Lets work out the Tesla Semi-Truck Technical Specs

[bambam4171]

I think this has Tesla Semi written all over it Exciting Developments In NMC 811 Lithium Battery Technology | CleanTechnica

You guys of course remember that Tesla has an agreement with Jeff Dahn

 

[mongo]

I think that Roadster 2020 has 200kwh 2170 battery pack. It's one 200kwh pack or two 100kwh, but my bet is on one 200kwh. The same packs are used in semi 4 of them so 800kwh of batteries. Making semi efficiency 1.6kwh per mile. No breakthrough, no solid state batteries... Just simple model 3 batteries.

They may use the same cells, however I do not think the same pack form factor works for both. Semi packs look to be rectangular blocks semi width by half cab length by half cab height (likely a triple stack internally to each submodule) . Roaster is likely tapered for offset crash and vehicle width by wheel base - clearance (likely a double stack internally).
Factor in things like connection and support equipment integration and they would be further differentiated.

Edit: The Roadster pack is also likely a support structure for the car whereas the semi frame appears to not need the packs for strength/rigidity.

 

[lklundin]

not accounting for wind/tire/etc resistance, just fighting gravity, and assuming 80,000lb (36,000kg)

The above is simply the extra energy required to increase the Semi's potential energy, compared to driving on a horizontal surface. So for every 1000m altitude gained, the Semi has to expend this much extra energy:
h * m * g = 1000m * 36000kg * 9.81 m/s^2 = 353 MJ.
(see what I did there? - relying on SI to not worry about units). Since 1 kWh = 3.6 MJ,
the battery will need to delivery an extra 98.1 kWh per 1000 m climbed. Similarly, the battery could in principle harvest up to that amount of energy when descending the same amount, assuming no change in speed compared to traveling horizontally.

 

[lklundin]

the battery will need to delivery an extra 98.1 kWh per 1000 m climbed

The above gives some idea of the extra battery capacity needed to ascend a mountain.

Given a speed on a horizontal surface, one can also consider the extra power needed to maintain that speed while climbing a given slope, assuming that the slope is gentle enough that the horizontal speed can be maintained.

Just like the change in potential energy is given by the force of gravity (m * g) times the distance over which is works, i.e. the altitude change (h), the power required to change the Semi's potential energy is given by the force of gravity (m * g) times the vertical speed component over which is works, i.e. the Semi's (constant) speed times the incline.

We know that the Semi can manage a 5% slope at 65 miles per hour (29 m/s), which is a vertical speed of 1.45 m/s.

The extra power required to maintain that speed when going up that slope is thus
36000 kg * 9.81 m/s^2 * 1.45 m/s = 512 kW,
an increment that seems entirely doable for both the four (M3) motors and for the battery.

When maintaining that speed down the same slope, friction losses will eat into the 512 kW, so I would expect regenerative breaking to be sufficient to absorb all of the excess power.

 

[BioSehnsucht]

You won't get back all that energy on the way down of course, since regen isn't 100% efficient, but you'll get a good portion of it - at least until the battery is full (such as if you got to the top of the mountain, and then megacharged part way through the fastest charging portion of the SoC, then regen'd to ~full on the way down the other side)

 

[Lasairfion]

LG Chem NCM 811 => 410Wh/kg, 963 Wh/L (Battery technology: what's coming soon? - Push EVs) Note the rather direct "will produce"... very nice.

I still think that the Gigafactory is going to pull some new battery out for the Semi and perhaps even the new Roadster.

Here lorry drivers can only drive for a maximum of 10 hours before stopping (under British rules). If they're driving under EU / AETR rules then a rest stop of 45 minutes must be taken every 4.5 hours. And then it gets complicated with various alternatives: 1. EU and AETR rules on drivers’ hours - Drivers’ hours and tachographs: goods vehicles - Guidance - GOV.UK
Each of those stops would be a recharging opportunity.

As an aside lorries here have limiters set under European standards to 56mph; and the legal speed limits for lorries on motorways are 60mph, with 60mph for dual carriageways (50 in Scotland), and 50mph for single carriageways (40 in Scotland).

 

[Cloxxki]

Roadster, to keep up with Porsche and the like, will need faster cells. A 200kWh car with Model 3 cells will only charge at 300kWh peak.

Semi doesn't need faster charging. Might even get cells from Energy side of things. Less output, better lifespan.

Both need higher energy density. And so do many other types of vehicles.

 

[KarenRei]

Roadster, to keep up with Porsche and the like, will need faster cells. A 200kWh car with Model 3 cells will only charge at 300kWh peak.

Not really.

1) If you're never charging at the top or bottom of the range, you're reducing cell damage. You can push them harder during charging as a result for the same amount of damage.

2) You presume that Model 3 LR's batteries can't take more than ~117kW. But that's a supercharger max, not a pack max. The pack maintains that charge rate up to ~50% SoC, give or take. This strongly suggests that there's much greater max capacity at the cell level. Tesla has already stated that Model 3 is ready to take advantage of V3 supercharging currents; we don't know how powerful V3 will be yet, but speculation is around 180kW per stall.

 

[Brando]

Roadster, to keep up with Porsche and the like, will need faster cells. A 200kWh car with Model 3 cells will only charge at 300kWh peak.

Semi doesn't need faster charging. Might even get cells from Energy side of things. Less output, better lifespan.

Both need higher energy density. And so do many other types of vehicles.

Improvements in Li-Ion have been historically 5-7% per year - we see no end to this curve in the near future, right?

 

[Cloxxki]

Improvements in Li-Ion have been historically 5-7% per year - we see no end to this curve in the near future, right?

Prove this using 2012's Tesla cells vs 2018's? It's simply not true, not even close. And, Tesla's cells are still among the best in terms of energy density. Progress must have been closer to 1% per year.

 

[Brando]

Prove this using 2012's Tesla cells vs 2018's? It's simply not true, not even close. And, Tesla's cells are still among the best in terms of energy density. Progress must have been closer to 1% per year.

sorry, you seem to have been misinformed:

"The 2170 cell is around 50% larger by volume than the 18650, but it can deliver almost double the current (the 18650 delivers 3,000 mA, and the 2170 has been tested at 5,750-6,000 mA)."
Tesla 2170 Battery Cells: Greater Power At Comparable Cost

for general technical details I suggest Jeff Dahn lectures on Li-Ion batteries:
Jeff Dahn Research Group
Also Stanford - SLAC

 

[Olle]

Depends how you

Prove this using 2012's Tesla cells vs 2018's? It's simply not true, not even close. And, Tesla's cells are still among the best in terms of energy density. Progress must have been closer to 1% per year.

Depends how you define improvements. Cost improvement or capacity improvement? I'm almost wondering if the cost improvement hasn't even exceeded 5-7% for Tesla since 2012.
In Tesla's/Panasonic's case it has been more important to take advantage of a production cost experience curve Experience curve effects - Wikipedia, where they wish to stay with one design that drops in cost as the aggregated production volume increases. Others may productify newer designs more often, could be cost prohibitive for automotive. Tesla might at some point before Semi make a big jump in energy density that takes advantage of the many development steps since 2012?

 

[Brando]

Tesla claimed to have better than 8:1:1 - I think at the May or June earnings call??
anyway, below is a good overview article on Li-Ion development:
Tesla Panasonic Quietly Outmaneuver All Lithium Battery Manufacturers

JB, Elon, Marc Tarpenning have all pointed out that Li-Ion batteries double in output and half in cost about every decade since 1990s.

reminder: Tesla all about continuous improvement as fast as practical.
40-60-70-85-75-90-100 (continuous improvement, right?)

 

[BioSehnsucht]

Not just cost or capacity improvement (per volume? per kg?) but also charge/discharge rates are all improvements. Just because you only see a few % increase in capacity available for purchase in the vehicle itself, doesn't mean that these other axes for improvement don't add up to a larger number of over all improvement. It is very likely they could have chosen to take 5% per year in density at the cost of the others staying the same, instead of improving all aspects to some degree.

 

[Cloxxki]

Nice to make them cleaner and cheaper. Where are the energy density numbers though?
Tesla 90's need SuC throttling, don't survive more than 1.1C peak frequently.
Doubling in a decade. where does the bring us compared to 2008 today?

 

[Cloxxki]

sorry, you seem to have been misinformed:

"The 2170 cell is around 50% larger by volume than the 18650, but it can deliver almost double the current (the 18650 delivers 3,000 mA, and the 2170 has been tested at 5,750-6,000 mA)."
Tesla 2170 Battery Cells: Greater Power At Comparable Cost

for general technical details I suggest Jeff Dahn lectures on Li-Ion batteries:
Jeff Dahn Research Group
Also Stanford - SLAC

4416 cells x1.5 should then compare to 6624 Model S/X cells how much capacity do those have as M3's density? How does that compare to the 75, 90 and 100 packs at the cell level?

 

[BioSehnsucht]

4416x 2170 cells at 50% more volume would be volume equivalent to 6624 18650 cells, which is roughly 80% of a 100 kWh S/X pack, which is about the size of a Model 3 LR pack.

So on the surface, there seems to be no improvement. But they have both (probably) reduced the manufacturing cost and also doubled the discharge (and likely also quite increased charge) current for the cells, in addition to reducing Cobalt usage.

If they had solely worked on making no changes other than getting more kWh per volume, or per kg, then they probably could have done that. But it seems that they decided that for now reducing costs and improving charge / discharge performance is a more useful improvement.

S,X, and 3 have shown that being a heavy vehicle is not the end of the world for EV performance, though clearly there are trade offs made (do cheaper batteries make up for more costly structural components? Perhaps it's moot for their safety desires). Range is good enough for real world already, but improving charge times is useful, as is making the cars cheaper through cheaper cells.

At some point, it is likely they'll turn towards more kWh per volume or kg, but as of yet it doesn't seem that has been a priority.

 

[scaesare]

Roadster, to keep up with Porsche and the like, will need faster cells. A 200kWh car with Model 3 cells will only charge at 300kWh peak.

Semi doesn't need faster charging. Might even get cells from Energy side of things. Less output, better lifespan.

Both need higher energy density. And so do many other types of vehicles.

Tesla's cells see ~1.7C in the 75kWh packs... Which is the supercharger limit. That would equate to 340kW for the roadster.

Model 3 with a similar size pack also maxes the supercharger.

And who knows what they are capable of if the supercharger isn't the limiting factor.

 

[Cloxxki]

Tesla's cells see ~1.7C in the 75kWh packs... Which is the supercharger limit.

1.6C with 75 true capacity is 120kW. I've seen charging close to that, but certainly not higher. Screenshot?
The 75 is an odd one. Do they ever get throttled? I've not heard of it and can hardly imagine them getting more durable cells.

A manufacturer will soon (oe even presently) have the choice to go for decently cheap 3-4C cells, allowing the battery to be smaller and still maintain high travel speed, simply with a few more stops. Cost and weight would be saved. Imagine a 50kWh Tesla (cheaper than 75kWh) which charges at 200kW. Higher may be on the cards even for a 50kWh pack cheaper and lighter than a slow 75kWh one, I don't know the exact ratios.

 

[scaesare]

Sorry, it's the 70, not the 75, that hits nearly 1.7C... again it appears that the Supercharger is maxed at that point.