Enthalpy

I proposed on Nov 13, 2017 that the wall material can influence the sound by an elliptic vibration around the tone holes http://www.scienceforums.net/topic/111316-woodwind-materials/?do=findComment&comment=1023070 and the effect of the toneholes' nature is consistent with my explanation. As explained by Miyazawa and many more, the toneholes of a flute as of some saxophones can be drawn from the body of soldered on it http://www.miyazawa.com/media-library/educational-articles/options/drawn-vs.-soldered-toneholes/ Drawing the toneholes from the body makes them even thinner, while soldered ones are thicker. The toneholes are the very item that stiffen the instrument against elliptic vibrations there, and the effect of soldered toneholes is said to resemble a thicker body: stronger darker sound. I won't be more positive than "consistent with my explanation", because: I never compared by myself the two hole constructions. So many sales arguments aren't justified! This feature is never an option within a flute series. It characterizes series that have more differences. Soldered tone holes are typically undercut, so the transition from the bore has a bigger and better controlled radius. This knowingly matters. The rim shape may differ. The flow gets easily nonlinear there, and angles would change the sound. I suggested to electroform complete flute joints, the body with the tone holes at once. With denser current or more time, including insulating masks, electroforming can produce locally thicker metal, for instance at and near the tone holes. Marc Schaefer, aka Enthalpy

Here's a table of metals ordered by reduction potential, aka standard electrode potential, aka standard reduction potential, from CRC Handbook of Chemistry and Physics, "Electrochemical Series" I've picked metals not very toxic and with a reduction potential not very negative: Mn can be electrowon. Nb, Cr, Zn, Ta are included despite their protective oxide layer because Zn can be electrowon, but alloys may worsen that; brass can be brazed but stainless steel is cumbersome. Only simple ions are listed. V+++ is missing. Polyatomic ions, with less direct reductions, would offer more potentials and enable metalloids. The cost per kg converts Usd, Gbp, Cny, Idr to metric units as of September 2018. I took troy pounds (20% difference). Prices are for multiton amounts of pure metal at stock exchanges except Ta in kg foil by a supplier; the cheaper metals depend more on the amount. The vertical bars extend from Ni, Cu, Ag by +-0.7V because CuNi can be electroformed - no better reason, and electrochemistry is tricky. Even CuZn (span 1.1V) can be electroformed. The oxidation number promises flexibility, but are there interactions? To the commonly electroformed Ni and NiCo, Mo may perhaps keep the corrosion resistance and increase the hardness and stiffness, Zn Cr Ta improve the corrosion resistance but hamper the solderability, Cr increases the stiffness and In Sn Pb Bi decrease it. In electroformed Cu alloys, Mn Zn seem difficult, but there may be tricks like polyatomic ions. Cu-Mn is a known vibration damper. Little Ni Co, preferibly together, and Ag too, harden Cu while keeping excellent conductivity. Sn alone is known to decrease abnormally the stiffness for bells while Mo Ru might increase it. Cu-Sn and Cu-Ni are usual hard alloys. Cu looks compatible with electroformed Ag alloys. Most flutes being just 92.5% Ag 7.5% Cu, could we make body parts at once, of sterling silver or other, with protruding tone holes, bent tubes and bells, for flutes, bass clarinets and so on? That would save work time and let strict woodwind manufacturers make metal bocals, boots and bells. Miyazawa's better Pcm alloy was rumoured to contain 65% Ag with Cu Au Pd; Au is doubtful according to the table but Cu Pd would be compatible, Ru Bi Rh Ir Pt too. ========== The table's three rightmost columns hint to relative resonance frequencies. Sound velocity looks like a plane compression wave in a wide solid. The oval frequencies are figures-of-merit relative to pure silver: sqrt(E/rho3) for bending at identical mass, sqrt(E/rho) for bending at identical thickness, having in mind the resonances of Nov 13, 2017 and followings http://www.scienceforums.net/topic/111316-woodwind-materials/?do=findComment&comment=1023070 Among precious metals for a flute, only Rh and Ru are stiffer than Ag at identical mass. Rh resists corrosion even at brazing temperature. I ignore their other properties, especially their vibration damping. I've added a line with Dalbergia Melanoxylon, the Grenadilla preferred for woodwinds, and taken 1/10th of the lengthwise 20GPa for want of its transverse modulus. It wouldn't be as thin as metal walls. At identical mass, grenadilla's flexural resonant frequency is 2.2 to 3.5* higher than wind instrument metals, and usually it's even thicker. Marc Schaefer, aka Enthalpy

I've already described the use of Darrieus rotors. They have run for decades in Québec. I've already suggested how to get the water onshore. The market for water in Capetown is to avoid a shortage. 3 years supply with 4 months reaction time is a permanent solution. Again, "there are problems" does not mean "it cannot work". I should like to remind that desalination was absolutely unrealistic two decades ago. And that not conceiving or reading all the solutions now does not make a project unrealistic. All the items around you are full of problems for which you ignore the solutions.

You meant fossil energies get scarcer and more expensive, I suppose? I've already suggested using wind to move the icebergs. Renewable energies get cheaper over time, very much so. Presently they're cheaper than nuclear electricity, and for the iceberg they make sense over a few trips.

Why should moving icebergs get more expensive over time? Technology gets cheaper as people make efforts for that. Never tell "impossible for humans" and even less "because I don't see how". A car for each family, a mobile phone that fits in your hand... all this was impossible before people made it possible. Even desalination was completely unthinkable on a commercial basis before people invented reverse osmosis. I wish all resources were as little finite as that one!

If wasting water like European agriculture does, desalination is prohibitively expensive, even by reverse osmosis. Up to now, desalination can feed houses where water is too scarce, typically in some Spanish touristic cities. Water-saving growing methods in greenhouses may afford desalinated water. But can it grow wheat, or even maize and soya? In all these projects I see tomatoes, more expensive per kg than cattle feed. This may make the difference. I'm surprised that they use concentrated heat for desalination. Reverse osmosis is so much more energy-efficient that it pays for solar cells. Not much information https://en.wikipedia.org/wiki/Sundrop_Farms http://www.sundropfarms.com/ Well, history tells me that new technologies emerge.

Hi Arc, Ken Fabian and the others, thanks for your interest! The biggest supertankers weigh(ed) 1 million tons. The iceberg pushed from Antarctica to Australia or South Africa is to arrive with 60 million tons. I had wondered if the tanker was to collect just the molten sweetwater during the trip, but even for that it's too small. A tanker would move 20* faster than 1knot, so trying a fair comparison: 60 million tons delivered need 200 trips by a 0.3 million tons tanker(s), or 10* more travel time. At 20 knots, the tanker consumes also more oil per hour than when pushing strongly at 1 knot. An other difficulty: tankers are dirty. They carry crude oil and seawater. Sweetwater would rather need a new design, capable of disinfection, and with a nicely profiled hull - but then it costs 100Musd for 0.2 million tons. Maybe this will exist if we transport sweetwater regularly in some future, but for a demonstration it's too expensive. The team considers probably leasing an old tanker. I made long comparisons about the fuel consumption, but fuel is rather cheap in this enterprise. Here the iceberg would be tabular, a shape usual in the Antarctica. I suppose they result from snow falling on the Ocean rather than from glaciers flowing to the sea. A 1km*1km*100m iceberg won't capsize. However, it may very well break in two big parts or lose some side fragments. I ignore what the team has foreseen. Pushing the berg means being close to it, which is pretty dangerous. Moore lines with several MN tension are always damn dangerous too. Maybe some parts built weaker by design, for instance at the anchors or at the tugboats, make the operation less dangerous. And some moore line materials don't fly around when they break - at least the supplier claims so. Very interesting indeed! I didn't know that ice trade had become international at some time. This could be a commercially more viable option: sell ice for cold, not for sweetwater. In Australia they could save electricity wasted in air conditioners. Certainly more value than water, and one century back it made sense with less efficient methods. Whether this trade pays for insulated boats? It needs investment to equip customers with ice-loaded air conditioners. The team considers anchoring the iceberg off the coast. Remember it's about 1km*1km, and far too deep for a port. It would be "classically mined" (but how cleanly??) and the molten water brought by tankers somewhere - to the port or to the end of a water pipe, which can be some construction at sea. I'd rather imagine to extend the pipe to the iceberg. The water must be fed in the existing distribution network, which has some storage capacity anyway, but may need to pump. Caring inhabitants of a dry country may use 100L/day, so a 1M city uses 0.1M tons a day, which is less than the iceberg's melt rate. 800 days consumption may well exceed the existing storage capacity. Or should the geotextile envelope be closed, so the meltwater is stored at the iceberg and brought to the consumers as needed? I've seen no details, and even less estimated costs, for this part of the operation. Neither am I. But I'm sure that problems are solved only by tackling them. The comparison with reverse osmosis wakened my interest. When speaking of water price, we should distinguish the use. For drinking water we can pay 1€/L. For general house water it's few €/m3. Agriculture as I know it presently in Europe needs water much cheaper. In the case of Cape Town, house water was getting scarce recently, with a risk of supply breaks.

This other arrangement of keys at the boot of the baritone oboe of Sep 16, 2018 has advantages: Pyramidal frames hold the boot firmly together so the keys fit precisely. Transmissions from the main joint are easy and the disassembled main joint protects them. The keys can be light and still stiff. The keys are easily synchronized and adjusted. The toneholes can be tilted so the covers move perpendicularly to them. Not displayed here. With easy adaptations, the arrangement of keys applies to similar instruments if designed with a boot. This includes the English horn, lower tárogatók, saxophones, rothphones, sarrusophones, optionally with my automatic cross-fingerings. Marc Schaefer, aka Enthalpy

The free jet downward a propeller makes a conus with small divergence. If the jet impinges on the iceberg, 790m wide and 100m deep, the thrust is annihilated. The ice may also melt faster, or the protection geotextile damaged. Several towing boats could be linked with a transverse mooring and pull a bit to the sides. If they are 100m apart, >1300m forward, and pull 20° outwards, their jets flow past the iceberg's sides. The forward component drops as cos(20°)=0.94. The jets must cross at different depths. Or the jets can be 5° downwards, as is often the case, and the boats >1200m forward, so the jets pass below the iceberg. Limits: the jets diverge, warmer surface water isn't desired under the iceberg, and surface water may buoy too strongly. The boats could have deflectors added behind the propellers to direct the jet down or by halves to each side. Sidewise cos(25°)=0.91 and 850m forward suffice, or downwards cos(15°)=0.97 and 400m forward. Affordable modification. If the boat has several propellers with one rudder blade each, decoupling the blades adds fewer parts and keeps the manoeuvrability. The simpler way is to push the iceberg instead. Upstream a propeller, the flow converges and accelerates steeply, so an obstacle doesn't annihilate the thrust. ========== Adapted propellers As seen on the previous sketch, a bigger propeller with V~v takes less power for the same thrust. That's why subsonic airliners have wide double-flux turbofans and helicopters a huge main rotor. The limit is when downstream V is but bigger than upstream v and P~F*v. At 25 knots, the Maersk Triple E nears it, but at 1 knot it's inefficient. So can the propellers grow? Few m/s downstream are far from the cavitation limit, so a special propeller's blades could be narrow like at aeroplanes rather than very wide and overlapping. This saves material but demands precise machining. The blades must be thick to resist bending and rather of steel. Whether existing boats can accommodate bigger propellers? Such adaptations look expensive. Build a big marine wing or centreboard that makes the big force by lift, let a boat or propeller move this wing sidewise. Beginning at 10m depth, it can be about 10m/s fast and pull ~0.3bar, so 60m*6m create 11MN. If L/D=10 it consumes 11MW, saved one half. Can the wing push the iceberg, or only pull? Have a wing at each iceberg's side? Add a big nozzle around the propellers. Without cavitation, a convergent increases the suction force. A divergent makes a back force but the convergent compensates it, and at the throat the propeller speed is better for it and the motor. No miracle, but the adaptation seems affordable. Combines with the jet deflectors. Build (strong) marine parachutes, pull them from the sides of the iceberg. Boats then pull them forward again by their centre so they collapse and move easily. To achieve 2*5MN 80% of the time, two D=100m parachutes drift at 1.1m/s or 1.5m/s versus the iceberg so the winches consume mean 12MW together, saved one half. ========== Fuel cost Some 7000t fuel for the tanker and tugboats unoptimized option cost 2M€ only. That's a small part of the operation. Saving on that line is more an environmental argument. Marc Schaefer, aka Enthalpy

Trying to reconcile the power, thrust and drag because the Bbc's paper reflects the innovators' natural optimism... ========== Thrust and power Instead of the undefined "supertanker with over 20 000 horsepower" I took data from the well documented Maersk Triple E 210 000 t container ships wikipedia two 32MW engines and D=9.8m propellers push them to 25 knots. An optimistic jet diameter and zero losses let deduce 14.2m/s downstream and 2*2.45MN thrust at v~12m/s before the propellers. If (?) the engines provide 20% more torque at slower speed, the thrust too increases to 5.9MN from 6.2m/s downstream and 2*9.1MW at the shafts. Over 90 days at 60% efficiency, 2*9.1MW consume 5400t fuel, and the tugboats booze a bit too. By the way, we can deduce Cd=0.072 for the Triple E, impressive. Due to its smaller propellers, the tugboat Abeille Bourbon pulls only 2MN with 16MW peak shaft power fr.wikipedia three such tugboats could double the force, but above all they manoeuvre better than a tanker. ========== Drag The tabular iceberg shall be 100+10.5m thin (possibly by melting), 1000m long and 790m wide. Broken edges might achieve Cd=0.8. In decent weather and with luck it could be pulled lengthwise, so at the lower 0.4m/s the drag is 5.1MN. 5.9MN+3*0.7MN with the tugboats achieve 0.5m/s. A 1.5km long iceberg, narrower or thinner, would attain 0.6m/s. This speed doesn't look critical. 0.4m/s cover 3000km over 90 days, apparently more than is needed. Maybe the speed is necessary for few days only. ========== Updates to wind propulsion 21MW including the tugboats need four 6MW wind turbines. 8MN need fewer Darrieus rotors than 14MN. Thanks for you opinion!

E and H. Or psi if you prefer, suits few particles better.

Hi Swansont, thanks for your interest! I don't know of usual examples of water delivery by train. I take it as an example of transport that saves energy.

Critics reproached the 4 000 t fuel consumed to deliver 4 000 000 t freshwater to Canarias, or 1kg/t. But this consumption is litte! Compare with a train. It picks 70t*50 freshwater from a source, and a 40% efficient 4MW engine pulls it in 10h over 500km to the users. It consumes 360GJ heat from 8t fuel to deliver 3500t, or 2.3kg/t: a bit worse, and it needs a nearby source connected by railway. Compare with seawater desalination by the most efficient process, reverse osmosis. It needs about 60bar to work wikipedia and the engine and pump may be 30% efficient, while the brine's pressure can be recycled: 4 000 000 t freshwater need 80TJ heat from 1800t fuel, or 0.45kg/t, slightly better. ==================== Is the new route better? Antarctica to Capetown is easier than Newfoundland to Canarias. The authors mention a >20 000 hp tanker and three (6000 hp?) tugs, say mean 20MW combined shaft power over 90 days. At 50% efficiency, that's 310TJ heat from 7 000 t fuel for 60 000 000 t freshwater now, or 0.12kg/t, which outperforms reverse osmosis. ==================== Are there better energy sources? Fresh meltwater and seawater can produce energy as they mix. This osmotic power is worth some 120m water height. wikipedia So if a 90 000 000 t iceberg melts to 60 000 000 t, the maximum available energy is 35TJ. Too little to supply the tugs - and osmotic power has negligible efficiency at present prototypes. Meltwater is 0°C cold and the surrounding Ocean maybe +8°C or even less. Make a thermal engine? At 3% Carnot-limited efficiency, 30 000 000 t water bring 58TJ work as an absolute maximum. Too little again - and existing oceanic gradient power plants are far from Carnot's limit. Let's forget about solar energy on the Southern Ocean. Wind energy sounds better in the roaring forties, furious fifties and screaming sixties. 30MW peak tug power can depend on the peak power of five 6MW turbines on the iceberg. Their D=154m rotors cumulate 9hm2, while a 700m*700m*178m iceberg (80 000 000 t at 917kg/m3 in 1013kg/m3 seawater) exposes 17m to the air, or 1.2hm2. In 25m/s wind, the 39MN drag bring the drift to 0.8m/s versus the Ocean, whose current is already faster wikipedia Looks feasible. As the turbines cost a few times more than the fuel they save in 90 days, they should serve for several icebergs. Sails convert more directly wind speed to vehicle speed. Tugs were to achieve 0.5m/s. To produce 14NM to the North and 7MN to the East in 20m/s westwind, sails must total 6.4hm2. Masts 50m high, big sails and jibs 20m long bring 0.1hm2, so it take 64 of them. Not cheap. Darrieus rotors can combine both. Sails are as big as the wind area they exploit, but turbines move faster than the wind so they can be slimmer than the swept width wikipedia (picture gratefully pinched there) and the Darrieus rotor can, if driven accordingly, produce a force strongly angled to the wind instead of power at the shaft. This rotor isn't so frequent but real-size demonstrators were operated in Canada. Marc Schaefer, aka Enthalpy

Hello dear friends! John Isaacs proposed it in the 40s for California. Peter Wadhams, Olav Orheim and Georges Mougin made studies for Saudi Arabia in the 70s and the Canarias and the UAE in 2010s. Together with Nick Sloane, they now have plans to tow icebergs from Antarctica to Australia or South Africa bbc.com Antarctica provides bigger, flatter and more sturdy sweetwater icebergs than the Arctic region do. Australia and South Africa are accessible by pushing the icebeg a bit North and letting cold oceanic currents do the rest wikimedia.org gratefully pasted here antarctica.gov.au This makes the current plans less difficult than targeting the Canarias or the Persian Gulf. These plans are far-fetched and nicely megalomaniac, so I perceived a golden opportunity to add my own frenzy. Not only for Australia or South Africa: comparable dry locations with cold currents linked with the circumpolar one are Namibia, the Chilean and Peruvian coast including the Atacama and Namib deserts.

After the positron-electron pair annihilates, we still want the energy to be present before it's absorbed by an other particle, for instance at a detector. Some effects of this energy, like its gravitational attraction on massive bodies, are perceivable before the photons are destroyed. Energy carried by the photons: if you wish. But it's not a reason to dismiss the field.

I put thoughts in the liquid hydrogen tank because compressed hydrogen needs a tank too heavy. Also, liquid hydrogen is less dangerous than compressed one.

I am. Circuit analysis tends to be very simple. Usually I need no "theorem" nor "law" at all. just currents, voltages, impedances. Exceptions are rare. Or maybe I've forgotten that the computations I make bear a name. But you shouldn't imagine that electrical engineering is only circuit analysis. Software: forget about that. Computer: no. And: if you imagine that a software or computer replaces the designer's understanding and ability to compute, you will fail. Beware: your question suggests that you imagine that software replaces knowledge, but a designer that understands his stuffs doesn't need any software. Electronic circuits are easy to compute by hand, and just a pocket calculator is faster and safer than any computer simulation. In fact, at least a friend and me often cable electronic circuits without drawing them in advance, and compute the components mentally. That's very far from needing software.

Hi abrogard, there is no simple answer. It depends on how your material hardens during the deformation. Usual stainless steel needs more force than mild steel. Knowing the yield strength prior to deformation does not suffice here. A first approach would compute a bending moment from the 'deformed!) yield stress and the thickness, then compare with a lever length deduced from the bending radius, and compute a force, not forgetting the symmetries and factor-of-two. Alas, this simple model is known to fail, by too much for practical purposes. So you have to use empirical knowledge. For 1m width it takes a big force, usually by hydraulic cylinders. This force suffices to deform the bending tools (matrix etc) and make the unusable when the item to be bent is narrow, so your machine needs a force limiter AND a smart operator.

How fast the interactions propagate may still be an open question. I have still to think more at it, but when charges are near to an other and accelerate slowly, you can't use potentials retarded by x/c to compute the interaction. Just like the Earth-Moon interaction is much more instantaneous than delayed, as is known since the 19th century. At least common "knowledge" of electromagnetism is wrong here. How complete the understanding by experts is, I can't tell. ========== Whether an alternating current by dielectric polarisation radiates a wave, I feel it safe qualitatively. The "1" part of permittivity, equal to vacuum permittivity, does not radiate. The rest, or electric susceptibility, results from charge movement and radiates. And dielectric antennas serve in most cellular phones. But algebraically, with nice equations... That could be a remaining task, and it looks badly difficult. You know, write Maxwell's mess, and deduce logically what produces a radio wave or not. ========== This relates directly with the computation of antennas. We evaluate the far field by computing the vector potential A from the conduction current I or J but we don't include the vacuum's polarisation current e*dE/dt. This fits experiments. Though, both are equivalent to create curl(B) in Maxwell's thingy. So why only the conduction current? A direct application are vertical long-wave antennas. They get additional wires at the top to increase the capacitance there, so more current flows and the radiation increases. Though, this additional current returns to the ground through the added capacitance, and the return current compensates the additional conduction current. It works, why? ========== Does the electrostatic interaction have inertia? When, and for who? For me that's very unclear, for experts I can't tell. https://www.scienceforums.net/topic/85377-relativistic-corrections-to-hydrogen-like-atoms/?do=findComment&comment=990276 ========== The magnetic field of planets remains mysterious. For Earth we have a plausible model since the recent (!) VKS experiment https://hal.archives-ouvertes.fr/hal-00492266/document but for other planets it seems to fail. ========== As far as I know, the coherer still needs an explanation https://en.wikipedia.org/wiki/Coherer it were about time, as the effect was observed in 1835, the devices entered service around 1880 and were abandoned a century ago. Possibly more material science than pure electromagnetism, but who knows, as we ignore the explanation. ========== How does vacuum insulation (or rather the electrodes?) break down in a strong electric field? Probably not pure electromagnetism, but it would be very useful. Devices already use vacuum insulation despite we have no good models for it. Most people keep trying "field emission" again and again despite all evidence is against.

Hi Aleksey, you're right almost everywhere. The rising air parcel works against its neighbours. It loses internal energy, that's why its temperature drops. Note that without any losses, energy would have to go somewhere, and the parcel isn't different from its neighbours... (Probably your next question!) The energy difference goes into kinetic energy. If the parcel was warmer than its surrounding, it becomes faster as it rises. The acceleration can happen at the bottom, the top or all along the way, depending on the profile section of the tube. In such a case, where air packets exchange no heat with anything else, the work they provide (or receive) is the variation of their enthalpy. For air, which is almost an ideal gas under usual conditions, the enthalpy is fully determined by the temperature, or by the initial temperature and the initial and current pressure. The profile section will act on the pressure indirectly by imposing a kinetic energy and so on. It's impossible to solve locally, you must consider an distribution over the full height. One funny thing: enthalpy is not fully contained in the considered air parcel. The internal energy is, but the additional P*V is stored in the surrounding. Despite this, (T, P, V) or the air parcel itself suffice to define its enthalpy. An extreme case is a liquid, say in a hydroelectric dam: as a water parcel goes down to the turbine, its temperature and internal energy keep constant, but it receives V*(P2-P1) from the neighbour water parcels, and this energy converts to speed in the injector and to work in the turbine that slows the water.

You can't get a voltage with that setup. But that's tricky, as electromagnetism uses to be. Essentially, the moving magnet induces zero voltage in your closed loop as is. The setup nearest to yours is a homopolar generator, see Wiki. It has moving and immobile conductors, plus (uneasy) sliding electric contacts. More generally, it takes sliding contacts to produce a DC voltage from induction. A collector for instance.

Sorry it won't work. Capillary action can suck water in some tube with proper filling, but the water will be equivalently harder to extract from the tube's top. It won't flow spontaneously out of the tube. Donkey, wind mill, solar cells...?

You don't need two bodies to have energy stored in a field. If a star emitted light a billion years ago and you telescope catches it now, the comfortable law of energy conservation wants energy to have been stored in the vacuum in the meantime. An extreme case is when an electron-positron pair annihilates. At the time you catch a gamma ray, both particles don't exist any more. How is that energy stored? I ignore it. I don't even know what energy is.

Two variations on the computation by Timo (nice to see you), which change the 45TWh figure hence the cost. Generating electricity in one country isn't necessary and does not correspond to present-day practice. If the wind doesn't blow in whole Germany, it does in Scotland, Brittany, Aquitaine or Galicia. Electricity is presently transported, typically on such distances, as the market is continent-wide. There is no need to store an amount of (Germany's mean consumption) 64GW over 29 days as the 45TWh imply. Even if the electricity came from Germany alone, wind wouldn't stop for that long. Europe-wide, you won't have more than 1 day without wind. That would be 1.5TWh storage for the country. It wastes some electricity but is globally cheaper. The Powerwall costs slightly over 340€/kWh but is a small unit for houses. It is guaranteed for 10 years so it will last rather 20 years. https://www.tesla.com/powerwall The utility-sized Powerpack is hopefully cheaper per kWh https://www.tesla.com/powerpack https://en.wikipedia.org/wiki/Tesla_Powerwall#Powerpack_specifications "should" cost 220€/kWh. 1.5TWh and 220€/kWh cost 330G€ every 20years or 16G€/year, not 1350G€/year. This is affordable and much less than what the inhabitants pay for electricity. ==================== The other point is that batteries are only one solution. It's a mature one, already in use at substantial scale, but not necessarily the cheapest one. I have good hope that flywheels are cheaper per stored kWh than batteries. The store-restore cycle is more efficient from night to day, less efficient over a week https://www.scienceforums.net/topic/59338-flywheels-store-electricity-cheap-enough/ and Prof. Seamus Garvey's underwater bags look cheap too and has been experimented https://www.offgridenergyindependence.com/articles/3358/compressed-air-energy-storage there are few more ideas.

Hi Frank, thanks for your interest! Thermal engines tend to be less efficient than the fuel cells' 60%, but they improved quickly in the past two decades, and the difference is small now. As thermal engines are much lighter than fuel cells, the alternative must be considered, sure. 600kg less would sell 6 seats more, big difference. Hydrogen is difficult to bring to a combustion chamber. Injected liquid in a piston engine prior to compression, it freezes air's water vapour and possibly the carbon dioxide, and as hydrogen vaporizes, the extra volume to be compressed spoils the engine's efficiency. Injected gaseous and lukewarm in a piston engine prior to compression, the extra volume to be compressed spoils the engine's efficiency. Still bad. Injected gaseous after compression is as bad as before compression. Injected liquid after compression is the least bad option at a piston engine. Hot air components won't freeze, and this squanders the least power. However, it demands a damn strong injection pump, much worse than the difficult pump of a common Diesel engine, and the whole pump and circuit must work at 20K. Design is badly difficult AND operations get complicated, as the whole circuit must be cooled before start. In a turbomachine, hydrogen should be injected after air compression for the same reasons, but this demands a pump much more powerful than now, and again pre-cooling before starting the engine. I put there http://www.chemicalforums.com/index.php?topic=91121.msg325955#msg325955 (log in to see the drawings) and especially this message for hydrogen alone http://www.chemicalforums.com/index.php?topic=91121.msg333549#msg333549 some pumping cycles with the necessary power, and they inject the hydrogen already hot in the chamber, which helps stabilize the flame. Meant for ramjets and scramjets but fit turbomachines too. It's more rocket than aeroplane technology, and hasn't been developed for airliners up to now. Understand: long and costly. Simpler cycles are possible with turbomachines. Hot output from fuel cells: the 40% wasted energy must go somewhere, yes. My doubt is: how heavy is a heat exchanger to make use of this? Each time I tried, heat exchangers were too heavy to outperform alternative solutions. And if you can operate the fuel cell at a pressure higher than the chamber of the thermal engine that uses the waste heat, to inject the fuel cell outlet in the chamber without a heat exchanger, then you must first pump the air and the hydrogen to that pressure, which is nearly as bad and pretty complicated too. The rest is less of a problem, for instance hydrogen combustion is already known from the more difficult scramjets.