• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 17 subscribers
  • Views 569 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.

Original question:

CD4051B: MUX input structure of CD4051B

CD4051B: Failed to work like p2p competitor NLHV4051

Xizi Luo
Prodigy 610 points
Part Number: CD4051B
Other Parts Discussed in Thread: MPC508, TMUX4051

Hi there,

My customer is replacing NLHV4051 with our CD4051B, and these two devices are p2p compatible. The mechanism is that a high voltage signal (could be 9/12/24V) goes under some conditioning and is clamped at about 4.8V at the input of pin14 of CD4051B. Meanwhile, the conditioned 4.8V signal goes to an OR logic circuitry to wake up the MCU, then the MCU enables a DCDC to supply 5V power to CD4501B, and the MCU receives the signal from CD4051B, shown as below.

Our customer has soldered these 2 MUXs on their PCBs. The NLHV4051 performs well as the MCU is successfully woken up and can receive signal from the MUX. But our CD4051B cannot wake up the MCU, and the voltage after conditioning is 0.8V but not 4.8V. BTW, CD4051B didn't get heated. The schematics below are for your reference.

Would you please kindly give us some comments on why our p2p CD4051B didn't work? Is CD4051B not allowed to be used like this? Thank you a lot.

  • 0
    Guru 318050 points

    Both the NLVH4051 and CD4051B have clamping diodes from the I/O pins to VDD. With VDD = 0 V, the I/O voltage cannot go above the forward voltage of the diode.

    For both the NLVH4051 and CD4051B, voltages above VDD + 0.5 V are forbidden by the absolute maximum ratings, and can damage the device if the diode current exceeds 10 mA. I do not know why the diodes on the NLVH4051 do not actually clamp the voltage, but the electrical characteristics of these diodes are neither specified nor guaranteed. In any case, your circuit should have used external clamping diodes (see figure B of the NLVH4051 datasheet).

    The only 8:1 mux with powered-off protection is the MPC508, with different electrical characteristics.

  • 0 in reply to Clemens Ladisch
    TI__Prodigy 610 points

    Hi Clemens,

    Thanks for your fast response. As this account is very cost-sensitive, I guess the cost of MPC508 couldn't compare with CD4051B, right? Do you think TMUX4051 with external pull-up and pull-down clamping diodes could realize the function? Thank you!

  • 0 in reply to Xizi Luo
    Guru 318050 points

    In theory, the TMUX4051 should behave similar to the CD4051B. (But in theory, the CD4051B should behave the same as the NLVH4051. I do not know what is going on there.)

  • 0
    TI__Genius 10552 points

    Hi Xizi,

    Thanks for reaching out here. Clemens provided accurate information here regarding the clamping diodes and their abs max ratings. It's possible that the competitor part is designed beyond what it is spec'd for which is why the customer has a working solution before dropping in the CD4051B here. Almost seems like they intentionally made the device powered off protected, but did not specify that in their datasheet. Unfortunately, the MPC508 cannot compare on cost. As you suggestion the TMUX4051 with external pull up and pulldown clamping diodes would replicate the needed function, but this would need to be simulated or tried in real time to guarantee performance. 

    I hope this helps. Let me know if you need anything else here!

    Regards,

    Alex

  • 0 in reply to Clemens Ladisch
    TI__Prodigy 610 points

    Hi Clemens, 

    I guess it is D43 in figure 3 above that prevent from overcurret.

    VDD should be supplied with power first, then I/O channels receive signals. Is this the correct way to use CD4051B? 

    Thank you!

  • +1 in reply to Xizi Luo
    TI__Genius 10552 points

    Hi Xizi,

    That is correct. Without the powered off protection feature, we must follow the "traditional" power sequencing method of first powering on VDD then applying the appropriate signal to the I/O channels. 

    Regards,
    Alex

  • 0 in reply to Alex Vish
    TI__Prodigy 610 points

    Hi Alex,

    Many thanks for the help. I'll keep you posted if this is solved :)

  • 0 in reply to Xizi Luo
    TI__Genius 10552 points

    Sounds good, Xizi! I'll go ahead and leave this thread in a WFC state. I look forward to the results!

    Regards,

    Alex