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.

SN65DP159: How to get an output (no hpd snk, no ddc, iic config mode)

Part Number: SN65DP159
Other Parts Discussed in Thread: TPD4E05U06, TMDS181

Tool/software:

Hi, so we are using the DP159 in a hdmi rx path.

Currently the hpd_snk has been left unconnected, same with the ddc lines both in and out.

After power up, we hold oe low for the required time and then release it. After that we use iic to set the registers so that...

DDC_TRAIN_SET has training disable;

TMDS clock is 1/10 (the hdmi tx device sending the signal is set to hdmi 1.4);

HDMI_SEL is HDMI;

HPD_AUTO_PWRDWN_DISABLE has don't power down based on hpd sink;

PD_EN has normal working;

TX_TERM_CTL is No termination (but i assume anyway the device would sense this and correct the register anyway??)

It appears that input video goes to the device, but we see nothing on the outputs (they are held high by the fpga they are connected to, but are constant)

Any ideas on what could be going wrong? Any settings we should try fiddling with.

I read on another forum question that is was ok for ddc to not be connected.

Is it ok for HPD sink to not be connected? In this case after oe, i guess we may go to the power down state, but after configuring the register to ignore HPD, will it go back to power up?

  • Hi,

    Please see Table 14 as shown below, the last two conditions would put DP159 into normal operation.

    For the input, are you AC coupling or DC coupling the DP159 input? Is the source a HDMI (DC coupled) or a DP++ (AC coupled) source?

    Thanks

    David

  • So just to be totally clear, we do have to tie the HPD sink high, or else have it connected to something that can bring it high (e.g. real hpd) ?

    The input is a HDMI source, but the lines pass through a TPD4E05U06 and then through some 100nF capacitors before getting to the 159.

  • Is it ok to leave HPD src unconnected? I notice it is in the table as well.

  • Hi,

    HPD_SRC can be left unconnected if not used. But you may need an external level shifter if FPGA can't drive the 5V HPD.

    What is being connected to the DP159 input? Is it an HDMI receptacle? 

    Thanks

    David

  • As i said we havn't got anything at all connected to HPD sink on the 159, this is what i am thinking of requesting a mod for. The HPD on the hdmi line goes through some circuits to fpga and is working fine, but the design just doesn't include connecting this through the 159. 

  • So some hand mod has been done to connect the hpd sink to the real one in the same way as is working on the tx side, but alas the rx side is still not working - output is flat but input has signal. any other ideas?

  • Hi,

    What is being at connected at the DP159 output? Is it a HDMI receptacle with AC coupling capacitor? Any chance you can share your schematic?

    I did verify on the DP159 EVM that even HPD_SNK floating, as long as register bit 3 is set to 0 and bit 2 set to 1, the DP159 will still be in normal operation.

    Thanks

    David

  • Hi, sorry i thought i had already replied here. We did now get the modification for the HPD_SNK, but it is still not working. At the output the original was a capacitor and then fpga which includes a pull up. We realised the resistor on the other side of the capacitor was missing so currently there is a mod to just bypass the capacitor so it is no-longer ac coupled. 

  • Hi,

    DP159 requires its input to be AC coupled, you can not removed the AC coupling capacitors at its input.

    For DC-coupled input, you have two possible solutions.

    • Use TMDS181 instead DP159 which supports DC-coupled input
    • Keep DP159, but add 50ohm termination network to 3.3V between the HDMI receptacle and the AC coupling capacitors. Please note this will create a double termination network and cut the input signal amplitude in half. So you may need to compensate for the drop in the signal amplitude.

    Thanks

    David