Video from the LRV - where to find?

[ka9q]

Regarding batteries, a major design choice is between primary (non-rechargeable) and secondary (rechargeable).

Though primary batteries aren't rechargeable, they have the highest energy densities of all batteries and they tend to have longer shelf lives. One of the lithium metal chemistries would be the way to go today. They have better energy density than the Apollo workhorse, silver-zinc.

Silver-zinc's energy density is roughly comparable to lithium-ion, but while the latter has hundreds of charge cycles most Ag/Zn batteries, if they're rechargeable at all, have cycle lives in the single digits (e.g., those in the Apollo CM).

The workhorse rechargeable battery for aerospace in the past few decades has been nickel-hydrogen. They're similar to nickel metal hydride except that the hydrogen is stored as a gas instead of as a metal hydride, making them lighter but bulkier. I think they're now being displaced by li-ion, but li-ion has yet to prove itself in the lifetime category.

There are no obvious winners here just as there aren't any in terrestrial applications. Each has its plusses and minuses.

Fuel cells are another good option, assuming the water can be recycled and electrolyzed back into hydrogen and oxygen. The cycle energy efficiency isn't great, but the service lifetime could be very good.

[cjameshuff]

OFDM is highly sensitive to doppler shift, which might make it a poor choice for general space communications. You can compensate easily enough if you know your relative motion, but...that brings you back to the localization problem.

Beacons limited to ground movement could be cheap micropower devices, so you could just use enough of them that there's always one or two in line of sight. Perhaps bury the sensitive electronics to reduce radiation and thermal issues, with only antenna and solar panel exposed.

[Allan F]

I just checked with a friend who knows about fuelcells (PhD-student). I wondered if heat energy in stead of electricity could be used to break the water down. That would provide a heat sink, but that didn't fly. I think we should go with lithium-based batteries.

What about inertial navigation? That should be possible to build quite compact with the new chip-size accelerometers.

[cjameshuff]

I just checked with a friend who knows about fuelcells (PhD-student). I wondered if heat energy in stead of electricity could be used to break the water down. That would provide a heat sink, but that didn't fly.

Thermal decomposition or high temperature electrolysis can certainly be used, but they don't provide a useful heat sink. They just make more efficient use of power sources than conversion to electrical power followed by plain electrolysis.

I think we should go with lithium-based batteries.

Lithium batteries are heat sensitive, short lived, prone to thermal runaway, and require lithium, which isn't readily available on the moon. Nickel hydrogen's got a good record. Nickel-iron is another good chemistry for robust industrial uses...heavy, but that's less of an issue on the moon, and they could eventually be made with easily accessible materials. They essentially don't wear out over multiple charge cycles, are reasonably efficient, and can last decades.

What about inertial navigation? That should be possible to build quite compact with the new chip-size accelerometers.

Even with very high quality accelerometers and gyros, it's only useful for short term position estimates, and tiny MEMS devices are far from the best available. A vehicle rolling, slipping, and bouncing across an irregular surface is about the worst possible use case for inertial navigation. It's useful for controlling the vehicle, but needs an external reference of some kind to compensate for drift, noise, limited frequency response, etc...

[ka9q]

OFDM is highly sensitive to doppler shift, which might make it a poor choice for general space communications. You can compensate easily enough if you know your relative motion, but...that brings you back to the localization problem.

Exactly right, which is why I proposed it only for terrestrial (lunarial?) communications. Communications with earth would use a traditional satellite modulation scheme like QPSK or MSK on a higher frequency band, Ku or above.

OFDM is highly tolerant of multipath with a reasonably small delay spread like you find in terrestrial communications, and that's why it has all but taken over mobile phones and WiFi, plus TV broadcasting outside North America. But that also makes it difficult to piggyback navigation on it. So maybe an independent ranging system using direct sequence spread spectrum would be necessary for navigation. I.e., something like a ground-based GPS or your beacon idea, with transmitters placed in surveyed locations around the site. You'd need at least three in view to get 2D positioning, or 4 to get 3D positioning. If the beacons are placed on mountainsides so they can see each other, direct optical links could be used for time transfer and baseline determination to avoid the need for highly stable clocks.

DSSS can certainly be used for communications too (my company made its name doing that) but it tends to chew up bandwidth when you need a very high data rate, e.g. for HDTV.

[ka9q]

Thermal decomposition or high temperature electrolysis can certainly be used, but they don't provide a useful heat sink. They just make more efficient use of power sources than conversion to electrical power followed by plain electrolysis.

There are several chemical cycles to produce hydrogen (and oxygen) from water with high grade heat: sulfur-iodine; zinc-zinc oxide; iron oxide; hybrid sulfur; copper-chlorine; and cerium (III/IV) oxide were the ones I could find easily on Wikipedia. Several include an electrolysis step that uses less electricity than straight room-temperature electrolysis.

These cycles might be a good use for concentrated sunlight, something that could be produced in great abundance on the moon, and provide a more practical way to store up electrical power for the long 2-week night than conventional batteries.

Electrolysis of exhaled water vapor and urine to produce breathing oxygen is already being done on the ISS, but they throw the H₂ away. That would be exactly the wrong thing to do on the moon, which is poor in volatiles like H₂ but rich in O₂.

Like any heat engine these thermal converters would require heat sinks, which would have to be radiation to deep space. How big they would have to be depends strongly on how cold a temperature they have to produce.

[ka9q]

Lithium batteries are heat sensitive, short lived, prone to thermal runaway, and require lithium, which isn't readily available on the moon.

I didn't say they were perfect, but if you're going to send them from earth they have the enormous advantage of one of the highest energy densities available.

There are a few other battery chemistries that look promising for bulk stationary storage, particularly sodium-sulfur if the corrosion problems can ever be worked out. (The problem isn't the sodium, it's the sodium sulfide compounds.) On earth they have the big attraction of using extremely cheap and plentiful materials, but I doubt they're as common on the moon.

I still think that in the near term the most practical power source for a lunar base will be nuclear, by far. Various solar thermal and chemical cycles look like they could be made to work, but seriously -- once you've got a working reactor, do you really need anything else? That leaves the portable power problem for rovers and PLSSes, though.

[raven]

Solar-thermal has easier self expansion possibilities, that's one advantage in my opinion.

[cjameshuff]

OFDM is highly tolerant of multipath with a reasonably small delay spread like you find in terrestrial communications, and that's why it has all but taken over mobile phones and WiFi, plus TV broadcasting outside North America.

A single communications and localization standard for spacecraft and ground vehicles seems desirable, but the multipath resistance is a good point.

You'd need at least three in view to get 2D positioning, or 4 to get 3D positioning. If the beacons are placed on mountainsides so they can see each other, direct optical links could be used for time transfer and baseline determination to avoid the need for highly stable clocks.

That's if you're doing full GPS-style localization. For path marking, single beacons with some amount of overlap could suffice. If you have directional antennas or prior rough location information to break the symmetry, you can also get full 2D localization with just two beacons, 3D with three (or two and a good elevation map). The same beacons could be used in higher numbers in locations where greater precision is required, but reducing the number of beacons you need to plant across the landscape seems worthwhile.

Another possibility is visual localization. There's no weather, no seasons, no plant life...the moon is ideal for a system that just recognizes the surroundings and infers its location from landmarks. This would be difficult to do on a portable device, but the needed cameras might be built into suits and vehicles, and would be useful when surveying large areas.