• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 17 subscribers
  • Views 800 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.

CBT and 5V PCI-Bus

Niels Koll
Expert 1950 points
Other Parts Discussed in Thread: STRIKE

Hello all,

is it ok to use a CBT device on a PCI bus with 5V signaling levels?

PCI-Spec defines 11V max. voltage level (for 10ns), but CBT has 7V as absolute max. input voltage.

Since the 11V from the PCI-Spec are just the reflection of a 5.5V rising edge (2*5.5V = 11V), there should be very little energy behind this 11V spike.

Will such a spike harm the CBT device?

Thanks in advance, Niels

  • 0
    TI__Guru* 76130 points

    Hello,

    Since the question here is whether a CBT can be damaged by a voltage spike I am moving this post to the voltage translator forum.

    Regards.

  • 0 in reply to Elias Villegas M.
    TI__Mastermind 30280 points

    This is a Signal Switches question, and has been moved accordingly. 

    I have notified someone in the appropriate group about this post.

    -Brian

  • 0 in reply to Brian Berner
    Expert 1950 points

    Hello all,

    since nobody has answered my little question, I am wondering if there is nobody at TI who has the necessary skill or if this is the way TI takes care of customer needs.

    Frustrated Regards,

    Niels

  • 0 in reply to Niels Koll
    TI__Mastermind 46480 points

    Hi Niels

      Sorry for the late reply . This is not a forum I normally watch.

     The answer to you question is that an 11V spike will damage the CBT device. 

  • 0 in reply to Chris Cockrill
    Expert 1950 points

    Hy Chris,


    many thanks for your reply.

    I have principle question of understanding:

    The CBT data sheet defines 7V absolute max. input voltage, and you say that a 10ns spike of 11V will damage the device.
    But in the footnote related to the absolute maximum ratings, you can find the following:
    "Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability."
    From my point of view, 10ns is not an "extended period".

    One other question:
    Does CBT16211 provide any ESD protection? I didn't find anything in the data sheet.

    But other CBT devices have more than 2000V ESD Protection. How does this match wich your statement that a pulse of 11V for 10ns will damage the device?

    I am looking forward to your answer.

    King Regards,
    Niels

  • 0 in reply to Niels Koll
    TI__Mastermind 46480 points

    Hi Niels

      The note where it mentions extended exposure referes to voltages between recomended operating max of 5.5V and absolute max rating of 7V.  going beyond absolute can damage the part. Also keep in mind that functionality is not guarranteed beyong recomended operating conditions.

     The spike could be clamped by using an external diode to Vcc. Something like thisTPS1100 or any of the TPS ESD  parts. Also a ringing more than .5v below gnd will turn on most FET switches. If you have this you should use the Crev of the part that will prevent this.SN74CBT16211C

    All of our devices have some sort of ESD protection circuits on their I/O's to protect themselves but are not really made to be protection devices for the system.

     

     

  • 0 in reply to Chris Cockrill
    Expert 1950 points

    Hi Chris,

    ok, this means that any voltage higher than 7V will damage the CBT-device.

    But on the other hand, the CBT-device has ESD-Protection that protects its IOs against ESD-Events up to 2000V HBM.

    For me, this is a contradiction. The device will survive 2000V that last for app. 20µs, but will be damaged if 11V are applied to it for 10ns.

    Can you please explain this to me?

    Thanks in advance,
    Niels

  • 0 in reply to Niels Koll
    Expert 1950 points

    Hello,

    is my question so stupid that it's not worth answering it?

    Regards,
    Niels

  • 0 in reply to Niels Koll
    TI__Mastermind 46480 points

    The Esd protection is set to trigger off of a fast, high voltage pulse and and the clamp diode is set to clamp a normal voltage above vcc or below gnd.   The clamp diodes are designed to clamp undershoot and overshoot and they will have a current limit specd as Iik or Iok. If you go above Vcc or below gnd you will need to limit the current or the diodes will be damaged. all parts dont have a standarard diode to VCC so they can be overvoltage tolerant.

    The ESD protection is usually a scr type of circuit  that channels the esd strike to gnd. I am not an expert on how this works.