Re: TI-H: Replacing Capacitorr with Crystal?

To: ti-hardware@lists.ticalc.org
Subject: Re: TI-H: Replacing Capacitorr with Crystal?
From: Bryan Rittmeyer <bryanr@flash.net>
Date: Wed, 08 Jul 1998 20:45:11 -0400
Delivered-To: ti-hardware-outgoing@towerguard.unix.edu.sollentuna.se
References: <000f01bdaa9d$d6fe0140$2d38d2cc@towers.san.rr.com>
Reply-To: ti-hardware@lists.ticalc.org

Actually TI wasn't at all sloppy. Quartz crystals are extremely
expensive (just look in a digi/mouser catalog) whereas an RC pair
is extremely inexpensive, especially when you buy 100,000 
resistors/caps at a time. There isn't any real reason for a precise
timebase on a TI calculator anyway. I personally would have chosen
a resonator just because they tend to be closer to the labeled value
than an RC pair (capacitor tolerances being around 20%) but even
a resonator would have been much more expensive than their SMD r/c's.

Additionally, Quartz crystals are prone to shattering and I'm sure
most of the students on this list have dropped their calculators at
least -ONCE-. All in all TI knows what its doing...

You can pretty easily yank out the RC combo and put in a crystal if
you so chose, as long as there is an inverting crystal driver on
the z80 in there. I'm not too familiar with z80 board design but
I'd imagine it has two pins (Like the PICs/AVRs) for its osc, one
to drive a crystal and the other as an input to its clock circuitry.
Chances are the RC pair doesn't touch the crystal driving pin and
just feeds its clock right into XTAL1.

Or you could go to your local K-Mart and buy a Timex watch for $10 :)

> SMC12 wrote:
> 
> I know TI was sloppy in that they didn't use a crystal, and instead
> used a capacitor to control the clock speed.  Which is why we can
> replace C9 (or whatever it is for that calc) with a different value to
> get faster/slower speeds.  The biggest problem with Clock programs,
> like the one I saw on my friend's HP48G, is that clock speed isn't
> constant, it varies with battery level.
> 
> My question is how difficult would it be to create an add-on unit that
> goes in the former C9 connection that would use a crystal instead of
> the cap?  It would allow for more accurate transmissions (possibly)
> and it would allow clock programs.  Might even open up the door to
> more sound development.

-- 
Bryan Rittmeyer
mailto:bryanr@flash.net
http://www.bridges.edu/horizon/

References:

TI-H: Replacing Capacitorr with Crystal?

From: "SMC12" <scross@san.rr.com>