• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 236 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.

DS16F95QML: Signal Integrity Parameters

Courtney Ballard
Prodigy 30 points
Part Number: DS16F95QML

Is there a specification for the minimum or maximum input rise time of this device, for both the single ended and differential inputs? Looks like only the output transition time is specified in the datasheet. 

I also see that the enable input voltage pin has an absolute maximum voltage of 5.5V, would the DE pin also tolerate -0.5V as a limit for undershoot? Would the DI pin have the same specifications?

  • +1
    TI__Guru 52172 points

    Hi Courtney,

    We don't spec a minimum/maximum input rise time for the device on either section - there is a minimum pulse width that needs to occur (based on the data-rate) so that the device can receive data correctly. That being said - its not that there isn't a minimum rise time to hit - because there is - but unless you are intentionally slowing down the signal input here (with like an RC network)  you most likely will be going fast enough. The concern is that if the input signal is too slow through the transition region you can shunt a lot of current and cause glitches. The slow rise time is more of an issue on the single ended side (DI , DE, etc..) as when we usually see glitches coming from the bus its because the received voltage is too low - not usually for it being too slow. 

    Essentially in the vast majority of cases  unless you are doing something very out of the ordinary the input rise times aren't a super concerning limiting value.

    DI and DE will have the same ratings - on the note for the negative limiting value; we don't actually spec it - so we don't guarantee it. That being said - there is most likely a diode for inputs to GND (there isn't one to VCC as abs max is independent of VCC) but it could be -0.3V or -0.5V and we don't guarantee it - so I'd try to avoid negative voltages on this pin as it is risk since we don't guarantee it.  

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    Prodigy 30 points

    Hi Parker, thank you for the thorough response! When you say the diode to GND could be -0.3V or -0.5V, I assume there is a single rating for that diode. Is that something you could find out? Ideally we like to not see any undershoot but knowing the actual rating of the diode would give a threshold for if we do see any.

  • 0 in reply to Courtney Ballard
    TI__Guru 52172 points

    Hi Courtney,

    Based on other RS 422/485 transceivers the diode is most likely -0.3V to -0.5V and there is a note of ESD protection which implies their existence (since VCC to inputs do not have a diode since max input is independent of VCC). We don't have this value spec'd or characterized so we don't guarantee it and assuming a value could add additional risk to the system. 

    With that being said - I understand that undershoot may be a concern in the system that is more or less unavoidable - its most likely safer to assume the -0.3V as an absolute min (there is some risk with this  - but honestly I don't think it would large - -0.3V minimum inputs are extremely common and have been for a while) and to increase the robustness  a series limiting resistor would be beneficial. 

    There is always going to be some risk since we don't characterize or guarantee it - - but if the undershoots can be attenuated and short lasting it will definitely mitigate a lot of risk as we don't explicitly state a minimum - but its safest to assume -0.3V as this is the most strict realistic threshold that you will see. 

    Best,

    Parker Dodson