Ian Smith wrote:
> On Tue, 4 Jan 2000, James Paul wrote:
> >  Regarding the 4 lead crystals.  Are these actually crystals or
> >  are they canned oscillators?  Do they look like a DIP package
> >  with 1 pin in each corner?  And all corners are rounded except
> >  one which indicates pin 1?
>
> That is it.
>
> >  If this in fact is what you have,
> >  then pin 1 can be either N/C or a TRISTATE pin, Pin 7 is GND,
> >  Pin 8 is OUTPUT, and Pin 14 is Vcc.   If you have the halfsize
> >  parts then Pin 1 is either N/C or TRISTATE. Pin 4 is GND, Pin 5
> >  is OUTPUT and Pin 8 is Vcc.  These are easy to see if they
> >  operate.  Just provide power to pin 8 or 14, Ground pin 4 or 7,
> >  and measure pin 5 or 8 with a scope of frequency counter.  If
> >  you don't see any activity, then pull pin 1 high in either case
> >  to enable the TRISTATE output.
>
> *tinker*  Ahh, I got my PIC running with a ferw of them at different
> speeds.  I was never able to get it to work before, thanks!  I had
> assumed that you needed to take two of those pins into the CLK1
> and CLK2 like a 2-pin crystal, but I see that was incorrect now.
>
> So the 4-pin units have a crystal and an oscillator.  I assume the PIC
> has to have an internal oscillator since it can run a regular 2-pin
> crystal?
>
> Is a resonator equal (with less accuracy) to a 2 or 4 pin osc?

(Offline for a bit here, just got back home after "a day".)

Basically, yes - Fewer holes / components, often somewhat smaller /
lighter, and cheaper than a "rock" plus caps, not as temperature variant
as using RC - for many projects it's a great answer (for example,
Hosfelt has 512kHz resonators, I'm using some on a time-smart monitoring
system that beeps but delays an extremely loud alarm, 3 minutes or so,
giving you time to move things without an alarm, or to handle the
problem before it "wakes the dead" with it's "Uh-Oh, things are going to
BREAK!" alarm <G>)  Also small enough that you could just SMD solder the
resonator on, under the PIC - hard to do with a large crystal.

> >  If you could supply some of the numbers on the top of the 4 pin
> >  crystals, that may help identify them easier.
>
> CTS SINGAPORE; MXO-55GB-2C; 1.8432 MHZ; 8747 1202
> CTS KNIGHTS INC; MXO 50-2; 19.584 MHZ; 8138 028
> 6N/S TD1145C; 25.175MHz; 8918 NDK JAPAN
> CK1145MC; 28.322200MHz; 8912
>
> Do the 8xxx numbers refer to some common standard?  They all seem to
> have them, even though they are made by different companies.

"Welcome to the wonderful world of date codes."  8747 means 47th week of
1987, typically;  Some places would have a date code like 87289, which
means 289th day of 1987, usually week numbers though.  Knowing this, you
can look at most any PC board and pretty accurately figure out when it
was manufactured - on any mass production board, they get most parts
right as production happens - so a board with a few 8023, 8105, and 8153
date codes, and a few 8244 and 8245 chips on there, was likely made in
fairly late 1982, for example.  Handy knowledge sometimes <G>

> --
> IanSmith@erie.net

  Mark

--
I re-ship for small US & overseas businesses, world-wide.
(For private individuals at cost; ask.)