The fastest flip-flop?

[neazoi]

fastest d-flip-flop

Hello
Which is the fastest flip flop? I need an ultrafast Dual D-Type Positive Edge Triggered Flip Flop to use it as a phase splitter and by-4-divider.
I have an 74F74 but the 120MHz is not enough. Something like 1-2GHz would be fine if possible, or at least much higher than 120MHz.

 

[FvM]

fastest cmos flip flop

Classical ECL logic provides several GHz toggle frequencies, see e.g. https://www.onsemi.com/PowerSolutions/taxonomy.do?id=273 for reference.

Basically you have to think about a suitable interface standard that still works at this high frequencies. Differential standards as ECL, PECL, LVDS and CML are first choice.

 

[erikl]

mixer using d flip-flop

I have an 74F74 but the 120MHz is not enough. Something like 1-2GHz would be fine if possible, or at least much higher than 120MHz.

ECL, PECL, LVDS and CML are first choice.

What about one of these?

 

[neazoi]

fastest flip flop

It is not needed only for frequency division but at the same time for phase splitting, so a Flip Flop must be used. So I am looking for a high frequency flip flop in general.
Any specific flip flop to propose?

 

[erikl]

sige flip-flop -patent -edu

It is not needed only for frequency division but at the same time for phase splitting, so a Flip Flop must be used. So I am looking for a high frequency flip flop in general. Any specific flip flop to propose?

And what - would you think - would these people use for frequency dividing, other than a Flip Flop ? This is a Flip Flop, nothing else! Better study your answers more thoroughly!

 

[neazoi]

high freq flip flops

Thanks for your answer, but please do not get it personal. We are not criticizing people here, we just need to find solutions.

Thanks once again.

by the way except from flip flops there is the mixer component too as you already know.

 

[erikl]

Thanks for your answer, but please do not get it personal. We are not criticizing people here, we just need to find solutions.

Right, I tried to help you, and was disappointed that you wouldn't look thoroughly at the 2 URL's which I sent you. Sorry for my overreaction!

by the way except from flip flops there is the mixer component too as you already know.

No, this is a (up to) 2.6GHz flip flop only (s. page 2 of the appended datasheet). Didn't you look at that one, either? Perhaps you didn't check the PDF which comes with the URL I sent you last time? I'll append it here directly.

 

[Teddy]

ONsemi has also SiGe differential flip flops which go way faster than those erikl showed. - you just pay more.....

 

[neazoi]

Thanks a lot for your replies, very helpfull, I will have a closer look over the data.

Just for reference, one major advantage of using flip flops instead of mixers for down conversion in radio, is the absence of local oscillator and the improvement of the stability of the divided frequency. The divided signal is square wave, but this may be considered an advantage in some applications as square wave is less affected by noise from the PSU.

 

[dick_freebird]

You can look up "dynamic DFF" which is capable of well more
than 2GHz in thin film CMOS-SOI at 0.5um. This is a single
ended, full-swing architecture. At 0.25um we are able to exceed
10GHz prescaler operation. I'm not going to advertise or say my
employer here.

Source-coupled logic in any decent short-channel CMOS ought
to be able to hit your numbers too, but at higher power.

Try looking for "CMOS RF prescaler" rather than asking for the
digital "flip flop".