Is there any way for the development board to be configured so that it will simulate that there is an HDMI audio port? I purchased the eval board and found out once it arrived that the audio output section is not populated on this eval board.
The EVAL_ADV7612-7511 Advantiv™ EVAL-ADV7612-7511 Video Evaluation Board has audio port headers. The ADV7612 and ADV7611 are the same die, just different packages.
Do i have to have the AD1938 on the EVAL-ADV7611EB2Z to be able to get the hdmi source to be able to see an audio channel it can send data to? I am trying to figure out what I can do with this eval board now that i have it. I was expecting it to be fully populated per the development kit bill of materials.
The ADV7611 is a receiver. It will only extract audio from the incoming video stream. You can see one I2S channel on J5. The board should come with the AD1938 installed as shown in the picture in the user guide.
What exactly are you trying to do?
The board that I received looks like this
The audio section is not populated.
What I am trying to do is to use the ADV7611 and audio chip to verify that we have all the data lines connected between our imx6 processor’s hdmi outputs and the ADV7611 on the development board (and on our test board in the future). I was planning on using the sound output to verify that all of the signals are making it to the ADV7611 by using the audio data output since the sound signals are split on the TMDS 0,1,and 2 lines. So if we are able to get the desired audio out then we can verify that the TMDS lines are connected properly.
Normally you would use a monitor, but we are not wanting any user interaction with our test.
Yes, the AD1939 is missing.
This board only has one I2S pin and the audio carried on this pin only goes through one TMDS channel, not all three. So you can not verify all 3 TMDS data channels by just checking one audio pin.
Assuming you want to verify TMDS connections without human interactions, then I'd do the following
1) Have the FPGA output RGB or YCbCr in 444 mode with an image of large vertical black and white bars.
2) Set the ADV7611 OP_FORMAT SEL to 0x40, SDR 4:4:4
3) monitor bits R7, G7 and B7 for a waveform the matches the input image bar pattern.
Another quick way to check it to look at the TMDS lines with a scope. The lines will be about 3Vdc plus have a +-300mV signal on them. The clock lines will have the pixel clock while the data lines will have 10 times the clock rate. This is easy to see with low pixel rate formats and 200 MHz scopes.
Trying to do this without a scope. I have a processor that has analog and digital inputs.
If I populate the AD1938 per the bom will I need to reprogram the board? If so what tools will I need?
The AD1938 only gets one I2S channel which only comes from one TMDS data channel. You won't be able to detect the other two data channels. Doesn't meet your requirements
As far as repopulating and reprogramming, don't know state the software is on that board, especially since it appears someone removed the part.
To verify all 3 data TMDS channels you must look at the video output port. Only that will indicate activity on all 3 data channels.
Due to the protocol used to deliver the HDMI audio over the TMDS lines wouldn’t you need all of the data lines to be able to get a valid packet transmitted?
If the AD1938 only supports one channel then how is it getting 4 stereo channels output on the development board?
Timing information is carried on channel 0. Audio packets are distributed over the other channels. LRCLK and MCLK are derived from infoframe packets. These packets are distributed over the there channels. So even if channel 0 was getting through and channel 2 was shorted, you still might get audio out, just that it's noise.
Now this brings up another possible way to check data channels. Read the avi infoframe and see if it makes sense, i.e., the checksum is correct.
Correct… we would put a sine wave out and evaluate it to make sure it was correct.
What byte would I look at for the checksum in the infoframe?
Who would I need to contact to find out if adding the parts will get the audio section working?
ADI HDMI Repeater Application Ver R1.55.2
PLATFORM: ATV 7611/2 EVAL Rev 0xff
HDMI-RX: ADV7611 Rev 0x100
HDMI-TX: ADV7511 Rev 0x14
Created: Jun 11 2012 At 11:16:23
The fourth byte is the check sum byte. Need to check out the HDMI spec for the full packet definition.
I didn't realize you had the ability to verify a sine wave is really a sine wave, I thought you were just looking for activity.
Follow the evalnote I attached above, If the software doesn't work then there's an email address you can get the latest software from. Now a note, this particular is not well supported anymore, this is why I suggested the EVAL-ADV7612-7511. If is our current eval platform for the ADV7612/11 an has better support. Besides it's cheaper and can do everything you've asked for above except it doesn't have and audio output jack for a headphone.
Will be running the audio back into a line-in on another chip. So will record the audio and analyze it.
So the entire audio packet will be in the infoframe bytes 0x1c through 0x29? And the checksum is in the fourth byte? Which bytes are these PB0..PB14? Shouldn’t there be 27 bytes?
Would have been nice to find out that the other kit would have been better… but we purchased this one already and have it in hand…. Now get to wait on samples to fix this one.
Looks like the audio section drivers for the eval board have been removed from the source code for the development board. Is there any way to get archived copies of the code that I can use to get the audio jack output on the board working?
I don't think we support analog audio output on any of our evaluation boards for ADV7611/ADV7612.
Retrieving data ...