Tesla Semi

[mongo]

Gotta wonder if the clutch is on the Cybertruck? First thought is that it's complex, heavy and expensive. Second thought is that it might be a lot cheaper and lighter than the batteries it replaces to get equivalent range.

Haven't seen much discussion here but this could be a bigger breakthrough than people realize. Maybe even end up on entry level vehicles where four wheel drive is wanted?

Just one more example of superior Tesla engineering.

Cybertruck going from 4 to 3 engines might indeed point to the clutch.

Semi needs the clutch because the boost axle has another stage of reduction and the motor's drag at high speed is an efficiency drain. Cybertruck has a higher top speed and needs less low end.

Cybertruck is going from tri-motor to quad, not 4 to 3. Tri-motor was like a Plaid/ Semi love child with a lot of payload mass (and thus traction) on the rear axle. Dual front motors is an interesting change, may be for increased power & regen.

Semi clutch packaging worked due to truck width and solid axle. An Independent suspension with half shafts needs as much length as they can get to save the joints. (Could go asymmetric and trade off longitudinal angle).

 

[mongo]

I wonder why the 2kWh/m figure was used during the presentation?

It was shown as <2 kWh/ mile. Protecting the worst case.

 

[Captkerosene]

Semi needs the clutch because the boost axle has another stage of reduction and the motor's drag at high speed is an efficiency drain. Cybertruck has a higher top speed and needs less low end.

Cybertruck is going from tri-motor to quad, not 4 to 3. Tri-motor was like a Plaid/ Semi love child with a lot of payload mass (and thus traction) on the rear axle. Dual front motors is an interesting change, may be for increased power & regen.

Semi clutch packaging worked due to truck width and solid axle. An Independent suspension with half shafts needs as much length as they can get to save the joints. (Could go asymmetric and trade off longitudinal angle).

Thanks for the response.

Just a thought: did Pepsi buy 300 or 500 mile range trucks? Everyone assumes that all the trucks being built are 500 mile rigs when I haven't seen that stated anywhere.

Also, I wonder if the statement that the semi weighs the same as a standard semi is referring to the "300" model. I've seen it speculated that given current battery density the 500 mile truck's battery would weigh close to 15K pounds. (Hard to make up for that much weight.) People are backing out tractor weights from gross weights assuming that the truck is a 500 mile variant and not a 300 mile variant. That would account for "the missing 5K" in tractor weight if it was a 300 model semi. Just a theory.

 

[mongo]

Thanks for the response.

Just a thought: did Pepsi buy 300 or 500 mile range trucks? Everyone assumes that all the trucks being built are 500 mile rigs when I haven't seen that stated anywhere.

Also, I wonder if the statement that the semi weighs the same as a standard semi is referring to the "300" model. I've seen it speculated that given current battery density the 500 mile truck's battery would weigh close to 15K pounds. (Hard to make up for that much weight.) People are backing out tractor weights from gross weights assuming that the truck is a 500 mile variant and not a 300 mile variant. That would account for "the missing 5K" in tractor weight if it was a 300 model semi. Just a theory.

Even if made from Model S packs (with structure and electronics), a 1MWh pack would only be 12k pounds.

 

[jhm]

So I got that wrong. As @jhm noted, Elon has updated what was shown during the Semi launch presentation to say 1.7kWh/mile, meaning the pack is around 900kWh. I wonder why the 2kWh/m figure was used during the presentation?

I would assume the same; 1.7 is fully loaded. Which is impressive!

I think the 2kWh/mile figure was hedging or worst case scenario, like massive elevation climb fully loaded.

 

[mongo]

I think the 2kWh/mile figure was hedging or worst case scenario, like massive elevation climb fully loaded.

Hedging on the flat with a bad trailer. It's definitely not worst case. 1% grade at 60 MPH is 1.6 kWh/mile just in elevation change.

 

[Doggydogworld]

Even if made from Model S packs (with structure and electronics), a 1MWh pack would only be 12k pounds.

I thought the 85 kWh pack with 7104 cells was 1200 lb. I don't remember anyone weighing the 100 kWh pack, but the 1152 extra cells would add ~120 lb.

 

[jhm]

I wanted to get some numbers on the differential mpg trucks get load vs empty. This was he best I could find: MPG of your truck full and empty?

It's pretty informal data. What I gather is empty is about 15% above average and loaded 15% below average (where average is just the midpoint between loaded and unloaded. (I'll ignore bobtail and partial load conditions.)

Applying this +/-15% rule to the fleet average of 6.5 gets 7.5 mpg empty and 5.5 mpg loaded. This seems plausible.

Next applying this to the Tesla Semi. Let's suppose 1.7 kWh/mile is loaded. Then average would be 1.445 kWh/mile (= 1.7 * 85%) and empty 1.257 kWh/mile (= 1.7 * 85% / 115%). (Remember that miles is in the denominator here, but in the numerator for mpg, so you have to invert everything.) This too seems reasonable to me and the average estimate is within the range I was expecting (between 1.2 and 1.5).

Obviously, when people want to understand the practical range, you want to use the fully loaded efficiency. If you can haul your load, you can make it back no problem. So to have a practical (loaded) range of 500 mile at 1.7 kWh/mile, you need at least 850 kWh of charge plus a buffer of maybe 50 kWh or 29 miles. Thus, a pack of 900 kWh seems about right.

BTW, for the return trip empty, 500 miles just requires 629 kWh of charge. So a 75% charge to return should suffice, whereas for the loaded trip you want to start with near 98% charge (just a little capacity for regen braking). This differential suggests that logistics firms will want to place more charging infrastructure and allow for more charging time near where they load trucks and less near where they unload. Of course if the loading area is at much higher elevation than the unloading areas, this could get reversed, but in that case, you're getting free energy from your load!

 

[JRP3]

whereas for the loaded trip you want to start with near 98% charge (just a little capacity for regen braking).

I see reduced regen at 85% SOC in my 3, and at even lower SOC in moderately colder weather.

 

[JRP3]

I thought the 85 kWh pack with 7104 cells was 1200 lb. I don't remember anyone weighing the 100 kWh pack, but the 1152 extra cells would add ~120 lb.

The new Plaid pack has better specific energy:

In the older Model S, it used to be 157 Watt-hours (Wh) per kilogram of the pack weight, with the Plaid, it has now jumped to 181.5 Wh / kilogram.

Sandy Munro reveals Tesla Model S Plaid battery secrets, Musk says Model Y structural pack will be next level - Tesla Oracle

Vehicle teardown expert Sandy Munro takes a very close look at the Tesla Model S Plaid battery pack (in-depth details of the findings and video analysis).

www.teslaoracle.com

That works out to right around 1200 lbs for the new pack

 

[jhm]

I see reduced regen at 85% SOC in my 3, and at even lower SOC in moderately colder weather.

Hmm, I wonder how that scales to a 900 kWh battery. I honestly don't know. It seems that some 18 kWh of charge capacity should be able to provide 50 kW or more of breaking, maybe not the max, but some. Would you really need about 135 kWh capacity to provide effective regen?

Of course, trying to go a full 500 miles on a single charge is testing the limits of the battery. Just a little charging along the way takes the edge off.

 

[mongo]

Hmm, I wonder how that scales to a 900 kWh battery. I honestly don't know. It seems that some 18 kWh of charge capacity should be able to provide 50 kW or more of breaking, maybe not the max, but some. Would you really need about 135 kWh capacity to provide effective regen?

Of course, trying to go a full 500 miles on a single charge is testing the limits of the battery. Just a little charging along the way takes the edge off.

1MWh is around 8,200 pounds worth of 2170 cells.

 

[jhm]

1MWh is around 8,200 pounds worth of 2170 cells.

10X that in gross weight when fully loaded.

 

[mongo]

10X that in gross weight when fully loaded.

Oh geeze, I blended your post with the previous one, sorry!
To your actual post:
If pack is charging rate limited it seems that limit would also apply to regen to the pack. Of course, it can use regen to generate heat to warm the pack up to the limits of the heat pump and motor thermals.

If not limited by temperature, regen should scale with pack size to the limits of the inverter. The gear ratio on the boost axle should theoretically increase regen capability by operating the motor at a higher backEMF voltage (less current for same power).

Model S Plaid does > 1C regen of 185kW, normal S/X does > 0.5 C or 50kW.

 

[MC3OZ]

Thanks for the response.

Just a thought: did Pepsi buy 300 or 500 mile range trucks? Everyone assumes that all the trucks being built are 500 mile rigs when I haven't seen that stated anywhere.

Also, I wonder if the statement that the semi weighs the same as a standard semi is referring to the "300" model. I've seen it speculated that given current battery density the 500 mile truck's battery would weigh close to 15K pounds. (Hard to make up for that much weight.) People are backing out tractor weights from gross weights assuming that the truck is a 500 mile variant and not a 300 mile variant. That would account for "the missing 5K" in tractor weight if it was a 300 model semi. Just a theory.

I am sure the trucks are 500 miles variants, I remember a statement about that somewhere.

300 mile variants would allow them to build more trucks with the same amount of cells, but they may be focused on filling customer orders in a particular sequence.

It is probably easier to start production with one variant.

As for the current build rate, we have had lots of rumours and speculation, but we don't have a lot of solid insight.

On the Semi weight we have also had a lot of speculation, I think the facts will come out soon enough.

One thing that occurred to me is that the cab could be lighter than a conventional Semi cab, in fact if the cab isn't lighter Tesla has probably missed an opportunity to reduce weight.

We might we waiting a while for a Munro teardown of the Semi, but I think weight and range info will leak out soon enough.

 

[Beltsbear]

Just take a 40% hit in those cold, snowy, windy conditions.

The old 40% hit was for pre- heat pump vehicles. Heating the cabin resistively is tough and is almost as much power as moving the vehicle. Now if you did that (resistively) in a semi it would not be nearly as bad, but luckily we don’t have to do that, we have a heat pump. With 4-5x the power use to move a semi vs a car, you also have substantially more waste heat to use in the cabin. Finally the thermal mass of the battery (greater Than 10,000lbs) works in your favor, and that is “paid for” during overnight charging. The hit for cold weather driving in a semi could be 5% or less.

 

[Captkerosene]

I am sure the trucks are 500 miles variants, I remember a statement about that somewhere.

300 mile variants would allow them to build more trucks with the same amount of cells, but they may be focused on filling customer orders in a particular sequence.

It is probably easier to start production with one variant.

As for the current build rate, we have had lots of rumours and speculation, but we don't have a lot of solid insight.

On the Semi weight we have also had a lot of speculation, I think the facts will come out soon enough.

One thing that occurred to me is that the cab could be lighter than a conventional Semi cab, in fact if the cab isn't lighter Tesla has probably missed an opportunity to reduce weight.

We might we waiting a while for a Munro teardown of the Semi, but I think weight and range info will leak out soon enough.

Rob stated tonight that the Sparks factory is a prototype factory and will only ramp to about ten trucks per week. The big factory will be in Austin. I question how they're going to build that factory when they have the Model Y ramp, Cybertruck ramp, RT design/production and 4680 ramp all happening at the same time. Of course it's just me speculating without any actual information.

PepsiCo does a lot of their own deliveries. So a driver might make 15 or 20 (?) stops in a day. I think a 300 mile truck would work for them since EVs do well at low speeds in city traffic. Maybe they use smaller trucks for that IDK.

The packs are modular I think. Two vs three battery modules for the two variants. (The math doesn't exactly work though.) So I think it's easy to switch production between the two models. Producing the "300" makes two tons of sense to me.

There are ways to save weight: aluminum wheels etc. I don't know the weight of an aluminum flat bed trailer vs a 10,000 lb enclosed trailer ... gotta be several thousand pounds. Nobody was accounting for that in their tractor weight calculations.

I still think it's a great product.

 

[MC3OZ]

Of course it's just me speculating without any actual information.

Yep, I'm on the fence about this Semi information a lot of it is coming form a single source.

Multiple sources have confirmed we should not expect anything too spectacular in Q4 2022, perhaps 100 semis built in total.

We will get an update in production and deliveries, and the question will probably come up in the earnings call.

We will know more in 3-6 months, trying to read the tea leaves now cold lead to a lot of jumping to the wrong conclusions.

I'm not sure if the higher volume line will be at Austin or Sparks, and I wonder how much credibility to assign to inside sources.

 

[petit_bateau]

Yep, I'm on the fence about this Semi information a lot of it is coming form a single source.

Multiple sources have confirmed we should not expect anything too spectacular in Q4 2022, perhaps 100 semis built in total.

We will get an update in production and deliveries, and the question will probably come up in the earnings call.

We will know more in 3-6 months, trying to read the tea leaves now cold lead to a lot of jumping to the wrong conclusions.

I'm not sure if the higher volume line will be at Austin or Sparks, and I wonder how much credibility to assign to inside sources.

In the Q4 of 2022 P&D numbers I would be expecting maybe 10-15 Semis rather than 100.

 

[Electroman]

The hit for cold weather driving in a semi could be 5% or less.

No challenging this, but this number seems too good to be true. Remember besides heating the cabin, there is also additional battery heating and the air resistance is higher due to denser cold air. The last one is the big factor between say 20F vs. 80F.

This probably belongs in the Semi thread.