This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TFP401 HDMI splitting problems

Other Parts Discussed in Thread: TFP401

Hi,

My project consists of 6 Adafruit HDMI 7" backpacks all using the TFP401 decoder.
www.adafruit.com/.../2406

I need all of them to mirror my desktop Mac Mini for video playback.

However, my problem lies in that I cannot get any picture through my HDMI splitter hub.

Are the TFP401's not capable to being split? Could the issue be the splitter?


All screens successfully display my desktop when plugged directly to the hdmi of my computer, so I know they work.

Please, some guidance is really appreciated. This is for my highschool project and I'm getting really close to the deadline.

  • Hi Rick,

    Are you sure the splitter is working? Did you tested it using a standard monitor? How your backpack is powered?

    ---
    regards,
    Igor
  • Thank you for the reply!

    The splitter does seem to work. When I hook it up to my flat screen TV, the computer desktop shows when running through the splitter.

    The backpacks are powered by their own 5v supply to an outlet.

  • Rick,

    It seems the problem is caused by the EDID data missing.

    According to the standard, every HDMI interface shall implement at least three physical channels:

    1. TMDS channel comprising of four differential pairs for transferring video data from source to sink;
    2. DDC channel (essentially an I2C bus) for transferring the sink's identification information (at least the EDID data) in the opposite direction (i.e. from sink to source);
    3. HPD channel (actually a single wire) that allows for source to sense the sink's presence.

    Now look at the backpack's PCB using the link in the original post. Fortunately the image is of enough resolution.
    The only HDMI-related part in the design is "the TFP401 decoder". Check the data sheet for that part. The only channel it does implement is the TMDS channel. Note that TFP401 does not implement neither DDC nor HPD channel. So it's the designer's responsibility to implement that channels using an additional parts.

    Now let's try to locate that additional parts on the PCB.

    Look at the HDMI connector. We can see that the pin 19 (HPD line) is wired to a resistor and that seems OK provided this is an 1 KOhm resistor connected to +5V rail.

    As for the DDC channel, there is nothing on the PCB that can suggest the implementation. Instead, there are five copper pads on the upper left corner. The pads of interest are marked with "SDA", "SCL", "GND" and "+5Vsomething".

    So the message from Lady Ada is that the DDC channel implementation is left to the user. After all it's a DIY stuff and on that market the price figure is much more important than the standard compliance.

    Actually the backpack is a sub-standard HDMI product and when it has connected it to the standard HDMI source, it's up to the source's vendor how to handle the issue (EDID reading failure). It seems while your computer just falls back to some default resolution mode, your splitter is more restrictive in that regard and refuses to output video data as it can't determine what video modes are supported by the display.

    Again, this is a DIY product so what you can do to fix the problem is to buy an I2C EEPROM part, program it with valid EDID data block describing the video modes the backpack supports, and connect the part with the mentioned pads on the PCB using a short wires. Alternatively, you can visit nearest electronic scrap yard in your area, find an old broken monitor (even that of the CRT type will be OK), locate and de-solder the EDID EEPROM part from it and try to fix the backpack using that part.

    Perhaps you can get more advises about EEPROM programming on the maker's forums. As for the TFP401 part, in your case it does exactly what is designed to do and does it well as you're not complaining on the picture quality.

    Finally, be careful when connecting the backpack to any more-less expensive equipment. The backpack does not have ESD protection device on its HDMI connector so the chances are a damage has occurred to the HDMI port of the equipment because of static discharge event.

    ---
    regards,
    Igor
  • Well, more close inspection shows that the EDID EEPROM is actually populated. It's the tiny SOT-23-5 part located just above the TI chip. And the pads are provided for programming it. Anyway it seems the splitter is not happy with the EDID block contents.
    ---
    regards,
    Igor
  • Hi Igor,


    Thank you for all the detail. I have reached out to the makers of the backpack and they have been unresponsive.

    since you mentioned the edid, I did come across this post, but am unsure as to how to make it work with my particular situation.

    From their forum ---

    awesome, we can burn the EEPROM with an Arduino.
    Go to this tutorial
    https://learn.adafruit.com/adafruit-tfp ... g-the-edid
    and wire up the Arduino to the backpack, there's some square pads next to HDMI port. You'll need to solder wires to them
    Connect 5V from Backpack to 5V on Arduino UNO
    Ground from Backpack to GND on Arduino UNO
    SCL from Backpack to A5 on Arduino UNO
    SDA from Backpack to A4 on Arduino UNO
    Then run that sketch from the tutorial to burn in the 800x480 EDID. You'll have to open the Arduino Serial console and send a character to get it started. Let me know what the output is (it should be "Verified!") and then it will work on any computer.

    We've gone thru all our stock here and fixed any that did not have the EDID programmed so thank you for helping us figure it out :)

    ---

    I do have an arduino and soldering equipment. Alternatively, my professor said that I can run two screens directly to a Mac Mini and he can let me borrow 2 more Mac Minis to get my 6 screens powered. Any thoughts?