TLV3802: Abnormal output from TLV3802

Part Number: TLV3802


Hello,

This is Moon.

I’m seeing abnormal output behavior on the TLV3802. In the attached screenshot, the orange trace is the comparator input and the green trace is the comparator output. With the threshold set to 2.1 V, I expect an output pulse width of about 7–8 ns, but the actual waveform does not match this.

The input to the comparator has a 2 V DC offset, and the supply rails are 0 V and +5 V, so I believe the common-mode range is within spec. However, the measured output shows a second LVDS transition with a width greater than 100 ns—that is, the output toggles twice (two rise–fall events).

 

ju.gif

Could you please advise on the likely cause of this behavior and recommended next steps?

Best regards,
Moon

  • Hi Moon,

    Beware that the input range of the TLV3802 is NOT ground sensing and has a limit of VEE + 1.5V (that's 1.5V *above* ground).

    So on a single +5V supply, the legal input range is +1.5V to +5V. It looks like they are right on the edge of the 1.5V limit. Going below the 1.5V limit could cause reversals, false or delayed triggering or other misbehavior.

    It looks like you are getting a valid transition a few ns after the pulse, but then the input pulse drops below 1.5V and the output eventually reverses ~100ns later.

    What happens if the input voltages are kept above 1.5V?