• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 46 subscribers
  • Views 1649 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Replacement of TVP5150 by TVP5151 - issues

Claus Maehringer
Prodigy 100 points
Other Parts Discussed in Thread: TVP5151, TVP5150

Dear all,

currently I am trying to replace a design using the TVP5150PBS (TQFP 32) by an equivalent TVP5151,
as the TVP5150 was set to NRND recently. The datasheet of the TVP5151 states on page 25:

"TVP5151 is pin compatible with TVP5150/A/AM1, and the following differences should be considered
when an upgrade is planned.
• IO_DVDD supply can be any voltage from 1.8 V to 3.3 V.
• AVID/CLK_IN is input during RESETB. If this input is used as the clock source, XTAL1/OSC pin must
be grounded."

So, regarding Hardware the case is quite clear (beside the beforementioned points the clock frequency
on the TVP5151 is now 27MHz instead of 14.318MHz), but there are some small differences in the register set:
Especially the contrast control reg. (0x0C) on the TVP5151 is marked to be "reserved".
Our application software writes contrast values settings from the user to this register.
The circuit worked so far, also varying the contrast is functional, but according to the spec
this shouldn't be done.

My questions:
1. Is it safe to still write values to this register, although it is marked as "reserved"?
2. What is the risk if I do so, what could happen?
3. Any other registers to better keep untouched on the TCP5151 that were used on the TVP5150?
4. How is contrast control be done without having a distinct register on the TVP5151?
5. Are there any other known issues, when replacing an TVP5150 with an 5151?

Thanx in advance
Bye
Claus

  • 0
    TI__Mastermind 43875 points

    Claus,

    Are you using patch code with the TVP5151?

    Please make sure you are uploading the latest patch code (available here... http://software-dl.ti.com/dsps/dsps_public_sw/dsps_swops_houston/ANALOG_VIDEO/Analog_Video_Decoder_Versions.htm)

    I am not aware of any other issues in switching to the TVP5151, as long as patch code is applied.

    BR,

    Steve

  • 0 in reply to Steve Clynes
    Prodigy 100 points

    Dear Steve,

    no, the TVP5150 was used without any patch code. We just replaced the "old" TVP5150 by an 5151, and changed the oscillator. The output signal of the TVP is handled by an FPGA, which build the interface to a TFT display. So some slight modifications on the FPGA 'code' were necessary. The TVP itself is controlled by a separate uController via I2C. We would like to avoid any software changes on this uC (-> avoid requalification), thus I came up with the question what happens, if register 0x0C (= contrast control on the TVP5150) is also written on the TVP5151.

    Again: Is it safe, to write to register 0x0C on the TVP5151 like it was possible to do so on the TVP5150 to adjust contrast - even without the patch code?

    The point is: It works, we can adjust the contrast on the TVP5151 by writing to register 0x0C. But why is it remarked as "reserved"? What could happen if it is written with values? Are there any values that must be avoided in any case (e.g. 0x00 or 0xFF)?

    Does the patch code provide a kind of "compatibilty mode" of the TVP5151 to act like the TVP5150?

    Best regards

    Claus

  • 0 in reply to Claus Maehringer
    TI__Mastermind 43875 points

    Claus,

    What did you need to change on your FPGA? In theory nothing on the output of the TVP should have changed between the TVP5150 and the TVP5151.

    It is always recommended to use the latest patch code for the specific device you are using.

    I am not sure why the contrast register is reserved in the TVP5151 datasheet since as you have found, it appears to be there. Additionally, the patch code documents for the TVP5151 reference the contrast control. If the TVP5151 patch notes it does say that the contrast value is clamped, so I believe that if you are not using this patch code then you should perform clamping in your control software.

    Patch code does not guarantee compatibility. It is usually designed to add features or fix bugs.

    Whilst the TVP5150 and TVP5151 are derivatives they are also different devices running different software so it is impossible to say that they behave exactly the same and unfortunately, just like any other device replacement, I suggest that you do as much validation in your system to ensure that any differences that might be present are either ones that don't affect you, or if they do, that you can accomodate the differences.

    BR,

    Steve

  • 0 in reply to Steve Clynes
    Prodigy 100 points

    Dear Steve,

    thank you for your instant reply. There is still one question I have, that had been asked by my colleage in the past:

    [Pls see here:   http://e2e.ti.com/support/data_converters/videoconverters/f/376/t/94237.aspx ]

    Larry Taylor wanted to check, if there is a additional checksum 'option' to improve confidence in proper data transmission to patch RAM and startup of the patch code. But the discussion unfortunately stopped there.

    Maybe you could find out for me, if there is a checksum integrated in the patch code that is checked within the TVP5151 during data transmission or startup phase.

    This would greatly improve my confidence ;o).

    Thanx in advance!

    Claus

  • 0 in reply to Claus Maehringer
    TI__Mastermind 43875 points

    Claus,

    Please check the notes supplied with the patch code (available here... http://software-dl.ti.com/dsps/dsps_public_sw/dsps_swops_houston/ANALOG_VIDEO/Analog_Video_Decoder_Versions.htm)

    In the TVP5151 you should be able to read back the patch code using the read register 0x8E. Please check section 3.21.53 in the TVP5151 datasheet in conjunction with the application note referenced above.

    For the specific device you are concerned with please check both the datasheet (search for "patch" to see if there is a readback register exposed) and the corresponding device patch download application notes (included with the patch code)

    Beyond this there is no additional way to validate the download.

    Having said this though, as Larry mentioned, if the version reads back correctly then the internal checks on the firmware passed and code execution started.

    BR,

    Steve