Dan's Data letters #163
Publication date: 25 March 2006.Last modified 03-Dec-2011.
Hello, officer!
A colleague of mine intends to start a horticultural business that would involve growing plants from seed under artificial lighting. My colleague has been told that the nature of the lighting required has a unique voltage "signature" that can be "read" remotely (presumably by the power supplier or a device fitted to the incoming supply).
My colleague has been told that detection of this voltage signature has been used to successfully identify buildings or locations used for growing plants of a particular kind.
I think this is all "big brother" nonsense and don't believe a word of it, but am curious if you have any knowledge of the matter. If such a thing is possible, then presumably it is also possible to use some form of filter to erase the signature.
Bob
Answer:
Great. One answer about LED growlights, and suddenly I'm Dan's Pot Tech Emporium.
Look, was it the picture...?
(As an aside: A while ago, I improved upon the semi-famous nosemonster picture, by exploring the ability of my rather popular Peleng fisheye to focus very, very close, when you unscrew the lens somewhat from its M42-to-EF adapter. It's the equivalent of using a very short extension tube. The result was, in one sense, glorious, but in most senses, utterly horrifying. I, therefore, implore you to at all costs avoid clicking here. I recommend even more strongly that you not set that image as someone's desktop wallpaper.)
On the subject of the mystic electronic signature, though - that's actually not out of the question. Various high-tech gear puts some signal out down its mains cord. Spooky types can apparently do nifty tricks like reconstruct what a laser printer's printing just by monitoring its input power, a stunt not entirely unrelated to van Eck phreaking. It could therefore be quite easy to find someone with a bank of short-arc lights by using a spectrum analyser on the power wires or, perhaps, just a directional antenna to look for simple RF noise. Ordinary filament lamps have no signature, but discharge lamps of various flavours have driver hardware that can, in theory at least, be detected.
The more stuff there is on a given circuit, though, the fuzzier becomes the signature of any given device.
A normal residential street will often have only a third of its houses on any one of the three phases offered by the wires up on the poles. Go for a walk and look - you'll see that the tapped wire changes as you go from house to house. So zeroing in may be easier than you think. But, the imaginings of unmedicated schizophrenics aside, Men In Black still aren't going house to house hooking up test probes to the power lines just on spec.
This could still provide a handy-dandy way for cops to check on premises they're already suspicious about without getting a warrant, though. Local laws and/or police lack of concern for them permitting, of course.
To my knowledge, they're seldom that cunning about it, though, because they don't need to be. If a normal-looking house is drawing a peak of several kilowatts in a slow grow-light-like cycle (easy enough to determine by just hooking a clamp meter around the power line), or if it just lights up like a star when they peer at it through the IR camera on the helicopter (the vast bulk of the energy that goes into your lights eventually leaks out of your house as heat, by definition; I suppose enterprising loons could use some of the waste heat to warm the swimming pool or something), that's enough. No power line snooping required.
(And then they come calling, and possibly find a nerd running his personal Pixar render farm, or someone with a big pottery kiln. Woe betide that last dude if he ever made a terracotta bong, though.)
Many serious urban agricultural criminals just steal their power, by running their own tap from somewhere inconspicuous, and/or modifying or plain old bypassing their electricity meter.
This is, of course, not the safest thing to do, and can be a pretty serious crime all by itself, and your power draw will still show up somewhere; people do tend to notice when a street's aggregate draw considerably exceeds what all of the meters say put together.
I dare say it is possible to, at least, mask the power signature from a bank of lights. A big old power conditioner (see a not-so-big one here) would probably do it quite nicely, and also protect any delicate lighting power supplies from surges and spikes.
(It'd also give you protection from brief power cuts - a ferroresonant conditioner isn't a proper standby supply, but it'll still let you ride out an interruption of a second or two without clocks saying 12:00 and computers restarting.)
I'm not kidding about the "big" part, though. A power conditioner for a several-thousand-VA load is likely to resemble, in size and weight, at least a bar fridge full of bricks. Possibly a chest freezer.
wheeeeeeeeeeePING!
I have a question regarding batteries for use in an external (hot shoe) camera flash. The flash takes 4 AA-size batteries and can accept alkaline, lithium, and NiMH rechargeable. I'm strongly considering lithium batteries because they will provide the most number of flashes, and also because I don't want to mess with the hassle of rechargeables since I won't be using the flash nearly as much as the camera itself. There will probably be months where I won't use the flash at all.
Will the lithium batteries retain their great shelf life if stored in the flash, or should I remove them from the flash when I am not using it?
Bob
Answer:
Yes, lithium batteries are a good option for you. Their nominal capacity is only marginally higher than that of alkalines,
but they deal much better with heavy loads, like a recycling flash.
Lithium batteries won't recycle the flash terribly quickly, mind you; they're faster than alkalines, but slower than NiMH or NiCd rechargeables, because the rechargeables have very low resistance and thus much higher current capacity. This is obviously an issue for professional users, but it can also matter for amateurs; if a decently meaty hotshoe flash fires at full power, you can easily be waiting 15 seconds for lithium AAs to recycle it (or as much as 30 seconds for alkalines). That's a long time to make people hold a pose.
Also, rapid shooting can overheat lithium cells, and they'll then automatically cut out until they cool down, leaving you with no flash at all for a while. Unless you take the cells out and blow on 'em or something, the flash will probably be dead long enough for you to reach the firm conclusion that you've blown the thing up.
You do need to get pretty enthusiastic with your machine-gun high power paparazzi flashing to have this problem, though. If you do, you may also find yourself wondering where that funny burning plastic smell's coming from. It's the diffuser on the front of the flash, melting.
NiMH and NiCd cells get pretty warm when they're heavily loaded too, but they're not full of highly flammable metal and so don't need internal thermal cut-outs.
And yes, they'll last as long sitting unused in a flash as they will sitting unused on the shelf. Lithium batteries, where available, are a great choice for standby purposes in many seldom-used devices. Rechargeables are a really bad choice, because NiCd self-discharges fairly fast and NiMH is even worse. Mind you, lithiums can be a lousy choice for a filament-globe emergency flashlight, since their higher terminal voltage kills bulbs. Some LED flashlights also don't like lithiums, but most are fine with them (and, so far, you can only get lithium versions of AA and AAA cells, not any of the larger "standard" batteries).
Non-rechargeable lithium batteries also, by the way, keep working at very low temperatures - their nominal minimum operating temperature is minus 40, where Celsius and Fahrenheit cross over. I talk about this here, in case you care.
If the neighbours complain, they're werewolves
I'm a grad student looking to build a speaker with a very high frequency response (above 20Khz). Thus far I have discovered that the most critical (and probably most expensive) component will be a tough light weight tweeter with sufficient stiffness to vibrate at higher frequencies. However, the only place I have found that sells such a component is pioneer electronics starting at around $1000. A bit out of my price range. The tweeter they sell is made out of beryllium. I understand that diamond would work well also. However, I'm looking for one that won't cost me up the wazoo to purchase or build. It will need to go at least as high as 40Khz and preferably as high as 100Khz. What other materials could it be made from?
Ethan
Answer:
Well, it's easy to find piezoelectric transducers that can go to 40kHz and beyond; medical
ultrasound goes up into the megahertz.
The response chart for transducers not made for audio work is probably up and down like a yo-yo, but high tweeter response is often pretty freaky, so they might still work.
As regards materials, I'm definitely thinking ceramic, here, though I'm no materials science dude and so may be way off. With ceramic you could at least cast and finish the part easily enough, and not be trying to machine or acid-etch or beat thin then form or something to make a super-thin metal dome.
Pyrolytic graphite? Just thinking out loud, here.
Corundum? Lab-made rubies and sapphires are very cheap, though good luck machining the stuff. Still, if you got a circular cabochon and just hollowed it out from the inside with a diamond cutter or something (laser erosion?), I suppose it could work.
I think the problem with any cheap option is that you'll end up with a ridiculously fragile dome if you make it as light as you want it to be. But as long as it doesn't shake itself apart, that ought to be an acceptable limitation. There may well be some wonder polymer and/or resin out there that'll solve your problem, though. Heck, you could inflate an otherwise floppy material with highly pressurised gas to make it stiff.
There's also, of course, the old plasma arc option. Unfortunately, most of them seem to need a supply of inert gas (to stop the arc filling the listening room with poisonous ozone and nitrogen compounds...), but helium's cheap enough.
Huntin' Hg
Re getting hold of mercury - I remember tons of globs of mercury in old thermostats... the ones with the bimetallic spirals of metal.
Might be a bit hard to find, maybe a friendly HVAC gent will save you some.
They actually are cool to play with, incidentally. They make great "level" switches and such. Tell them you are building a "battlebot".
Phil
Answer:
Two problems with that:
1: The mercury switches in thermostats are just standard mercury switches, with only a large drop of mercury in them - not really enough to make them worth scavenging. You only need a drop of mercury to make a switch, of course.
2: I live in Australia, where central heating is a rarity. Air conditioning's common enough, but wall thermostats aren't.
I live up in what-passes-for-mountains now where central heating is more common, but it's a late retrofit to most houses, so they often have digital controls.
To the moon and back on a AAA battery
Parallel Path Electromagnetic Motors - another one to add to the list.
The idea of conservation of energy is just a theory. Just ask the Kansas school board?
Dan
Answer:
The Open Source Energy
report seems
to be what's put the "even over the 100% barrier (?)" part into the description of these motors; I can see no mention
of any such thing on the site of Flynn Research, who're actually making
the things. Maybe Flynn Research are cranks, but they don't seem to be trying to sell
an
over-unity
scam.
If shaping of magnetic flux with permanent magnets really can make motors more efficient and give them more useful torque curves, then that's great. Sure, it's no big deal to get a modern brushless motor to run at 90 to 95% efficiency, but for practical applications in things like electric vehicles it may dip significantly lower at times, while the transmission's trying to keep up with unexpected terrain and throttle inputs, or something. And efficiency remains a limitation to motor miniaturisation; you can get outrageous amounts of power out of quite small motors (cases in point: All kinds of extreme radio control applications, whose protagonists will go to jail forever if cruelty to motors and rechargeable batteries is ever made illegal), but if something the size of your fist is producing six horsepower at 92% efficiency, it'll be getting very hot very quickly. Push efficiency to 98% and even if you don't care about getting that much more run time, your motor should run cool enough for continuous operation, with a bit of forced air cooling.
Of course, with 114% efficiency, there could be far greater problems. Leave the handbrake off and the car may drive round and round the world forever!
Solid smoke
Aerogel is cool, but too expensive for me to play with. Would you please ask for donations for an aerogel article and then do entertainingly geeky things with a few chunks? I'd donate.
Joel
Answer:
Not a bad idea; I've been checking into the availability of worthwhile-sized bits of the stuff for some time now,
not least so that I could take my own versions of those
pictures-that-look-like-Photoshop-jobs.
Unfortunately, as you say, you still can't get pieces bigger than a matchbox for anything resembling a reasonable price United Nuclear are a representative retail supplier, and they're charging $US125 for a matchbox-sized piece. You won't be doing a lot of frog-one-side, oxy-torch-the-other photography with a piece that small.
Incidentally, this may well make aerogel the world's most expensive commodity by weight, not counting illegal drugs (the street price per gram of LSD is astronomical, given its tiny dose size) and really exotic stuff like obscure radioisotopes, experimental drugs and so on.
Saffron is often quoted as taking that prize, and it's certainly a much more commonplace substance than aerogel, but it only costs about six US bucks a gram. Silica aerogel's got a density of around three kilos per cubic metre (if we come up with a solid below about 1.2 kg/m^3, we'll have to tether it down), and costs $US125 for... oh, 20ccs at best.
That's about 60 milligrams. $US2083 per gram, less whatever discount you get for buying bigger pieces, if indeed you get one at all.
