480KHz or 3.3768MHz

Agreed. The easiest frequency to get out of a quartz crystal is 10MHz, and 3.3768MHz is close enough that both stability and accuracy will be a lot better than you'd get from a crystal resonant at 480kHz.

There is also the point that the 3.3768MHz crystal is a commodity item for the TV receiver industry, so you get maximum economies of scale.

---------- Bill Sloman, Nijmegen

Almost certainly the original poster meant "3.2768 MHz", seeing as how he's in a 50-Hz part of the world and wants to use a 2^16 divider.

NatSemi used to sell chips (MM5369 with appropriate variations) to divide down from colorburst frequency to 50Hz and 60Hz but those haven't existed for years.

PAL colorburst is something like 4.43MHz, but certainly not every place that uses 50Hz uses PAL. I'm pretty sure HK is 50Hz and uses PAL (although one of my charts says NTSC).

I'm pretty sure long after NTSC has left us that we will still have

3.58xxx MHz crystals. Just like we still have 6.3VAC and 12.6VAC filament transformers but only rarely do we have filaments :-).

Tim.

The color crystal!

Digikey stocks them, although due to pricing the 6.5536 MHz units will be a lot cheaper (factor of ten) if you've got an extra flip-flop to do the divide-by-two.

If this is for some kind of timekeeping, a 32kHz crystal is dividable with integer math (a divide-by-ten followed by a divide-by-64 for example) and probably a better tolerance (commonly used in watches, although not quite as common as 32.768kHz) and certainly results in lower power consumption if this is a critical parameter.

Tim.

Well..... it's only a commodity item for NTSC TVs. There's rather more non_NTSC sets made in China I'll wager.

I've seen the suggestion before that it makes that particular frequency 'cheap' but never seen this in practice. Maybe because we have PAL TVs here ? I suspect the assertation is bogus however or maybe one of those things that was true long ago.

Graham

Yes, that's was my thought. They are not very common from what I can see, but the crystal manufacturers should be happy to make them if you buy enough, and it shouldn't be that hard to find a few. Mostly a potential problem if you want them to be available for long-term small quantity production.

Yes, PAL, as does mainland China (maybe different flavors of PAL).

A little 8-pin microcontroller would be one way to get 50Hz from something common like a 4MHz, 10MHz or 8MHz crystal or resonator without much fuss.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Well, here in the US, at at least, it has always been 3.579545MHz.

jp

"amdxjunk" schreef in bericht news:861a8$4312f2b5$18d6b479$ snipped-for-privacy@KNOLOGY.NET...

Hahaha, that 3.3768MHz must be a typo (3.2768Mhz).

-- Thanks, Frank. (remove 'q' and '.invalid' when replying by email)

Yes, the fact that you were using a rounded off figure. Of course it's easier to say "3.58MHz" but it was never that, always 3.5768MHz.

Michael

Notwithstanding the freq. of the NTSC color burst is 3.579454 MHz. ;-)

I thought he was taking about a 3.2768 MHz watch crystal (3.2768 X 10), but made a typo.

Albeit, for all I know, PAL might use a 3.3768 MHz burst.

Cheers! Rich

Er... The American color TV crystal is 3.58 MHz (or 3.579 something). Is

3.3768 a British one?

And I realized I had it wrong after I posed. Nevertheless, the point holds, that the "3.58MHz" is a rounding off.

Michael

I read in sci.electronics.design that Rich Grise wrote (in ) about '480KHz or

3.3768MHz', on Mon, 29 Aug 2005:

M/PAL uses 3.57561149. N/PAL uses 3.58205625. All other official versions use 4.43361875. All official versions of SECAM use 4.406250 and

4.250000.

--
Regards, John Woodgate, OOO - Own Opinions Only.
If everything has been designed, a god designed evolution by natural selection.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Nope. 4.43361875 MHz, to be precise :-)

Meindert

No at all; the color crystal is best for the various reasons given. If that seems to you to be a third choice or one not originally mentioned, then you get a few points for keen observation.