• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 366 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

THVD4431: Is there a dedicated RS232 transceiver that meets these conditions?

Lee SungYun
Intellectual 365 points
Part Number: THVD4431
Other Parts Discussed in Thread: TRS3122E, THVD4421, SN75155, SN65C1168E, THVD1512, SN65LVDS049, THVD2410V

Hello?

I'm currently considering an RS232 only transceiver. The conditions I'm looking for are as follows.

- Receiver 2-channel, Transmitter 2-channel

- Control Logic Level: 1.8 V

- Power Level: 3.3 V ~ 5 V

- Output Swing Level: Selectable between ±5 V or ±9 V

- Maximum Data Rate: 1 Mbps

- The size of the circuit should be as small as possible

During the review, I saw the only product called THVD4431, but it seems to be large in size with too many channels and features. (40 pins with 5R3T+RS485)

If there is a product other than this that meets these conditions, please recommend it.

Thank you.

S.Y.Lee

  • 0
    Guru 297970 points

    The THVD4431 does not have a selectable output swing; its RS-232 outputs use charge pumps and just double the supply voltage.

    ±5.5 V are valid RS-232 voltages. What problem are you trying to solve with ±9 V?

  • +1
    TI__Mastermind 48377 points

    Hi S. Y. Lee,

    We don't have any device that meets those characteristics listed. 

    The closest device would be the TRS3122E which has the following parameters:

    - 2 Transmitters and 2 Receivers

    - Control logic voltage - 1.65V to 5V (through VL pin)

    - Main Power Supply - 1.65V to 5V (through VCC pin)

    -Output swing voltage -  +/-5.4V (typical) when VCC = 3.3V to 5.5V

    -Maximum Data-rate: at least 1Mbps  @ 500pF loading

    -24 pin RGE package - 4mm x 4mm footprint. 

    The only parameter it doesn't meet is the selectable swing voltage. None of our devices have a selectable swing voltage - and the device that is closest to all your needs is +/-5V (minimum). 

    Why do need a selectable swing voltage - this is an extremely uncommon need (RS-232 valid voltage levels are +/-5V to +/-15V - so both +/-5 and +/-9V are valid RS-232 voltage levels and would be rated to be placed into the same exact systems). So I am wondering why this is required because I have never seen this as a need in an RS-232 system and the vast majority of RS-232 transceivers, TI or otherwise, also do not give the option of changing swing voltage while also granting a lot of modern features that you are needing. I am just bringing this up because you may be very hard pressed to find a device that meets your system's needs if that is a hard requirement so I'd double check why that is necessary. 

    Best,

    Parker Dodson 

  • 0 in reply to Clemens Ladisch
    Intellectual 365 points

    To put it briefly, this is an examination of whether it is possible to create a product that can respond at high speeds & low voltage in a weak noise environment and at low speeds & high voltage in a strong noise environment.

  • 0 in reply to Parker Dodson
    Intellectual 365 points

    Thank you for the detailed review.

    I understand that the reason the usable range of RS-232 is ±15 V is for a communication line that is resistant to noise.

    So, in my opinion, is it possible to create a product that can respond to ±5.5 V (VCC=3.3V) of 19200~ for short distances (weak noise env.) and ±9.0 V (VCC=5.0V) of 9600 for long distances (strong noise env.).

    THVD4421 (currently only EVM is confirmed) seems to match the content except for multi-protocol, but I couldn't find anything other than what I expected, so I asked.

    S.Y. Lee

  • 0 in reply to Lee SungYun
    Guru 297970 points

    There are RS-232 transceivers that need external high-voltage supplies, e.g., SN75155.

    But in practice, high-noise environments are much easier handled with differential signals like LVDS (e.g., SN65LVDS049) or RS-422/RS-485 (e.g., SN65C1168E or 2× THVD1512).

  • 0 in reply to Clemens Ladisch
    Intellectual 365 points

    I reviewed the SN75155 you mentioned, but it is not small either. I will also reexamine the overall plan.

    Thank you.

  • 0 in reply to Lee SungYun
    TI__Mastermind 48377 points

    Hi S.Y Lee,

    As Clemens alluded to - older devices may support multiple supply voltage cases like the SN75155 but they are larger in size and generally have less than ideal performance in many applications. 

    I will say the THVD4421 - which is only available in the EVM at the moment - is going to have the same issue as our THVD4431 and most of our RS-232 transceivers because you cannot change the swing output voltages for the RS-232 - they are going to be the same as listed in the THVD4431. The THVD4421 is basically the same device as the THVD4431 but with only a 2T2R rs-232 configuration instead of 3T5R.  So that won't really work either - the TRS device I listed in my first reply is still the best option that hits almost all of the RS-232 requirements you have. 

    If you are very concerned about noise I agree with Clemens - RS-485 and RS-422 are most likely better because your speed is only 1Mbps (LVDS is made for very fast communication - like 100s of Mbps - so I don't think that is necessary here). They are differential devices which inherently are more resistant to noise than RS-232 devices are. The two devices Clemens mentioned are okay but they would need a level shifter to be able to use them with a 1.8V controller. However a solution that uses 2x THVD2410V (which is offered in a 3mm x 3mm VSON package - so it is relatively small on the board) would also most likely be a good solution as it allows for a separate 1.8V supply so that it can interface with 1.8V logic - it would include a few other features but ultimately it may be the simplest solution using RS-485. In general if noise is the concern and you have flexibility on what standard can be used I'd highly advise RS-422 or RS-485 - all of our devices that wouldn't require a additional level translator are going to be RS-485 and RS-422 (RS-485 can be thought of as basically a more robust version of RS-422). 

    Please let me know if you have any other questions and I will see what I can do!

    Best,

    Parker Dodson