nelson_p@apollo.uucp (02/29/88)
Paul Dietz summarizes the options discussed so far for my electronic time capsule rather nicely: >Power sources: >Batteries Get serious... >Solar How do you make sure that the cells remain exposed? I was planning to locate the devices above tree line in the mountains, trying to find flat places (minimize effects of avalanches) well off the beaten path to avoid casual hikers from stumbling across them prematurely and also to take maximum advantage of the pipsqueak RF signal. During the winter they would probably be covered with snow. Anywhere else they would probably get covered with vegetation. I assume that the output of the solar panel will go down over time due to dirt and dust, degradation of the cells and darkening of the covering (tentatively Lexan XL, which is UV-stabilized) due to UV exposure. My prototype 6x10" panel yields 6 volts at about 150 mA in sunlight. However, my very low duty-cycle, gated-bias transmitter produces a power output of 200 mW (short pulses) with a steady state current drain of only about 10 mA. A small bank of polyester caps is charged from the cells to provide a relatively high current pulse for the transmitter. The CMOS oscillator and control circuitry draws less than 1 mA. I assume that polyester capacitors are more likely to last out the years than electrolytic ones. >Nuclear Not feasible for the the basement project... >mechanical (i.e. tides, wind, etc.) subject to failure. >thermal Very good reliability, low power output. If I can get a reliable 1 or 2 mA for the control circuitry and 10 mA when it 'wakes up' to transmit, I would be happy; what do you have in mind? > >Timing sources: >crystal mechanical reliability problems. >rc reliable, accuracy ~1% fairly easy to achieve But this requires that the system have a 24 hr source of power to save the count from one day to the next. >mechanical won't last if built in a basement... >chemical (clock reactions, or how about an Hg filled > speaker cable, with a copper plug that gets dissolved > away, and allows the mercury to close a circuit?) > Reliable, but timing accuracy is poor. >daylight Same problems as solar power... >thermals Could be a possibility since unit would not have to be > exposed, and earth/water (depending where you put it) > give a nice slow time constant. (these effects are both > daily, and yearly...) > >As for putting it underwater, I'm sure it is possible to find materials >which would be extremely resistant to things like barnical growth. >(How about teflon, aluminum, or maybe lead?) I know the Navy has had a big problem with this. I know they've tried aluminum without success and are currently using an anti- fouling coating that is very controversal due to its environmental effects but, in any case, it has to be re-applied every so many years. Anyway, how would I get the antenna up where it could be heard? --Peter Nelson
phd@SPEECH1.CS.CMU.EDU (Paul Dietz) (03/01/88)
In article <3a935422.44e6@apollo.uucp> nelson_p@apollo.uucp writes: >>Solar How do you make sure that the cells remain exposed? > > I was planning to locate the devices above tree line in > the mountains, trying to find flat places (minimize effects > of avalanches) well off the beaten path to avoid casual > hikers from stumbling across them prematurely and also to > take maximum advantage of the pipsqueak RF signal. During > the winter they would probably be covered with snow. > Anywhere else they would probably get covered with vegetation. The problem I see with this scheme is that it gives you lots of reliable power now, but not one hundred years from now, when it really needs it. > >>thermal Very good reliability, low power output. > > If I can get a reliable 1 or 2 mA for the control circuitry > and 10 mA when it 'wakes up' to transmit, I would be happy; > what do you have in mind? I'm no expert on thermo-couples, but I would think it would be possible to reliably generate a milliamp or two by digging a deep hole (say 10 feet or so) and using the temperature difference between the surface and the ground underneath. I tried to find some real numbers for the current you could expect from a junction, but I couldn't find any. Maybe somebody on the net could help with that one. As for the voltage, Nickel-10% Chromium versus Constantan will give you about 50uV/degree C per junction. I'm not sure what sort of temperature difference you could expect, but I suspect that careful placement could give something that averages near 10 degrees C. (In a dessert, for instance.) >>Timing sources: >>rc reliable, accuracy ~1% fairly easy to achieve > > But this requires that the system have a 24 hr source of > power to save the count from one day to the next. Yup. But by careful design, this can be made exceedingly small. One of those multi-farad caps made by Sohio could probably drive it for several days before it self-discharged. >>As for putting it underwater, I'm sure it is possible to find materials >>which would be extremely resistant to things like barnical growth. >>(How about teflon, aluminum, or maybe lead?) > > I know the Navy has had a big problem with this. I know they've > tried aluminum without success and are currently using an anti- > fouling coating that is very controversal due to its environmental > effects but, in any case, it has to be re-applied every so many years. Actually, I've had the extreme displeasure of applying bottom paint to a boat. The poisons that they put in don't seem to work particularly well. However, my own observations were that those beasties tended to prefer porous surfaces. I don't recall ever seeing any on glass surfaces, for instance. (By that I mean real glass, not fiberglass.) > Anyway, how would I get the antenna up where it could be heard? That's the best part of putting it underwater! You keep it submerged until the very last moment. Then, you can trigger a small chemical reaction that either: 1. releases some weight, allowing it to be floated to the surface; or 2.blows the tanks, which causes it to float to the surface. At the same time, you can use all this energy to expose your solar panel, so that when it gets to the surface, you have enough power to put out a serious signal. (Here's a real simple version: attach the weight with something that you know will corrode away. Encase the electronics box in another box that will explode (well, maybe just break) off when its internal pressure becomes very large compared to its external pressure. Fill it with something viscous, like Vasaline, under high pressure. At this point, use straps to hold it together that will decay quickly under water. Thus, when the weight breaks off, the whole thing quickly rises to the surface and the sudden pressure change bursts off the external casing, leaving your long protected electronics free to float to the surface and do their dirty work. You don't have to worry about a perfect seal on the outer box because everything still works fine so long as the leak time constant is long compared to the time to float to the surface.) Hope this is helpful. It's certainly an interesting diversion! Paul H. Dietz ____ ____ Dept. of Electrical and Computer Engineering / oo \ <_<\\\ Carnegie Mellon University /| \/ |\ \\ \\ -------------------------------------------- | | ( ) | | | ||\\ "If God had meant for penguins to fly, -->--<-- / / |\\\ / he would have given them wings." _________^__^_________/ / / \\\\-
cgs@umd5.umd.edu (Chris Sylvain) (03/02/88)
In article <3a935422.44e6@apollo.uucp> nelson_p@apollo.uucp writes: > > Anyway, how would I get the antenna up where it could be heard? I remember reading about a safety device for skiiers in avalanche-prone areas.. The thing was about the size of a small thermos bottle, and it transmitted a signal around 30 kHz, which was supposed to allow detection of the skiier even if buried by the entire mountain. No antenna per se, but I guess the requirement is that the search team be close enough to pick up the signal via a loop antenna (the transmit antenna is a loop as well). Hmm, that's going to be a problem.. If in the next century cables, fiber optic or RF, become the rule instead of wastefully radiating energy all over the place, then how do you know there will be anyone listening? Will the important frequencies of today (international marine distress, et. al.) be the same tomorrow? That is, what frequency (or band of frequencies) will you need to pick in order that someone will notice the little 'hello, I'm here. Come and get me.' signal in the first place? It would be a shame to build the time capsule, only to have no one notice the signal in a century... -- --==---==---==-- .. he went galumphing back. .. ARPA: cgs@umd5.UMD.EDU BITNET: cgs%umd5@umd2 UUCP: ..!uunet!umd5.umd.edu!cgs
jbn@glacier.STANFORD.EDU (John B. Nagle) (03/02/88)
Thermocouples may be a good idea here. Most of the time, there's a
useful temperature difference between the surface and 50' underground.
An old well shaft should be a good source for a few hundred mA.
Or you could drill your own small hole, which, for moderate depths, is not a
big deal. Put all the electronics in a cylinder, lower it into the
shaft a short distance, plug the shaft with cement, and cover it over
with dirt, and, in a reasonably undeveloped area, it is not likely to
be bothered for centuries.
Underground antennas, just below the surface, are used in military
installations where blast-resistance is desired. So even the antenna
can be underground, if you can store up enough power to get out despite
the losses.