a couple of observations we have:

1. The ALP tools not detecting the 949 correctly, we have manually added 949 devices, still show incorrect device profile:

2. It seems that ALP is a 32-bit tool, actually, we are using a 64-bit, Windows 10 system, does it work? if not, is there any alternative solution for it?

We are using the ALP 1.57 version, please confirm if any different version needs to be used. Also, referring to the user guide "DS90Ux949EVM User's Guide", "Literature Number: SNLU169 December 2014".

Let me know of any newer versions of documents and tools to be referred to.

I would like to add a few of my colleagues at BCC.

Thank You,

Madan

Hi Madan,

Madan Avhad said:

From the schematics side, we looking to get from the hardware team.

Once you get these, feel free to send me a direct message if you do not wish to share the schematics on this public forum.

Madan Avhad said:

The ALP tools not detecting the 949 correctly, we have manually added 949 devices, still show incorrect device profile:

Madan Avhad said:

It seems that ALP is a 32-bit tool, actually, we are using a 64-bit, Windows 10 system, does it work? if not, is there any alternative solution for it?

This version of ALP is sufficient for the x64 OS being used. Version 1.57 is also appropriate.

It seems that the ALP screenshot you have is no longer showing the 949 in the info tab, whereas the first screenshot did include it. Has there been any changes in your setup between these two submissions?

Aside from ALP 949 detection issues, I believe emphasis on getting lock established between the 949 and 948 would be a great start to prioritize. Please let me know if any other items or issues are preventing this link. 

Best,

Miguel

Hi Miguel,

I'm debugging a bit more, Can help me with the power sequence:
 - VDD18, then VTERM, then VDD11, and wait for settle then activate PDB.

Can you please point out me jumper number and the setting to be applied? and When?
And also how can we know that power is settled and ready PDB and HMDI input?

 Thanks,
Madan

Hi Madan,

Madan Avhad said:

I'm debugging a bit more, Can help me with the power sequence:
 - VDD18, then VTERM, then VDD11, and wait for settle then activate PDB.

These power-up requirements including the PDB pin are shown in the chip datasheet below:
DS90UB949-Q1 1080p HDMI to FPD-Link III Bridge Serializer datasheet (Rev. B) (ti.com)

This is section 9.1, power up requirements.

With the sequence you mentioned, it may be better to flip your sequence of VTERM and VDD18.

Madan Avhad said:

Can you please point out me jumper number and the setting to be applied?

If you are using the EVM, there should be an external voltage supply of 12V as a barrel jack, which would take care of the timing.

However, if you are powering the individual pins, the test point jumpers correspond to J35 for VDD11, J39 for VDD1V8, J42 for VDD33, and J40 for VTERM. This can all be found under the schematic shown in "A EVM PCB Schematics": DS90UH949-Q1EVM User's Guide (ti.com).

"Power is settled" refers to the steady state reached for each voltage level.

Best,

Miguel

Hi Miguel,

I have followed the mentioned power sequence, but neither EVM detects ALP tools nor any of the LEDs turn ON.

May I know how long needs to wait to settle power?

Regards,

Madan

Hello, 

Since today is a US public holiday, we will resume activity on this thread tomorrow. Thanks for your patience. 

Regards, 

Logan

Hi Madan,

The EVM should turn on immediately upon powering on, what is the result of using a 12-V barrel jack? If the EVM does not turn on correctly, there may be an issue with how the 949 is operating on the DC Voltages. Have you taken DC measurements from the test points to confirm that the voltages are correct?

Best,

Miguel

Hi Miguel,

We have a 12V DC barrel jack with a 1A current rating. 

With the default setting of jumper and switches, EVM gets ON and detects in ALP.

But with power sequences mentioned in the datasheet,  J35 for VDD11, J39 for VDD1V8, J42 for VDD33, and J40 for VTERM, EVM didn't detect in ALP, seems no power ON

Hi Madan,

Yes, continue with the default setting and the 12V DC barrel jack, this will follow the power sequencing for you and is the best option for power stability.

When the EVM detects in ALP, are you now seeing 949 EVM?

Best,

Miguel

Hi Miguel,

We continue on 12V DC barrel jack power and default setting, and the device gets detected.
Wondering the Revision shown is 9.

As per the user guide, it should show Revision :1
Is it correct? we using rev-B EVM.

Hi Madan,

The revision of the board may differ from what's shown in the datasheet, but I confirmed that the revision for the 949 EVM in-house is also Rev 1 (Rev K).

What you're seeing in the info tab may be confirmed with the following procedure:

The command to run in the scripting window under the scripting tab is:
"board.ReadI2C(0x18, 0x0D)", this should return 16 (0001 0000) if it is revision 1.

Where 0x18 is the SER IDx address, 0x0D is the register holding revision info, and the first 4 bits in register 0x0D represent the board revision.

Please let me know if ALP is able to detect link between SER and DES.

Best,

Miguel