On the subject of creating fake vacuum tubes
OK, here's the skinny. If you're at all into building electronic stuff, this is where I need your help - and if you don't but know somebody, have them reply, please. Note, no anonymous posts, as always. =)
I'm in the middle of coming up with some costume concept here: a fake vacuum tube (or valves if you're British). The purpose is for a steampunk airship pirate fae for Faerieworlds, which is happening in a few months. To explain, I thought it would be really incredibly cool to do this, and started branching off of using the real McCoy, but there are four problems: one, they're expensive, two, they're hard to come by outside of Guitar Center, three, they require a decent size power supply to even get hot, and four, well, they get HOT. So, instead of that, we fake it.
With the help of dicemistress, I came up with that the glass envelope should be a test tube. Optimally, I will find one that's about 1" dia x 3" long, but I don't expect that to happen - so some REALLY FRACKING LONG firebottles are probably in my future. In addition, one of my girlfriend's roommates suggested UV LEDs, and use some blacklight reactive paint to simulate the glowing. Since this is for a fantasy event (and because you can apparently get tubes at Guitar Center that glow something other than the characteristic warm orange of tubes from back in the day), this is a feasible option. He had also noted that the use of heavy copper wiring to simulate the anode and cathode would add to the archaic flavor that is appropriate to steampunk - so the "anode" would do this, while I'd be doing more of a rigid support system and finer gauge wire for the "cathode", for reasons that will shortly become obvious. For plugging them in, he had also suggested something where I could get a "twist-lock" motion going.
A quick search on Ebay turned up a few different suppliers of UV LEDs, and general rule I'm seeing is that they're 3.3v/20 mA, and require a limiting resistor if the power supply is >4W. Most efficient use, as near as I can tell, would be to wire the LEDs in parallel rather than series, so that I can actually get a rigid support for the LEDs and not have to worry about the "cathode" collapsing inside and simply looking funny.
This is where I'm stuck - the general schematic. I have, in short, forgotten Ohm's Law, as I've been way out of practice since something like 1998.
I have two plans for the wiring, keeping in mind that the guts of each "tube" would be wired in parallel rather than in series, and would contain two or three LEDs each.
The following, in short, are the design plans:
1) wire the power supply as a central power supply, and run series wiring throughout the assembly, additionally wiring to avoid what I call "old christmas light string syndrome" - that is, one burned out bulb (or, here, not connected tube) breaks the entire circuit and accordingly kills all lights (upshot: I can plug and unplug tubes on the fly for effect)
2) Again with the central power supply, but wire all sockets parallel to the supply (cures the OCLSS problem as above)
3) Wire each "tube" with its own power supply (probably a series of watch batteries, or a pair of AA batteries, but creates multiple switches - one for each tube)
On this note, for the plug end and any additional gribbles, I can probably work with ksharra for other ideas. But that aside, here's where I have the questions.
1) Which of the above design options is most optimal for the general power feed?
2) Given that I'm soldering at least two LEDs for each tube, how do I calculate resistance and power requirements?
3) In the case of 2, where do I solder the resistor?
4) If series soldering of the LEDs is more efficient, what would the schematic look like?
I'm in the middle of coming up with some costume concept here: a fake vacuum tube (or valves if you're British). The purpose is for a steampunk airship pirate fae for Faerieworlds, which is happening in a few months. To explain, I thought it would be really incredibly cool to do this, and started branching off of using the real McCoy, but there are four problems: one, they're expensive, two, they're hard to come by outside of Guitar Center, three, they require a decent size power supply to even get hot, and four, well, they get HOT. So, instead of that, we fake it.
With the help of dicemistress, I came up with that the glass envelope should be a test tube. Optimally, I will find one that's about 1" dia x 3" long, but I don't expect that to happen - so some REALLY FRACKING LONG firebottles are probably in my future. In addition, one of my girlfriend's roommates suggested UV LEDs, and use some blacklight reactive paint to simulate the glowing. Since this is for a fantasy event (and because you can apparently get tubes at Guitar Center that glow something other than the characteristic warm orange of tubes from back in the day), this is a feasible option. He had also noted that the use of heavy copper wiring to simulate the anode and cathode would add to the archaic flavor that is appropriate to steampunk - so the "anode" would do this, while I'd be doing more of a rigid support system and finer gauge wire for the "cathode", for reasons that will shortly become obvious. For plugging them in, he had also suggested something where I could get a "twist-lock" motion going.
A quick search on Ebay turned up a few different suppliers of UV LEDs, and general rule I'm seeing is that they're 3.3v/20 mA, and require a limiting resistor if the power supply is >4W. Most efficient use, as near as I can tell, would be to wire the LEDs in parallel rather than series, so that I can actually get a rigid support for the LEDs and not have to worry about the "cathode" collapsing inside and simply looking funny.
This is where I'm stuck - the general schematic. I have, in short, forgotten Ohm's Law, as I've been way out of practice since something like 1998.
I have two plans for the wiring, keeping in mind that the guts of each "tube" would be wired in parallel rather than in series, and would contain two or three LEDs each.
The following, in short, are the design plans:
1) wire the power supply as a central power supply, and run series wiring throughout the assembly, additionally wiring to avoid what I call "old christmas light string syndrome" - that is, one burned out bulb (or, here, not connected tube) breaks the entire circuit and accordingly kills all lights (upshot: I can plug and unplug tubes on the fly for effect)
2) Again with the central power supply, but wire all sockets parallel to the supply (cures the OCLSS problem as above)
3) Wire each "tube" with its own power supply (probably a series of watch batteries, or a pair of AA batteries, but creates multiple switches - one for each tube)
On this note, for the plug end and any additional gribbles, I can probably work with ksharra for other ideas. But that aside, here's where I have the questions.
1) Which of the above design options is most optimal for the general power feed?
2) Given that I'm soldering at least two LEDs for each tube, how do I calculate resistance and power requirements?
3) In the case of 2, where do I solder the resistor?
4) If series soldering of the LEDs is more efficient, what would the schematic look like?