Le petit défouloir des pro et des anti Tesla

R2D2 a écrit:Mes remarques portent sur la rentabilité de l'entreprise Tesla, pas de son cours en bourse dont je me moque totalement

fxmx86 a écrit:C'est risible : Tesla a les meilleures marges, enviées par tous les constructeurs

fxmx86 a écrit:
Aucune discussion possible : de l'incohérence et juste un narratif négatif à faire passer de ton côté

Shockingly Virtuous Boring Tunnels

Thank you @boringcompany and @elonmusk for solving the traffic problem and saving a LOT of energy.

Some people, when they see what the Boring Company are doing, decry the system as cars in tunnels and they proclaim that trains are much better.  But, nay nay - the shockingly tasty main course is yet to come.

In this post I am going to look into the tunnels to find a whole train of first principle improvements that add up to shockingly virtuous boring tunnel systems.  

For a start, electric cars are already something like 3 times more efficient than ICE cars.

If my clumsy physics and calculations below are anywhere near right then mass transit in Boring Company tunnels will be at least 3 times more efficient again.  Perhaps someone cleverer than me at physics and mechanical and more stuff might want to calculate some better numbers.  What this means is, quite remarkably, that mass transport in tunnels will approach TEN times the efficiency of ICE cars running on surface roads.

The Boring Beginning

At the moment the Boring Company employ mainly Model Y cars with drivers to transport passengers around the LVCC (Las Vegas Convention Centre).  That system is already being expanded to the closest hotels by the LVCC and will eventually have something like 70 stations all around the city in a total system referred to as the Vegas Loop.

The advantages we know about already are that the cars will be able to travel point to point without traffic intersections or pedestrians or other traffic getting in the way, anywhere in the city.  When large conventions are on it can take an hour to travel by taxi from a nearby hotel to the LVCC.  

With the Vegas Loop the travel time will be almost constant at any time of day or night, busy conventions or not.  At very busy times there may be some queueing but so far in the operation of the first stations this has been minimal and with huge improvements to capacity it will likely remain that way.

First Principles - Always

Elon Musk always thinks in first principles.  How many atoms (physical) and electrons (power) do we need to achieve the work that we want to do.  To reach the ideal first principle solution for any new technology takes some time.  We have seen steady progress in cars from the first Roadster to the latest updated Model 3 Highland and soon to be in around 20 months time, the new Gen 3 architecture.  (Remember, Gen 3 is not a car, it is not a platform, it is a vehicle manufacturing architecture).  And, in rockets from the first early explosions before NASA invested and now, even bigger explosions as we get closer to Starship reducing the cost to orbit per kg by orders of magnitude compared to earlier NASA launches.

So, to the atoms and electrons of doing boring, what will a first principles Boring Company system look like in a few years time?  We explore some aspects of this below.  Note, in this case a few years is probably something like 20 years before we see all of the principles implemented but, have no doubt, Tesla and the Boring Company will implement all of them and more that I have not thought of.  

Doing the Boring

The actual boring of the tunnels is not really what this post is about except that to calculate the overall cost of transport in tunnels the construction cost needs to be accounted for.  However, what is rather nice about tunnels is that they tend to last a very long time with minimal maintenance - far less than normal roads because they are shielded from the weather - so the cost of boring could easily be accounted for over 20 years and maybe much more.

Elon and the Boring company have stated that they aim to be tunneling with novel improvements to Tunnel Boring Machines at unheard of rates so fast they will outpace Gary the Snail.  Incidentally, I have not heard how Gary is doing recently, I hope he is ok.  

Snails pace sounds slow but when it comes to boring tunnels it is extremely fast.  Whether they will achieve this slippery pace or not, what we do know is that it is not wise to bet against Elon.  Prufrock 3 is already being deployed at Giga Texas (see @JoeTegtmeyer for details).  Prufrock 3 will undoubtedly have several improvements over Prufrock 2 and there will of course be more versions in the future as the Boring Company aspire to the dizzy depths of Gary's snail pace.

Even at a snails pace it would take some time to build many miles of tunnels with a handful of TBMs.  But, first principles again - Tesla builds machines that builds machine.  As with all of their technology new generations of Boring TBMs will be progressively easier, cheaper and faster to make.  

When people realise just how shockingly good this system will become, there will be a ChatGPT moment and many cities and nations will start to queue up for Boring Company tunnels.  At some point, Tesla / Boring will build a TBM factory that makes a TBM a week - they are pretty big and heavy so maybe they won't ever get to the point like SpaceX of making a new rocket engine a day.  

At that rate of building TBMs two years would mean over 100 TBMs.  A bit of quick maths using the medium term goal of tunneling at 7 miles a day, even assuming a low utilisation of say 20% because the machines have to be moved around to new locations, and 300 days a year for round numbers, means that the fleet could dig 42 thousand miles of tunnels a year !

The Principles

Now, below, for the actual factors that we can consider using first principles.  Just like in electric cars already where a lighter car allows for a smaller battery which in turn allows for an even lighter car, many of the factors also compound together for even bigger improvements.  Ah, compound - do not forget the tires - we are talking first principles through and through here.

Safety  

The first principle that Tesla aim for when designing and building cars is safety.  A Model Y weighs around 2 tonnes or 4,000 lbs.  That is a lot of weight (made up of a lot of those atoms that we need to think about) and a very large factor in that weight is safety.  

As we begin to explore why a Tesla car is so heavy it will quickly become apparent that if we think first principles all the time and we think tunnels then our ideal tunnel vehicle might be a little different to a Model Y.

Le Buttle (Boring shuttle)

What form will the ideal boring company shuttle vehicle take?  Clearly a Model Y with a driver is not the ideal form factor for mass transport.  Fortunately the new gen 3 manufacturing architecture will allow for the rapid creation (relatively speaking) of several different vehicle body shapes and sizes.  

Some people will want to travel in style and book a sleeper buttle to travel overnight in luxury while they sleep but for the sake of calculating the maximum efficiency possible for the whole system more van like shapes will be used in many cases. With a squarish flat topped design we can even have various sizes simply by making longer wheelbase (with longer structural battery pack centre pieces) very easily.  It might take some statistical analysis of what vehicles give best overall capacity in any particular situation (city or area etc) as they would have to balance overall journeys and traffic volumes but perhaps in high volume situations an average vehicle might have 12 seats and carry an average of 7.5 passengers.  

For maximum throughput in busy stations vans could have wide double sliding doors on both sides - an in and an out side to match station designs throughout the system.

The van format clearly has huge advantages over a Model Y with driver as a Model Y would likely be limited to 3 passengers most of the time so if we optimistically guess an average of 2.5 passengers we could potentially be 3x more efficient with a Buttle.

But, there is a down side of a van shaped Buttle, much higher drag and air resistance. The good news here is that drag has a much lower effect at lower speeds - Buttles will not be travelling at 80mph around city systems.  

No Impacts

Cars sometimes hit things, or get hit by other things, including cars and even heavier vehicles, in the worst case where the other vehicles are going in the opposite direction.  Potentially even worse, cars even hit solid immovable things like giant pieces of concrete.  

Boring company tunnels will always be one way so the vehicles will never hit anything. And there is only one lane so nothing will hit us from the side.

When the cars become autonomous the already very low possibility of even minor collisions of one vehicle hitting the back of another will be further reduced.

With almost no chance of any significant impacts boring company vehicles can be made much lighter with far less strength than current vehicles.  Less strength in major parts of the car means significantly less weight in castings, structural battery pack, body panels, glass, seat belts and air bags (lack of) etc.

Speed

Tesla cars are incredible racing machines. Even the base model of the new Highland Model 3 has a top speed of 125 miles per hour.  Buttles could be designed for far lower maximum speeds - maybe even as low as 60mph.  A lower maximum speed has compounding effects in many other areas.    

Weight

Buttles are likely to be significantly lighter than cars across almost all of their parts.  Less weight means less power needed to drive it.

Acceleration and Deceleration

Buttles will never encounter the head thrown back to the headrest acceleration of a Model S Plaid.  In fact, passengers might hardly ever notice acceleration or deceleration at all.  The most efficient way to drive is to accelerate and decelerate slowly and smoothly.

Even with regenerative breaking some power is lost due to the less than perfect efficiency of pulling power from the battery to the motor to accelerate and getting the electrons back into the battery while braking. Therefore, the best acceleration and deceleration is no acceleration and deceleration.

While we do need to speed up at the start of a journey and slow at the end, there will be none of that incessant and repetitive faster and slower of normal road travel, especially in cities. With long tunnels Buttles can gently accelerate up to speed, travel a distance towards their destination at almost constant speed and then gently decelerate again at the other end.

Corners

Corners are quite stressful for cars on roads.  Modern cars are marvelous mechanical machines that shield many people from almost all of the stress of corners.  But, take your Tesla up to 100 mph and go round a corner and the stresses on the car go up very quickly.  A large part of the need for the very high specification of fast cars is to allow them to go round corners fast and wet roads adds another twist to those already twisty corners.  

Take away the need to be able to go round sharp corners fast and in wet weather, because buttles will not race, it does not rain in tunnels and tunnels will be mainly designed without sharp corners on longer sections and suddenly, tires, steering, and more can be much lower specification and lower weight.

Flat smooth surfaces

The driving surface inside tunnels will be almost perfect nearly all the time.  With no rain or freezing conditions ever happening inside the tunnels there will be far less degradation of the surface.  The 'road' in the tunnels will also have far lower inclines without the same need to follow the ground contours of the surface above.  Even where inclines are needed they will be less dramatic and gradual.  

Low levels of acceleration and lateral force (when cornering) applied to the surface by vehicles will also mean less wear and tear on the surface which in turn means smoother more efficient driving.

Smaller and lighter wheels
 
Why do cars need big wheels?  It is for the pumps and bumps - fast acceleration and performance and to make the ride smoother over bumps.  We already know that Buttles will not accelerate sharply (they don't do jerk) and the driving surface will be smooth so we have eliminated both reasons for big wheels.

A Model 3 with 18 inch wheels gets about 8 percent more range than one with 19 inch wheels.  A Model Y with 19 inch wheels gets about 6 percent more range than one with 20 inch wheels.  That means that potentially between 18 and 20 inch wheels we are looking at something like 14 percent more range !

At first this seems rather surprising but the reason is that big wheels with even bigger tires are very heavy.  When accelerating or braking a very large amount of energy proportional to the weight of the tires and wheels has to be spun up to speed or slowed down and that takes a lot of energy.  

Perhaps the wheels of Buttles could be even smaller because we don't need to accelerate quickly and we will be driving on very flat surfaces that are always dry.

Hold on, there is something else.  Wheels have to be very strong in case cars hit big bumps.  With very flat surfaces and no big bumps the wheel does not need to be as strong or heavy and that means we save even more energy accelerating, possibly significantly more.

Tires

Like the wheels, tires can be smaller for the same reasons.  But, there is more if we remember we have less acceleration, less cornering and no water on the driving surface.  The tires can be thinner and smaller again because they need far less grip.  

One more thing - Tesla and SpaceX have material scientists that are good at inventing new materials - the ideal tire material for in tunnels would be lighter, more hard wearing, have less rolling resistance and be much cheaper.

That all means more savings on weight (and energy) and replacement tires (cheaper each time and less often).  

Weather

Lack of weather in tunnels has been covered elsewhere but the absence of rain and ice means various savings.

NIce Body

Buttles will have NIce (Not ICE car) bodies.  They will be made of lightweight materials, rather than much heavier metals.

The unboxed and modular Gen 3 manufacturing Architecture also means that more variants of body shape and size can more easily be created so that the best design can be used for each purpose from single passenger short hop Buttles to larger sleeper modules and mass transit Buttles.  And all in nice NIce lightweight materials.

One penalty for a NIce body though, bigger Van like Buttles will have higher drag coefficients than cars.

Magic Trains

Remember that some people said trains have much higher capacity that boring cars in tunnels could ever have?

Hah - we have them beat with convoy mode.  
Autonomous Buttles will automatically form convoys (perhaps initially with a lead driver).  In van form Buttle Convoys will have the mass transit capacity of trains and look remarkably like small trains.  However, they will have huge advantages over trains in that they can dynamically form and break up depending on demand based on time and location.  And by dynamically, in fully autonomous mode, we literally mean on the move.  

As Buttles come in from outlying areas and enter busier sections of tunnels they can dynamically form convoys and conversely as convoys leave busier areas they can break apart to deliver their passengers to their respective destinations thus solving the last mile problem of current public transit systems.

In busy stations Buttle convoys will load and unload in convoy mode as well with entrance on one side and exit on the other (like some roller coasters have).  The Tesla app will of course direct you where to stand to board the correct Buttle in the convoy to be with other passengers travelling to similar destinations.

And convoy mode means slipstream savings for energy use too.

Air Flow

We have to slip this one in.  Professional cyclists will know what we mean - slipstream.  Even birds do it.

The reason that cyclists and birds try to get into what is called the slipstream of the one in front of them is because when travelling at a constant speed most of the energy used to keep travelling at that speed is consumed by pushing the air out of the way.  If you can get the distance right when following then the one in front has already caused the air to be moving in a way that the air resistance for the one following is less.  

Unfortunately there is quite an art to this as when out in the open air the effect of this is quite local as the air is pushed in various directions and the effect also dissipates quickly so that follow at much of a distance and the effect is lost.

It also turns out that the amount of energy used to push the air out of the way goes up by the square of the speed travelled.  That is, travelling at twice the speed uses up four times as much energy.  That is why fuel economy is so much worse when travelling at higher speeds.  

At higher constant speeds like 80mph the amount of energy used to push air out of the way is by far the biggest part of the total energy used (less so in ICE cars, they waste a lot of energy as heat before they even start turning the wheels or pushing air ! ) .    

Tunnels are, by definition, not in the open air. Therefore, once the air has started to move forwards when it is pushed by a vehicle it has to continue moving forward because there is nowhere else for it to go.  This effect is clearly seen in subway systems as a rush of air is felt every time a subway train approaches the station.

What this means is that in tunnel systems the slipstream effect caused by vehicles travelling is far slower to dissipate than in the open air.  With busy tunnels with many Buttles travelling through them this would create a semi permananet airflow through the tunnels.    

How much of an effect would this have on efficiency of Buttles in long tunnels?  I don't know - this really needs some very clever engineers to work it out - but imagine that the air moves at 10mph in a tunnel where the Buttles are travelling at 60mph. That would mean thirty percent less energy used to push the air once the system was running !  

One Small Motor

Given we do not need a lot of acceleration or deceleration we definitely do not need two motors.  And, the one we do need can probably be quite a lot less powerful, smaller and lighter than a Model Y needs.  That is yet more weight saved.  

Material Science

This is Tesla we are talking about.  The next decades will undoubtedly see incremental improvements in materials and therefore weight, strength and wear and tear.

Smaller Batteries

Given all the weight and power savings that means Buttles will need significantly smaller batteries than cars.  And, like any electric vehicle, a smaller battery weighs less which means even an even smaller battery is needed.

Air Bags and Seat Belts

Are not needed saving a bit more weight and cost.

Wear and Tear

As we discussed earlier, Tesla cars do some an amazing things and have to be very highly specified to cope with the stress of performing.  Buttles, on the other hand will have a very easy life with far less stress and wear on many major parts of them.  

Buttles could either be specified to be very long lasting or to a much much lower specification than a car.  Or, probably somewhere in between but, with a still significantly greater life than a car, not least because they will never be rained on or frozen.

Summary

We have a number of compounding factors that will make Buttles far more efficient than cars.  Cost per passenger mile will therefore be significantly better than surface cars.  We have to be a bit careful not to compound everything because some factors cancel each other out - for example, smaller wheels save a lot of energy when accelerating but we do less accelerating and braking because we don't have anything in the way in tunnels.  

Below are some of my wildly uneducated guesses at what we might save in energy for some of the factors.  Please do comment to help me get some better numbers in here.  Maybe, if I am very lucky @teardowntitan might comment and maybe @BoringPrufrock might look at tunnel costs.

Unless my guesses are all wildly high (and they probably are ?) it looks possible to me that Buttles may even be two thirds more efficient than surface cars even if we do not compound the percentage savings and we just add them up.

More passengers per vehicle minus higher drag - 30%
Slipstream - 15%
Tire rolling resistance - 2%
Smaller wheels and less accelerating / braking - 20%
Convoy mode - 5%
Less weight therefore less power used - 10%      (this would have been much greater except that we accelerate and decelerate much less)
Weather - 2%  (Better grip)
Less corners - 2%
Flat surface - 1%
Less inclines - 5%   (obviously dependent on local terrain).

Antoine a écrit:Perso, j'ai l'impression d'une fuite en avant vers des promesses de moins en moins réalistes. J'y vois une énorme bulle (je disais la même en 2017 avec l'action qui a fait x10 depuis)

fxmx a écrit:Je suis étonné qu'en tant qu'ingé, tu ne remettes pas en cause tes certitudes