[Alchemy]

Here goes my BillA impression:

Originally posted by JSimmons
My idea is to have (three or more, depending on desired granularity) IR receiver/emitter pairs aligned along the vertical axis of a reasonably tall reservoir. These IR pairs would each be responsible for altering the state of an associated status indicator depending on the water being low enough to trigger an electronic signal.

We are apparently graced by the reincarnation of Rube Goldberg.

1) Is lucite transparent to IR? Don't laugh - some glass isn't, so I don't want to just assume that lucite is.

Lucite is not glass. And most polymers are transparent to IR, though they can easily be treated to be otherwise.

2) How much heat would be associated with three or more emitters (IR LED's) being on 24/7 (or at least while the computer itself is on)? Negligible amounts?

Maybe 3 watts apiece. And if you're using IR LEDs, you will be emitting IR just beyond the range of visible light. So the answer to question 1 would be yes - if it's transparent to red light, it'll be transparent to near, near-infrared light.

3) If the water cannot be made to block/diffuse IR light, I would have to consider visible light. This has it's own problems:

If it diffused the IR beam, it would hit all of the sensors. Same thing if it did any other sort of light scattering. You want the beam to be absorbed.

A) Ricers (my term for people that put lights in their case for no reason except that "it's pretty"), would not be able to make use of such a kit. The same goes for people with windows in their cases.

It would also work poorly if there was an LED on the mobo or the case was not sealed completely.

B) Lucite has an annoying (in this case) tendancy to edge light when a light source is place in close proximity.

"Edge"?

C) Photocells are not exactly reliable as far as their light sensitivy values are concerned, so even in a dark case, one light illuminating the lucite AND the water could errantly activate an receiver that is not associated with the emitter in question.

Photocells are designed for detecting visible light. IR is at the very edge of their detection. I see no reason to say they have decreased sensitivity in the visible range.

Since the visible light concerns are enough to convince me to try IR first, the IR's spectrum needs to be able to exist in a dark case, a case with just a clear window, a case with enough lights in it to illuminate a small city, and a case that contains a black light.

A photocell *will* detect visible light shining on it, and make no differentiation between it and IR. If the lights in the case are anywhere near as intense as the detection limit of the photocell, the sensor will give a signal.

However, before any of this can work, I need to make sure I can create a fluid environment that can effectively block/diffuse IR light.

If it blocks visible light, it'll probably block the sort of IR frequencies you'll be making with a LED. So just fill your water with enough dye to make it dark as night and you're all set.

There's a number of different indicator styles that could be used:
- Multiple single LEDS
- a bank of LED's (I've seen them arrange in straight and arched row packages)
- a LED character display
- a LCD character/graphics display
- a VFD

Bit ahead of ourselves, aren't we?

Alchemy