LM317 adjustable supply

// An LM317 adjustable supply.
//
// There is no microcontroller here. Press Run and the simulator solves the
// circuit; turn the knob and watch the number under the LM317.
//
// WHAT IT IS. A 12 V wall adapter into the LM317's IN. The LM317 holds its
// OUT 1.25 V above its ADJ leg, and that is all it does. R1, 240 ohms from
// OUT to ADJ, has those 1.25 V across it, so 5.2 mA through it. The same
// current goes on through R2, the potentiometer from ADJ to ground, and lifts
// ADJ off ground by R2 times 5.2 mA. So:
//
//     Vout = 1.25 V x (1 + R2 / R1) + 50 uA x R2
//
// The 50 uA is the current that comes out of the ADJ leg itself, the small
// second term on the datasheet.
//
// THE KNOB. At mid travel R2 is 1 k: 1.25 x (1 + 1000/240) + 0.05 = 6.51 V,
// and the readout says 6.52 V, well inside the spread of the reference (TI
// gives 1.2 to 1.3 V). All the way down, R2 is nothing and the output is the
// 1.25 V reference. All the way up, R2 is 2 k and the formula wants
// 11.77 V, but the LM317 needs about two volts across itself: from 12 V in it
// gives about 10.5 V, and the rest is the dropout.
//
// WHY 240 OHMS. The LM317 has no ground leg, so its own operating current,
// a few milliamps, leaves by OUT. R1 has to take it. Change R1 to 10 k and
// take the LED off: the output climbs toward the input, because nothing is
// there to carry that current.
//
// THINGS TO TRY. Swap the adapter for 9 V in its properties and the top of the
// knob comes down by about 3 V. Or put a 47 ohm resistor on the output at
// 10 V and think about the heat: (12 - 10) V x 210 mA is 0.4 W in the
// regulator, and at 5 V out the same resistor costs it (12 - 5) V x 106 mA,
// 0.74 W. A linear regulator burns what it does not pass on.