• State TI Thinks Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 7 answers
  • Subscribers 18 subscribers
  • Views 351 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.

SN74CB3Q3257: what will happen if pin7 short to GND

user4979989
Intellectual 2800 points
Part Number: SN74CB3Q3257

Hi Sir,

Could you please check what will happen if SN74CB3Q3257DGVR pin7 short to ground?

Will it result in system/board damage or burn out issue?

Thank you.

  • 0
    Guru 243090 points

    Pin 7 is 2A. This pin will be connected to either 2B1 or 2B2.

    The switch itself does not care about the voltage. The other devices connected to 2B1/2B2 must be able to cope with a connection to GND.

    If other devices connected to 2B1/2B2 actively drive high, they will source current that flows through the switch into GND. This will exceed the absolute maximum rating of 64 mA, which is likely to cause damage.

  • 0 in reply to Clemens Ladisch
    Prodigy 60 points

    What if the pin7 has low resistance to GND (~10 ohm) due to IC itself damage, will that cause board burnt or system shut down?

  • 0 in reply to Aaron Chiu1
    Guru 243090 points

    The effects are the same.

  • 0 in reply to Aaron Chiu1
    TI__Mastermind 43245 points

    Hi Aaron,

    As Clemens mentioned - this configuration by itself will not cause damage, however since this part is low resistance with a typical on resistance of 4 ohms. This means relatively small voltages across the switch can create relatively large currents. For example, this parts max current is 64mA; you would only need 256mV across the switch to hit the max current limit before you may start seeing damage. 

    In short as long as the absolute max ratings, such as voltage and current, are respected the switch will not sustain damage.

    If you have any other questions please let me know!

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    Prodigy 60 points

    Hi Parker,

    We suffer from bus check error due to this IC damage recently. And our factory sent 2pcs IC for failure analysis. Attached is the FA report we got today.

    The FA report says it's caused by EOS. Do you have idea which one (ESD, over voltage/current, over heat) may cause this from the picture you see?

    We want to get this issue repro by ourselves first.

    Thanks,

    AaronQEM-CCR-2010-00484 Final Report.pdf

  • 0 in reply to Aaron Chiu1
    TI__Mastermind 43245 points

    Hi Aaron,

    Based on what I am seeing it seems the cause of the issue is an overvoltage/overcurrent event. These events can cause ESD protection on the device to short the pin to ground or VDD. 

    Also due to the fact that only 1 pin is suffering damage I'd think its not temperature related. Based on the setup, I'd try to measure the voltage being put across the switch when it you connect the switch to ground, I'd imagine there might be some transient that occurs when switching to ground that could cause issues.

    However if the voltage across the switches induces too high of a current, or the voltage with reference to ground is higher than abs max that will also increase the chances of damage significantly as well.

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    Prodigy 60 points

    Hi Parker,

    Thanks for the advise. 

    We did the voltage across the switch measurement when it's at static state, no matter in low or high level, it's lower than 256mV.

    However, we observed that the high voltage delta when it's in transient, such like '0' to '1'. But we think it's normal behavior as there has delay between the switch. 

    We'll try to follow your step to see if the issue could be repro.

    Thanks,

    Aaron

  • 0 in reply to Aaron Chiu1
    TI__Mastermind 43245 points

    Hi Aaron,

    Another thing that you may want to check, if possible, is to see what the current being injected through the switch is. Overcurrent is typically the big killer for the mux, so if a transient current spike violated abs max, that could be possibly the source of the issue as well if the voltage tests don't prove to find anymore issues. Please let me know if you need anymore assistance on this problem!

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    TI__Mastermind 43245 points

    Hi Aaron,

    One thing that may help reduce these transients is to add a bit more capacitance to the input of the switch can help smooth those out, at the cost of slightly slowing the response of the circuit. If possible adding a capacitor here may prove to be helpful - if you can get a reading of the peak of the transient I can probably help narrow down a capacitor value. 

    Best,

    Parker

  • 0 in reply to Parker Dodson
    Prodigy 60 points

    Hi Parker,

    Thanks for the suggestion. We will consider to add the small capacitor for future product.

    We did some experiments but not yet to repro the issue. We'll perform the ESD and over-current test on these two days.

    Two questions that I would like to asks:

    1. If the over-current on the pin 7, then it should break this switch path, right? That means both switch side should have low resistance, pin 5 or pin 6.

    2. The fail symptom that we observed so far is pin 7 low resistance. Is this pin weaker than other pins?

     

  • 0 in reply to Aaron Chiu1
    TI__Mastermind 43245 points

    Hi Aaron,

    1. Overcurrent on pin 7 can cause damage either to the switch pathway or to the ESD cells on the pin, I can't say which one will happen first because it depends on other conditions as well, such as during these high current transients there could be voltage swings on the pins that cause the ESD diodes to conduct as well. 

    2. This pin isn't weaker than the other source pins on the device, they are essentially copy/pasted designs so that each channel behaves similarly. The low resistance path to GND observed on Pin 7 looks like the ESD cell on pin 7 has shorted to ground, causing the issue you are seeing. This is a result of electrical overstress.

    I am sorry about the late reply on this thread - I believe we are on a few email chains as well for this question. If you could move your future questions about this to the email threads that we have running currently it would be of great help so that I don't miss any question. I will still be the person responding to the questions I just don't want to miss any question!

    Best,

    Parker Dodson  

  • 0 in reply to Parker Dodson
    Prodigy 60 points

    Hi Parker,

    I am not sure which mail thread that you mean to we are chain together? Please allow me to ask further questions here.

    What if there is a negative voltage on VCC or the undershoot is out of the spec., will that break the ESD diode on the FET low side?

    Thanks,

    Aaron

  • 0 in reply to Aaron Chiu1
    TI__Mastermind 43245 points

    Hi Aaron,

    1. It doesn't necessarily have to break the switch; it could also break ESD protection structures depending on all of the conditions at the time of failure.

    2. Pin 7 is essentially copy pasted version of the same type of source pin that is seen elsewhere on the device. This pin is rated the same as the other Source/Drain channels on the Mux.

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    Prodigy 60 points

    Hi Parker,

    Yes, we're considering to make the negative voltage (-0.5V) on VCC or increase the undershoot voltage on pin7. Will that break the ESD protection structure as attachment?

    Thanks,

    Aaron

  • 0 in reply to Aaron Chiu1
    TI__Mastermind 43245 points

    Hi Aaron,

    A larger undershoot will cause the highlighted ESD structure to conduct, I am not sure however the duration of pulse needed to cause it to break down though. Overvolting the pin can also possibly cause the ESD structure to break down as well and short. 

    So in short yes, it can break the ESD diode, but I am not sure how long the duration of the pulse would be required to break the ESD structure, and it will vary based on the magnitude of the pulse with respect to the absolute ratings.

    Best,

    Parker Dodson