Tesla Semi

[electracity]

Do we know if the two prototype Semi have their full stated battery pack sizes or could they be using a smaller pack for testing purposes?

I doubt the cells Tesla will use in production trucks have been made in volume yet.

 

[electracity]

It would be great PR if the empty Semi back to Sparks only use up the energy regenerated with full payload on the way down. Similar to the Iron Ore Line in Sweden ("The regenerated energy is sufficient to power the empty trains back up to the national border" Iron Ore Line - Wikipedia ). The Semi never has to Supercharge!

I think they will build the semi in Sparks with some parts made in Fremont, so there will be no deadhead.

 

[jhm]

From the TMCer who talked to them, they are both 300 mile versions. I'm guessing for more/ deeper cycles.

This reinforces in my mind the idea that Tesla will put enough kWh in the 300 so that it can go from Fremont to Sparks fully loaded. This is the basic test case they need to optimize around, both in terms of reliable range in all conditions and in minimizing degradation. After they test and optimize that, perhaps the 500 can be optimized to do the round trip in a single charge.

BTW, for load back up to Sparks, they are not limited to hauling Tesla freight. They could pick up anybody's load anywhere in the Bay Area and take it to Reno area. This would force them to test different routes and work with different kinds of load. For example, hauling a milk container could expose them to challenges of turbulence in liquids. Stabilizing a liquid load could be an interesting software challenge: can you use the highly responsive powertrain to minimize the sloshing around of milk and keep tractor and trailer stable on the road? Fully self-driving trucks will have to be able to do that, right?

Ironically, there are a lot of refineries along the way in the Bay Area, so Tesla could try hauling fuel to Reno. Would make fascinating PR.

 

[Reciprocity]

Isn't it a one million mile breakdown guarantee as opposed to a part failure/ wear warranty?
If Telsa can use 80% capacity packs for Grid applications, then replacing packs (for nominal fee) might be more cost effective than over sizing the pack for lifetime degradation.
Routes will be analyzed before tractor sales to enure customer happiness and charging infrastructure (including weather). I do not think this level of dynamic routing (different grades) will be in play for a while.

A couple of problems with that.. the routes vary too much even from season to season (winter vs summer) and many customers won't have set routes. @jhm alluded to another reason, which is charge speed. If SOC is typically 10-80, then battery will last much longer and charge faster for most of it's life. Which is a minimum of 10 years, at which point I agree with you that the pack could be replaced with 2030 tech and the old batteries could do another 10 years as storage.

Is running out of battery at 450 miles with a 500 mile semi a break down or nah? Some degradation will be ok, but 5 year old cars that don't fully charge 2x a day are losing noticable amounts of range. To me, this is probably the real breakthrough of this battery pack. It's not the energy density but the cycle rate. It would be to allow for 4000+ cycles with less then 10% degradation, unless you oversized the pack by 15% total including anti brick and a small reserve.

 

[jhm]

A couple of problems with that.. the routes vary too much even from season to season (winter vs summer) and many customers won't have set routes. @jhm alluded to another reason, which is charge speed. If SOC is typically 10-80, then battery will last much longer and charge faster for most of it's life. Which is a minimum of 10 years, at which point I agree with you that the pack could be replaced with 2030 tech and the old batteries could do another 10 years as storage.

Is running out of battery at 450 miles with a 500 mile semi a break down or nah? Some degradation will be ok, but 5 year old cars that don't fully charge 2x a day are losing noticable amounts of range. To me, this is probably the real breakthrough of this battery pack. It's not the energy density but the cycle rate. It would be to allow for 4000+ cycles with less then 10% degradation, unless you oversized the pack by 15% total including anti brick and a small reserve.

Let's be clear what a million miles on the 300 looks like. Suppose an average cycle of 250 miles. This is 4000 cycles. So they really do need a long life battery. Even then, I think this needs to be oversized, which of course allows for variable cycle depth to accommodate extreme conditions. Also they really do need a software layer to minimize battery abuse.

We should also remember that the price difference between the 300 and 500 is just $30k, $150k vs $180k. Even priced at battery cost, $100/kWh, this affords just 300 kWh for the extra $30k from which to get an extra 200 miles of range. So an incremental 1.5kWh/mile. But really, we may be talking about just and extra 200 kWh, or 1kWh/mile. So this all looks underpriced, unless the 300 miler has an oversized battery. The longevity requirement is not nearly so high on the 500, just 2500 cycles at 400 miles per cycle gets to 1 million miles. So the 500 does not need to be as oversized as the 300 to satisfy the 1M mile warranty.

So the 300 could have a 800kWh pack while the 500 needs a 1000kWh pack. This would allow the 500 to generate about $10k more in gross profit than the 300. Moreover, the 300 would have a lot of surplus battery range, so 300 miles of range could just about be guaranteed under every circumstance. In the extreme, it could be operating at 2.6 kWh/mile for 300 miles. If this can't go from Fremont to Sparks fully loaded on the coldest day of the year...

 

[mongo]

Let's be clear what a million miles on the 300 looks like. Suppose an average cycle of 250 miles. This is 4000 cycles. So they really do need a long life battery. Even then, I think this needs to be oversized, which of course allows for variable cycle depth to accommodate extreme conditions. Also they really do need a software layer to minimize battery abuse.

We should also remember that the price difference between the 300 and 500 is just $30k, $150k vs $180k. Even priced at battery cost, $100/kWh, this affords just 300 kWh for the extra $30k from which to get an extra 200 miles of range. So an incremental 1.5kWh/mile. But really, we may be talking about just and extra 200 kWh, or 1kWh/mile. So this all looks underpriced, unless the 300 miler has an oversized battery. The longevity requirement is not nearly so high on the 500, just 2500 cycles at 400 miles per cycle gets to 1 million miles. So the 500 does not need to be as oversized as the 300 to satisfy the 1M mile warranty.

So the 300 could have a 800kWh pack while the 500 needs a 1000kWh pack. This would allow the 500 to generate about $10k more in gross profit than the 300. Moreover, the 300 would have a lot of surplus battery range, so 300 miles of range could just about be guaranteed under every circumstance. In the extreme, it could be operating at 2.6 kWh/mile for 300 miles. If this can't go from Fremont to Sparks fully loaded on the coldest day of the year...

I like the extra buffer on the 300 mile and then $/kWh delta up for the 500, maybe take a little cost hit to keep same buffer %.
If Tesla uses the minimum number of semi on the Fremont-Sparks run, they will hit one million miles in 3 years. 1,000 miles/ 4 charge cycles a day.

 

[Reciprocity]

I like the extra buffer on the 300 mile and then $/kWh delta up for the 500, maybe take a little cost hit to keep same buffer %.
If Tesla uses the minimum number of semi on the Fremont-Sparks run, they will hit one million miles in 3 years. 1,000 miles/ 4 charge cycles a day.

I wouldnt be surprised if they ran those things 24x7 just to accelerate the testing.

 

[mongo]

I wouldnt be surprised if they ran those things 24x7 just to accelerate the testing.

2 sets of three semi convoys doing 2 runs a day just for packs, with some drive unit haulers mixed in. Probably at least 3 sets of 3.

 

[AudubonB]

I see these trucks no longer are doing the Fremont-Sparks run only at Oh-Dark-Thirty.
And that is why they now are being reported - this run has been going on for the better part of a year. Tesla has a LOT of real-world data on all sorts of weather and road conditions.

 

[Brando]

Another thing to consider is that Tesla is setting a 1 million mile warrantee.

Technically a 1 million mile motor/gear box is different from 1 million mile battery.
IF the battery is "perfectly managed" perhaps it never "wears out" ?
We shall see - only time may tell - not a lot of experience especially as change is continuous.

 

[mongo]

I see these trucks no longer are doing the Fremont-Sparks run only at Oh-Dark-Thirty.
And that is why they now are being reported - this run has been going on for the better part of a year. Tesla has a LOT of real-world data on all sorts of weather and road conditions.

Oh reahhhhhhhealy????
Do tell.

 

[Brando]

Let's be clear what a million miles on the 300 looks like. Suppose an average cycle of 250 miles. This is 4000 cycles. So they really do need a long life battery. Even then, I think this needs to be oversized, which of course allows for variable cycle depth to accommodate extreme conditions. Also they really do need a software layer to minimize battery abuse.

We should also remember that the price difference between the 300 and 500 is just $30k, $150k vs $180k. Even priced at battery cost, $100/kWh, this affords just 300 kWh for the extra $30k from which to get an extra 200 miles of range. So an incremental 1.5kWh/mile. But really, we may be talking about just and extra 200 kWh, or 1kWh/mile. So this all looks underpriced, unless the 300 miler has an oversized battery. The longevity requirement is not nearly so high on the 500, just 2500 cycles at 400 miles per cycle gets to 1 million miles. So the 500 does not need to be as oversized as the 300 to satisfy the 1M mile warranty.

So the 300 could have a 800kWh pack while the 500 needs a 1000kWh pack. This would allow the 500 to generate about $10k more in gross profit than the 300. Moreover, the 300 would have a lot of surplus battery range, so 300 miles of range could just about be guaranteed under every circumstance. In the extreme, it could be operating at 2.6 kWh/mile for 300 miles. If this can't go from Fremont to Sparks fully loaded on the coldest day of the year...

NOT every SEMI needs to meet the 1 million mile promise.
IF failure rate is small enough (if Tesla set aside money sufficient) that is OK. Analogy would be insurance.
Actually all warranty programs work on this basis. Failure just need to be "affordable".

 

[Brando]

I doubt the cells Tesla will use in production trucks have been made in volume yet.

Seems the SEMI battery pack may well be multiple Model 3 packs - not just one huge battery pack - modular.

 

[mongo]

Seems the SEMI battery pack may well be multiple Model 3 packs - not just one huge battery pack - modular.

Form factor doesn't seem to work for 500 mile version, portrait vs landscape. But you make me think... These 300 mile versions might be able to Model 3 modules with extra space. 8 packs worth, 32 modules, 600 kWh, ~375 mile range at 1.6 kWh/mile.

 

[jhm]

NOT every SEMI needs to meet the 1 million mile promise.
IF failure rate is small enough (if Tesla set aside money sufficient) that is OK. Analogy would be insurance.
Actually all warranty programs work on this basis. Failure just need to be "affordable".

That's correct. That is why I talk about adding batteries to reduce the warranty cost. There is a tradeoff between paying more on batteries or paying more on battery warranty claims. It's good to find a sweet spot that creates maximal value for customers at minimal cost for Tesla.

I do think that going into an undeveloped market with an unproven product, you probably want to error on the side of spending too to much on the product than to much on the warranty. We need commercial customers to have absolute confidence in a Tesla Semi as a reliable product. Once the market heats up, electric truck makers will compete more on price. Even Tesla marketing a 300-mile range product will be able to scale back battery costs and size over time to be price competitive. If they marketed it as a 800 kWh truck, they would not be able to scale back that capacity if it were to prove unnecessarily large for a 300-mile range. Maybe 750kWh will prove to be just as reliable for 300 mile range.

 

[mongo]

Recycling:
Today: 500 mile semi pack
3 years : 300 mile semi pack
6 years: 4 delivery truck packs
9 years: grid storage

 

[AudubonB]

Oh reahhhhhhhealy????
Do tell.

I just did.

You could, of course, go back to earlier cryptic posts of mine and connect the dots.

 

[mongo]

I just did.

You could, of course, go back to earlier cryptic posts of mine and connect the dots.

Hey did you edit your post? I remembered your reference without prompting,

Nah, you did Sept 29,2017

It's hiding in plain sight. On pain of a very painful excommunication-and-then-some, I cannot say more than that.

Other than:

Patience!

 

[AudubonB]

 

[scaesare]

Let's be clear what a million miles on the 300 looks like. Suppose an average cycle of 250 miles. This is 4000 cycles. So they really do need a long life battery. Even then, I think this needs to be oversized, which of course allows for variable cycle depth to accommodate extreme conditions. Also they really do need a software layer to minimize battery abuse.

We should also remember that the price difference between the 300 and 500 is just $30k, $150k vs $180k. Even priced at battery cost, $100/kWh, this affords just 300 kWh for the extra $30k from which to get an extra 200 miles of range. So an incremental 1.5kWh/mile. But really, we may be talking about just and extra 200 kWh, or 1kWh/mile. So this all looks underpriced, unless the 300 miler has an oversized battery. The longevity requirement is not nearly so high on the 500, just 2500 cycles at 400 miles per cycle gets to 1 million miles. So the 500 does not need to be as oversized as the 300 to satisfy the 1M mile warranty.

So the 300 could have a 800kWh pack while the 500 needs a 1000kWh pack. This would allow the 500 to generate about $10k more in gross profit than the 300. Moreover, the 300 would have a lot of surplus battery range, so 300 miles of range could just about be guaranteed under every circumstance. In the extreme, it could be operating at 2.6 kWh/mile for 300 miles. If this can't go from Fremont to Sparks fully loaded on the coldest day of the year...

Forgive me if it's this thread already... but I remember seeing discussion that the Semi battery chemistry may be similar to the PowerWall/Pack chemistry for cycle-life reasons. Was this stated by Elon... or was that just conjecture?