Requirements for LED on IR shield

[microtherion]

You guys know you can use the tone and noTone functions built in to Arduino to output ~38khz to any digital pin on the digispark just with tone(1, 38000), right?

Certainly a plausible idea.

This is the second thread in a week I've seen on this forum with people talking about messing up all the timer registers

Well, ultimately, tone() uses exactly the same timer, just in a different mode. You can choose to read the tone() documentation or the timer data sheet.


I’m used to working with lower clocked ATtinys; on an 1MHz processor, handling the 80K interrupts a second that tone() generates in this mode would take a fairly significant part of the available processing power, but at 16.5MHz, this is obviously far less of an issue.

[dougal]

I think in the current design, there's some pin overlap with I2C, yes? It might be nice if it could use different pins, so one could receive IR data (and maybe send) and then display on an I2C LCD. So if you do any significant redesign, keep that in mind?


This is off the top of my head, haven't looked to see how the pins would map out.

[jim]

So I take it this is consensus on using P1 for the LED?

@microtherion and @dougal - adding an IR LED spot was an 11th addition because we had the room on the board - it was untested and unplanned - P1 certainly makes more sense in retrospect - I thinkt he easiest way to make the mod would be to cut the trace in the solder jumper near P2 and then solder a jumper form that pad to the side of P1's pad. You could also use the MIT high low tech core.