• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 33 subscribers
  • Views 759 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.

TRS3238E does not seem to propagate the start bit in powerdown mode

Jim Sung
Prodigy 50 points
Other Parts Discussed in Thread: TRS3238E, MAX3238E, TRS3237E
As far as I can tell, when the TRS3238E goes into powerdown mode, it fails to propagate the start bit from RIN* to ROUT*:

      Tx      Rx        ROUT* (not including the stop bit)
    ----    ----        ----------------------------------
    0x00    0x80                                    136 us
    0x01    0xc0                                    122
    0x02    0x81                                    136

The ROUT*, at 57600 bps, should be 156 us, not including the stop bit.

In the "0x01" case, since we miss the start bit, and the LSB is "1" (RS-232 is LSB first), the whole thing is off by two bits.

Jim
  • 0
    TI__Guru**** 164796 points

    Hello Jim,

    The ROUT [DOUT] output lines should be high impedance when the device is in shutdown mode. Do you mean when it comes out of power down mode?
    What are preconditions for the power status during the start bit?

    Regards,
    Ron Michallick

     

     

  • 0 in reply to Ron Michallick
    Prodigy 50 points

    Hi Ron:

    > The ROUT output lines should be high impedance when the device is in
    > shutdown mode. Do you mean when it comes out of power down mode?  What

    Sorry, I wasn't too clear.  Yes, the TRS3238E seems to fail to propagate
    the start bit when it comes out of the powerdown mode.

    Here is my set-up:

         FORCEOFF_N strapped to HIGH
         FORCEON strapped to LOW

    so the "auto-powerdown plus" feature is enabled, and it is a paper-pin
    loopback on the DB9.

    What I see is if I wait > 30 s, I see DOUT goes to Z, and since I have a
    paper-pin loopback, RIN also goes to Z, and INVALID_N gets asserted.
    I'm not sure why you said ROUT should be Z because all RS-232 receivers
    should have an internal pull so that on an open input, it outputs HIGH
    to the UART, which is what I see.  So far, the TRS3238E behaves exactly
    as I expected.

    As far as I can tell from the data sheet, the only case ROUT* should be
    Z is if FORCEOFF_N were strapped to LOW.

    After 30 s, if I send a character, then I see the anomaly that I
    reported.  Subsequent characters, if < 30 s, do not have this problem.

    When the "auto-powerdown plus" feature is disabled (i.e., both
    FORCEOFF_N and FORCEON are HIGH), we don't see this anomaly.

    > are preconditions for the power status during the start bit?

    The V_{CC} is applied throughout this.

    Jim

  • 0 in reply to Jim Sung
    TI__Guru**** 164796 points

    Hello Jim,

    The start bit is missing because the charge pump that generates the +/-5.6V for the drivers takes a finite amount of time to power up. I captured waveforms using both 3.3V and 5V VCC.  It took up to 30uS for DOUT line to become active. 

    The top trace (CH4) is the input to DIN, the middle trace (CH2) is the ROUT pin, and the bottom trace (CH3) is the DOUT/RIN line.

    In cases where the device might be shutdown, set FORCEON high and wait about 50uS before sending data.

     

     VCC=5V 

     

    VCC=3.3V

    Regards,
    Ron Michallick

  • 0 in reply to Ron Michallick
    Prodigy 50 points

    Hi Ron:

    I understand the constraints that the charge pump is under.  However, the bottom line is that unless the baud rate is very low, one really cannot use the "auto-powerdown plus" feature.

    Jim

  • 0 in reply to Jim Sung
    TI__Guru**** 164796 points

    Jim,

    It is true that the feature requires caution. Becuase shutdown is done autonomously and without a status flag, the application software driver must delay start of transmission if shutdown is a possibility.

    There are alternatives to the standard shutdown plus mode.
    By tying "~Invalid" to "Forceon"., the part will stay powered unless all RIN lines go dead for roughly 30 seconds (and no DOUT change) and will power up when one comes live. Manual power up down can be achieved by controlling "ForceON" shorted to "ForceOFF".

    Is there a case where DOUT must be live when RIN is lines are dead?

    Regards,
    Ron Michallick

     

     

  • 0 in reply to Ron Michallick
    Prodigy 50 points

    Hi Ron:

    > There are alternatives to the standard shutdown plus mode.  By tying
    > "~Invalid" to "Forceon"., the part will stay powered unless all RIN
    > lines go dead for roughly 30 seconds (and no DOUT change) and will
    > power up when one comes live. Manual power up down can be achieved by
    > controlling "ForceON" shorted to "ForceOFF".
    >
    > Is there a case where DOUT must be live when RIN is lines are dead?

    Yes, we can transmit asynchronously and we have no control over the
    receive, so I don't think this is going to be an option.

    I assume the MAX3238E (either Maxim's or TI's) works the same way?

    Jim

  • 0 in reply to Jim Sung
    TI__Guru**** 164796 points

    Jim,

    Yes, the TI and Maxim MAX3238E work the same way.

    Regards,
    Ron Michallick

     

  • 0 in reply to Ron Michallick
    TI__Guru**** 164796 points

    Jim, 

    The drivers can be enabled by setting ForceOn high to prime the charge pump before transmission.
    However, pulling ForceOFF low at the end of transmission will disable the receivers per the data sheet.
    To keep the receivers live, ForceONcan be set low and ForceOFF can remain high; the device will shutdown later on it own if the RX lines are at zero volts.

    The TRS3237E has separate driver and receiver shutdown control, perhaps it is a better fit. 

    Ron