Hi folks,

For many years, I have used a standard D type FF (74x74) to divide an incoming frequency. Hook the D input to the Q output, & works great.

I just used one of the single section D-flops (see title & it's a little bitty part), and it didn't work.

Both the Pre & Clr lines are tied directly to 3.3v, and the supply has good bypassing.

To make it work, I had to put a 10k resistor in place of the D-Q connection and add some capacitance to the D side. (slowing the signal down).

I substituted a standard 74HC74 in it's place (kludge on the board), and it worked fine without the extra circuitry.

Am I seeing some kind of race condition due to the higher speed of the part?

Looking at the data sheet, you have to hold data on the D pin at least 1.2ns after the clock rises, and the prop-delay through the part is 2.2ns. So, at least spec-wise, it looks like I have 1ns to 'spare'.

Ideas?

Thanks,

Tim

 

Peculiar. I think you're on to something here. I cannot see a reason why this is happening. These are very fast logic, 200MHz clock rates. The only thing I can think of is there is something peculiar about your set up - rise time/fall time, board capacitance, power supply etc. Perhaps contact the chip maker.

Let us know when you find an answer.

 

Is the chip decoupled with a small ceramic cap (0.01 to 0.1μF) directly between the power pin and ground pin?