Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 11 replies
  • Subscribers 33 subscribers
  • Views 490 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.

MAX3232: D2OUT Does Not Transmit

Michael Boren
Prodigy 40 points
Part Number: MAX3232

I use the MAX3232 in several products as an interface between the onboard MCU and an external RS-232 bus. Recently I have received reports from customers that the product stopped communicating. When I received some units back for analysis I isolated the failure at the MAX3232. Interestingly, it is a partial failure - the RS-232 receiver is still functional but the transmitter has failed. I found this by observing the signals: A message can be seen to arrive from the RS-232 bus at RIN2 (pin 8) and the properly translated message leaves from ROUT2 (pin 9). The MCU processes the message and sends a response that can be seen arriving at DIN2 (pin 10). However, no corresponding RS-232 output is ever observed at DOUT2 (pin 7)!

Please help me understand what the possible causes of this failure might be. I have included a schematic of the MAX3232 as used in my circuit:

  • 0
    Guru 283370 points

    The schematic does not show what is connected to DOUT2.

    I suspect that the absolute maximum rating (±13.2 V) was exceeded, or that some ESD was too strong. Is it possible for a human to touch this pin? How is this pin protected?

  • 0 in reply to Clemens Ladisch
    TI__Guru 70992 points

    Michael,

    I agree with Clemens here, it sounds like there is some kind of damage on the DOUT2 pin. Was there anything connected to DOUT2 that may have caused a transient, or some kind of ESD event due to handling?

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    Prodigy 40 points

    Thank you both for your responses.

    The DOUT2 pin is connected directly to a pin on an external connector (DB15). The user connects their own wire harness at this point to attach my product to their existing RS-232 bus. So, transients are certainly possible.

    The MAX3232 datasheet states that " RS-232 Bus-terminal esd protection exceeds ±15 kV using human-body model (HBM)". I expected this to be sufficient for my use case, but is additional ESD protection required?

  • +1 in reply to Michael Boren
    TI__Guru 70992 points

    Michael,

    With 15kV integrated HBM protection, I agree that the device is likely protected against random ESD due to handling. The HBM rating is tested against a specific transient signature, and other forms of ESD have wildly different transient signatures that the HBM protection may not protect against. That being said, since this is on an external connector, there is also a chance of exceeding the absolute maximum, as Clemens said.

    Additional ESD protection isn't always required, but it is heavily dependent on the application itself. One TVS diode recommendation for MAX3232 that would work is the SMBJ8.0CA. You can also find more information in this blog on choosing TVS diodes for interface signals in general.

    Also, I know you said communication no longer works, but do ou know how it manifests specifically? That is, is the output shorted to GND, attenuated, or something else?

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    Prodigy 40 points

    Thanks for the article link. That was very informative.

    The DOUT2 pin on the faulty chip reads as 0V (product circuit ground) constantly. However, it does not read as a short if I check connectivity using a multi-meter...

  • 0 in reply to Michael Boren
    TI__Guru 70992 points

    Michael,

    You're very welcome, I'm glad it helped. What does the multimeter read when you check DOUT2 to GND? And can you compare it to a known good unit?

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    Prodigy 40 points

    On a known good unit, DOUT2 to GND is -5V. 

    On a bad unit, DOUT2 to GND is 0V.

  • 0 in reply to Michael Boren
    TI__Guru 70992 points

    Michael,

    Sorry for the confusion in my question, I meant the impedance measurement of D2OUT to GND.

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    Prodigy 40 points

    I see. On a bad unit the impedance is very inconsistent reading anywhere from 3.3K to 100K.

    A known good reads consistently at 23.7K.

  • 0 in reply to Michael Boren
    TI__Guru 70992 points

    Michael,

    Okay, thanks for the quick measurements. We can at least deduce that there is some kind of difference between the two devices, maybe relating to damage. Since the output is stuck at 0V, my thought was that there was a short to GND or a low-impedance path to GND due to an internal circuit getting damaged on the D2OUT pin. Not sure why it is showing an inconsistent reading, but this may be damage to another part of the D2OUT circuit, one that actually drives the signal from the buffer to the pin.

    Regards,

    Eric Hackett 

  • 0 in reply to Eric Hackett
    Prodigy 40 points

    Understood. Thanks for all of your input. This gives me enough to move forward with. I'll look into adding some additional external transient protection and/or migrating to the MAX3232E which seems to have better built-in ESD protection.