The AM Forum

Ok, being almost illiterate in digital design - that stuff that uses TTL chips, or used to, I have an idea, one that I really don't know what I am talking about, but I seem to think will work. So, anyone who actually knows something can possibly help me out (the window?) and set me straight I need help?

Take 60cycle AC, divide by 10 get 600hz square wave (logic level).
Now count to 12, toggle a flip-flop, count to 12, toggle again, etc... output = 50Hz. logic level.

Yea or Nay?

Better ways?

If this scheme is reasonable, I am totally unsure about things like the triggering part where the divide by 10 has to be made to turn on and off (synchronize) to the 60Hz... or how to count to 12, things like that.

Help/guidance/pointing to sites or designs is appreciated.

Tnx.

                    _-_-WBear2GCR

Dividing 60 hertz by 10 gets you 6 hertz, not 600...

An easier way to do this is with a 100 KC xtal or similar, use 7490 decade counters to divide down to 50 hz.

Ummm... multiply if you wish.
CHOP is maybe a better term?

The xtal is an idea that bears some investigation, but the goal is to have the frequency accuracy long term equal to line frequency stability. The question is will the divided 100khz xtal have little enough drift?

                _-_-bear

A crystal will give much better frequency accuracy than the line freq.  And it's pretty difficult to multiply frequency with logic.  You'd need a PLL to do it.

Thanks John, I will give this some thought... I had PLL in the back of my head, but as I said, somewhat unsure as to which way to think about it.

The Xtal has some positives...

So going up in freq is not so simple. I was thinking about how to sync to the start of one half cycle and the start of the next and get the divisions in between equal.

The best thing to do may be to just try the xtal and see how it measures up, seems like a simple enough breadboard, now to find any time!

                       _-_-bear

                     

The line freq is ok, however as JN says, several 100 ppm or better... not great. I have (see the sales postings)... A 100 MHz VCTCXO JUST WHAT YOU NEED :D Just divide it down to whatever you desire. Or get two and mix them up. Or get ten and really have a ball.

Alan

What JN said.

A PLL, is no better than its source, the master time base. A square time base clock waveform is easy enough to wave shape from sine. But no question about it, a PLL can operate as a multiplier or a divider of an input clock and is an excellent way to go for a flexible source vs. simply dividing, multiplying, or mixing a family of xtals (unless you have a junk box of excess xtals!)