> On Jun 8, 2016, at 9:59 AM, Charles Steinmetz <csteinmetz@yandex.com> wrote: > > Nick wrote: > >> I’m contemplating trying my GPS board with an FE-405B. That’s a different kettle of fish, but at the end of that, if I’m successful, one of the goals would be to be able to use it for the external reference of my 53220A. Unfortunately, 15 MHz isn’t one of the options - only 1, 5 and 10. >> >> So I did some googling and found a divide-by-3 circuit using flip-flops, and then designed a board for it > > You can achieve substantially lower jitter (phase noise) with a regenerative divider, which also allows you to divide by 3/2 for a 10MHz output. I've built several like that, and they work extremely well. > > There are simpler divide-by-three logic circuits (generally, the simpler the circuit the closer to an exact 50% duty cycle and the lower the jitter). See the attached image for one approach. > I got the boards back, and Charles’ version with just two D FFs and a single XOR works perfectly. It works even without the delay line he indicated (sample size 1, FWIW). The version with the three D FFs and 3 NORs works just fine too, but of course it’s not as small as Charles’.

Nick wrote: > I got the boards back, and Charles’ version with just two D FFs and a single XOR works perfectly. It works even without the delay line he indicated I have never needed the delay in practice, either (I use mostly NC7SZ, 74AC, and 74HC logic). Modern logic is vastly more forgiving WRT setup and hold times compared to the bad old days. I showed the delay just in case someone built it without and it didn't work. What specific chips did you use? Best regards, Charles

https://www.digikey.com/product-detail/en/on-semiconductor/MC74VHC1G86DTT1G/MC74VHC1G86DTT1GOSCT-ND/2705092 https://www.digikey.com/product-detail/en/nxp-semiconductors/74HC74D,653/568-1490-1-ND/763395 > On Jul 6, 2016, at 8:39 PM, Charles Steinmetz <csteinmetz@yandex.com> wrote: > > Nick wrote: > >> I got the boards back, and Charles’ version with just two D FFs and a single XOR works perfectly. It works even without the delay line he indicated > > I have never needed the delay in practice, either (I use mostly NC7SZ, 74AC, and 74HC logic). Modern logic is vastly more forgiving WRT setup and hold times compared to the bad old days. I showed the delay just in case someone built it without and it didn't work. > > What specific chips did you use? > > Best regards, > > Charles > > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there.

Nick wrote: >> Charles wrote: >> >> What specific chips did you use? > > Nick wrote: > > MC74VHC1G86 > 74HC74D Good to know, thanks. Just FYI, you may want to look into the NC7SZ86 (and other NC7SZ "TinyLogic" series parts). The NC7SZ86 is pin-compatible with the 74VHC1G part, with several added advantages: 1). Output drive current is +/- 24mA rather than only 8mA. (This doesn't matter for the divide-by-three circuit, but in my book sturdy outputs are always preferred.) 2). It is a bit faster. 3). Speed doesn't drop off as much on lower-voltage supplies. 4). The NC7SZ series incorporates ground-bounce reduction for lower EMI and noise. Best regards, Charles