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.

Problem Using a TVP5150AM1 in PAL Mode

Other Parts Discussed in Thread: TVP5150AM1, TVP5146, TVP5151, TVP5150

Hi, Everybody.

 

I faced problems using a TVP5150AM1. 

I am developing a car AVN with PLDS Loader + TCC8900 CPU(Telechips).

In NTSC mode, there is a little bit flicker noise also. But in PAL mode, there are lots of flicker and 'Rainbow' effect(Please, forgive me. i don't know the exact term instead of Rainbow effect.)

 

DVD Title what i tested is PAL format. 

I took photos in NTSC and PAL mode as belows.

 

 

 

1'st Picture is one In NTSC and 2'nd is in PAL Mode Setting. Since above pictures are pause images, it is difficult to see flicker noise.

Same problem is occurred even i use a TVP5146 +  DVD CVBS configuration.

 

For reference, TVP5150AM1 PAL Setting Values are;

Register Value

03h 29h // enable Auto switch signal for SECAM

04h DCh   // enable Auto switch signal for SECAM

07h 40h   // insert 656 ancillary header for raw data

11h 01h   // AVID start pixel MSB, (01h in NTSC)

12h 04h   // AVID start pixel LSB,  (00h in NTSC)

13h 01h   // AVID stop pixel MSB, (01h in NTSC)

14h 00h   // AVID stop pixel LSB, (00h in NTSC)

 

I have tried to figure out what makes different between NTSC and PAL Setting during but failed. 

 

If you have any suggestion or advice, please don't hesitate to reply me!

 

Regards,

Sam

 

  • Hello Sam,

    Do you use our latest patch v048CAA for TVP5150AM1?  You can download it at:

    http://dleinspect.itg.ti.com:82/dsps/dsps_public_sw/dsps_swops_houston/ANALOG_VIDEO/Analog_Video_Decoder_Versions.htm

    Please let me know if you still have problem after patch.

    Regards,

    Viet

  • Dear Viet,

    Thank you for your reply.

    I applied V048CAA patch but still  have same problem. 

     

    After using patch, check ID results are;

    ROM major version 82h -> 04h

    ROM minor version  83h -> 8ch

     

    Do you have another recommendable try?

     

    Regrads,

    Sam

  • Hello Sam,

    There is always more flicker in Pal than NTSC since Pal refreshing rate is at 50Hz while NTSC is 60Hz.

    For the false color, can you read the register 8Ch to make sure Pal is detected (bit 3:1 should be 001b)?  If not, can you force it to Pal by writing 04h to to I2C register 28h?

    Have you try with PAL input from standard generator?  The false color also might caused by non-standard input.

    Regards,

    Viet

  • Dear Viet,

     

    The register 8Ch is 83h(Autoswitch mode and (B,G,H,I,N)PAL ITU-R BT.601).

     

    By the way, the Status register #1(address 88h) value is read 6Eh or EEh(Peak white is detected or not).

    Especially the picture such as posted in my previous question, EEh is read more frequently than 6Eh. 

     

    Do you have any possible clue?

     

    Regards,

    Sam

     

     

  • Hi Sam,

    That is a very good clue.  Look like your input signal peak to peak is out of range some time.  Can you disable peak detection to see if it help (set to1 to both bit 3 and 1 of I2C register 02h)?  Do do have anti-alias filter at the input like the data sheet suggested (see figure 6-1 of the TVP5150AM1 data sheet)?

    Regards,

    Viet

  • Dear Viet,

     

    After changing the peak detect option, Status register #1(88h) was read 6Eh always. But i have a still problem.

    It seems that the peak detect option is just enable or disable to detect a peak presence.

     

    In Color Bar pattern, Input video signal at No.1 pin of TVP5150AM1 was not over 0.75Vpp.

    And i did insert AAF(Anti-Aliasing Filter) part to my circuit using jumping wire because my current hardware doesn't have a AAF stage.

    Maybe wire length is long, AAF doesn't effect any improvement.

     

    Viet, what's the next step to struggle this problem.

    I will look forward your advice.

     

    Regards,

    (Almost dying) Sam 

     

     

  • Hi Sam,

    You said there is still problem, could you tell me which problem do you have now?  Is it flicker or false color?  The flicker is caused by 50Hz refreshing rate in PAL so we could not do anything about it.  For the false color, can you try to adjust the saturation and hue registers to see if it helps.  Does the problem happens only to this input signal?  Can you try it with color bar from video signal generator?

    Regards,

    Viet

  • Dear, Viet,

     

    The Problems are both flicker and rainbow noise(false color?).

    The flicker what i mentioned is different than a normal flicker caused by refreshing rate. 

    Due to my poor English, i can't explain problem detail. So, i attach some files for explaining my problem state.

     

    SG_Color Bar is taken from a signal generator. 

    DVD_PAL-1 and -2 are taken from a composite of PLDS Loader.

    Both Case, input video signal level was about 0.69Vpp.

     

    In NTSC mode,  i have same problem but much less noise than PAL mode.

    As i mentioned at my 1st question, there was a same problem in TVP5146 application using a composite signal of a loader. It was ok using RGB signal.

     

    Please see the attached files and advice me.

     

    Regards,

    (near to death) Sam

    Click here to play this video

    Click here to play this video

    Click here to play this video

  • Hello Sam,

    We are very sorry about this problem.  Is there any way you can send us your schematic for review?

    I cannot download the avi files yet.  Will get back with you after see the video.

    Regards,

    Viet

  • We need to you IE to watch the video.  Firefox does not come up for these attachment.

    Regards,

    Viet

  • Hi Sam,

    Can you try different setting for peaking gain (bit 3:2 of register 08h and Luminance trap filter select (bit 1:0 of register 0Eh) to see if it helps?

    Please send us your schematic and the crystal you are using for review if possible.

    Regards,

    Viet

  • Hello Viet,

     

    My current browser is a google chrome and i experienced same problem not to see my attached AVI files in TI web.

    However, it is possible to see AVI files as i did right click on the file box area and selected 'save link as different name..' in pop-up menu.

     

    I want to know your e-mail address to send my schematic and crystal specification for your review.

     

    Regards,

    Sam

  • Sam,

    Are you using the correct load capacitors for the cyrstal that you are using?  The cyrstal needs  to also have <= 50ppm stability.

    Do you have a TVP5150AM1 or TVP5146 EVM that you can compare to your board?

    How do you display the TVP output?  How is the ouput being processed and displayed on the monitor?

    Regards,

    Larry

     

  • Dear Larry,

     

    In my thought, my circuit has a correct capacitor value. Two 27pF are adopted in current. 33pF was tested also, but still same problem occurred. 

    Using 27pF is more close to 14.31818MHz than using 33pF when i measured by a frequency counter.

    I attached the applied crystal specification.

    6114.FMS(DSM4-14.31818M-18pF).pdf

     

    Until now, i haven't obtained a TVP5150AM1 EVM. I did request the EVM to my supplier but i heard that it is difficult to find the one because a TVP5150AM1 is not a brand new IC.

    So i am looking for a TVP5151 EVM instead of a TVP5150AM1.

     

    The TVP5150 output is a BT656(YCbCr) with embedded sync and connected with a CPU(TCC8900). 

    I don't know exactly about processing inside a CPU but CPU's output is LVDS signal to interface with a LCD display.

     

    Larry,

    If you have another questions or need other information, please reply to me.

    Regards,

    Sam

     

     

     

  • Hi Sam,

    I can play the AVI files you attached last week.  Have you got a chance to try the filters I suggested last week?  My email is vdinh@ti.com if you send your schematic.

    Regards,

    Viet

  • Sam,

    The crystal should be OK. 

    I looked over your schematic.  It appears that you are using a single ground plane.  Is the analog input circuitry and crystal circuit well isloated froom the digital circuitry on you board?

    We recommend using a divide by 2 attenator with the TVP5150AM1.  R75 (75 ohms) could be replaced with two 37.4 ohm resistor for atteneation.  This does not explain why you see the same problem with the TVP5146, however.  The TVP5146 does not need attenuation.

    We recommend a 0.1uF cap for C428 instead of 0.47uF.  May not be the problem, however.

    The NJM2725 appears to be rated for 1V p-p input.  A composite video signal could actually exceed this with the color subcarrier excursions.  I am not sure how much distortion could occur with the NJM2575, if this occurs.  You could try attenuating the NJM input some to see if it has any effect, or try bypassing it.

    Your TVP5146 board has the same issue?

    Regards,

    Larry