[TI-H] Re: PIXpander & Stuff

To: ti-hardware@lists.ticalc.org
Subject: [TI-H] Re: PIXpander & Stuff
From: Adam Thayer <krevinek@mac.com>
Date: Wed, 24 Apr 2002 11:07:29 -0700 (PDT)
Reply-To: ti-hardware@lists.ticalc.org

 
On Wednesday, April 24, 2002, at 02:28PM, Nick Foster <zaphod@coe.neu.edu> wrote:

>
>To regulate your voltages, consider a network of Zener diodes acting as
>shunts. If you need help with this, you can ask me, or look it up in an
>electronics textbook. You can probably Google it, too. This would use far
>less power. Also, you could use three LM317 adjustable voltage regulators.
>This would be even more power efficient, and the LM317 is very easy to
>find. IIRC, they still sell it at Radio Shack.
>
>--nick

Both solutions require more expensive components. Zener diodes would be a good way to do it though, and I can limit the current the diodes sink rather easily, and would be simple using an 8V regulator for the main source voltage. The LM317 *IS* very easy to find, but requires a few support components along with it, making it more complex and expensive overall than using the Zeners. Overall, this is not a vital matter at the moment, since all my test cards are 3.6V, and as long as my battery is okay, run well off my poor man's power supply (3-12V AC->DC converter from Radio Shack).

However, there are 2 odd behaviors I noticed with my breadboarded project:

1) I have a switch hooked up oddly. One pin has an 18k pullup resistor and a connection to the switch pin I use on the PIC, the other pin of the switch is hooked up straight to +5v like the pullup resistor. However, the switch still behaves normally. I haven't managed to find the exact reason why this is happening, although it is rather trippy to observe, and possibly saves a little power.

2) I am using a TI<->TI cable, and the documentation says the outer band is Red (1) and the middle band is White (0). Yet my code which assumes this when getting/sending data to the TI winds up getting the bits flipped. So 00000110b (0x06, or a Var Header packet type) is seen as 11111001b. The data is all valid once I flip all the bits, or switch the wires. Anyone else see this and know what is going on? Is the inconsistency due to the difference in a unit link cable and a computer link cable (the little one that comes with the Black and USB links)? I have checked my data more than a few times, and every time my data seems to be coming out flipped.