• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 20 subscribers
  • Views 929 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.

SN74LVC1G80: Why Does 74LVC1G80 Make a Terrible Divide-by-2 Counter?

Jim Olson42
Prodigy 185 points
Part Number: SN74LVC1G80

I need a 60Hz square wave whose edges are synchronous with zero-crossings of AC mains voltage.  My zero crossing detector works fine and generates a clean 120Hz pulse train.

I feed this 120Hz signal into the CLK pin of a 74LVC1G80 "Tiny Logic" D flip flop that is configured as a divide-by-2 counter.  The Q-not output (pin 4) of the FF is tied back to the D input (pin 1). 

But this beast does not function as a divide by 2.  The Q-not output does not wiggle, period. 

The TI datasheet for this part shows a complicated implementation of a divide-by-2 counter where an external CPU pin with tri-state capability contends briefly with the Q-not output pin to preset the flip flop to a known state before it begins to divide by 2.

Why is this power ON preset strategy needed?  The part's truth table is simple and doesn't hint at the need for a power ON preset. 

I don't care about the flip flop's first state at power ON as long as the circuit eventually gets around to dividing my zero-crossing signal by 2.

Thanks.

Jim Olson

Lafayette, IN  

  • 0
    TI__Guru* 97046 points

    Hey Jim,

    Can you post scope shots of the clock, D, and Q\ pins? I find it very interesting that the device isn't switching.

  • 0 in reply to Emrys Maier
    Prodigy 185 points

    Thanks for the interest, Emrys.  To quote Gilda Radner's character, Emily Latella: "Never mind...".

    Don't take a knife to a gun fight.  And don't take a $1k Tek scope to an engineering task best suited to a $6k scope.

    The hint came from the 74LVC1G80's data sheet where it mentions the part toggles at 160+MHz.  The clock signal on the FF had a 1V/msec slew rate on its falling edge.  This was applied to an input pin on the 74LVC1G80 that offers zero hysteresis.

    A 160MHz gate can see stuff that a $1k scope can't.

    Solution: use an old MC14013BCP metal gate CMOS part.  It's as blind as a bat to noise.

  • 0 in reply to Jim Olson42
    TI__Guru* 97046 points

    I had a suspicion that there was a slow edge involved.  Hopefully one of these days we'll put out a FF with a Schmitt-trigger input to combat that issue.

    Thanks for letting me know what was going on, and good luck with your project!