Searching \ for '[OT]: A crystal question.. ][][' in subject line. ()
Help us get a faster server
FAQ page: techref.massmind.org/techref/index.htm?key=
Search entire site for: 'A crystal question.. ][]['.

Exact match. Not showing close matches.
'[OT]: A crystal question.. ][]['
1999\01\05@225652 by

Just a question that struck me late last night... why do they use 32.767
crystals for clock crystals? Would it be easier to use 100 hertz, for every
milisecond, or something in that multiple?

-Greg

-----------------------
Greg Cormier
Kathmandu, Nepal
Local time : GMT + 5h30m
ICQ # : 565465

At 09:40 AM 1/6/99 +0530, you wrote:
>Just a question that struck me late last night... why do they use 32.767
>crystals for clock crystals? Would it be easier to use 100 hertz, for every
>milisecond, or something in that multiple?
>
>-Greg

Greg,
The reason is that 32767 = 2^15 = 32768-1

Since digital logic is inherently binary which is
base 2, powers of 2 are the most convenient to use.

It is far less efficient, though not impossible,
to use base 10.  Chances are if humans had 2, 4 8 or 16 fingers
we too would be using base 2 to count instead of base 10.

73,
Dan Welch
w6dfwqsl.net

Hi,
It is very simple, 2 on the power 15 = 32768 . If You have a simple 15 step
binary divider You have 1 SECOND pulses at the output.
Cheers,
PeterS

>Just a question that struck me late last night... why do they use 32.767
>crystals for clock crystals? Would it be easier to use 100 hertz, for every
>milisecond, or something in that multiple?

>>Just a question that struck me late last night... why do they use
>32.767
>>crystals for clock crystals? Would it be easier to use 100 hertz, for
>every
>>milisecond, or something in that multiple?
>

As several people have pointed out, a binary multiple of 1 Hz is
a little easier to deal with.  Also, a 100 Hz crystal would be REAL BIG
compared to a 32.767 KHz crystal (which is more of a quartz tuning fork
than a typical crystal "plate").

Harold

Harold Hallikainen
haroldhallikainen.com
Hallikainen & Friends, Inc.
See the FCC Rules at http://hallikainen.com/FccRules and comments filed
in LPFM proceeding at http://hallikainen.com/lpfm

___________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
or call Juno at (800) 654-JUNO [654-5866]

>>Just a question that struck me late last night... why do they use
>32.767
>>crystals for clock crystals? Would it be easier to use 100 hertz, for
>every
>>milisecond, or something in that multiple?
>

|        As several people have pointed out, a binary multiple of 1 Hz is
|a little easier to deal with.  Also, a 100 Hz crystal would be REAL BIG
|compared to a 32.767 KHz crystal (which is more of a quartz tuning fork
|than a typical crystal "plate").

Note also that converting the 32,768Hz signal into 100Hz isn't all
that hard in a PIC.  If you have a routine that is executed every
64 clock cycles (i.e. 128x/second), then in that routine simply
code:
movlw   200
addwf   TimerAcc        ; A spare 8-bit register, used to hold fractiona
l
; 'ticks'.

Of the 128 times that code is executed, each second, the carry flag
will be set on precisely 100 of them.  If need be, the code may be
easily adapted for being called 256x/second, 512x/second, or 1024x
per second (assuming the CPU's running faster than 32KHz!)

Note, by the way, that executing the routine 1024x/second (with a
'movlw' value of 25) will yield a 100Hz output which is much cleaner
than that given above (much less jitter) but for purposes of stop-
watch display and such the above should be adequate.

More... (looser matching)
- Last day of these posts
- In 1999 , 2000 only
- Today
- New search...