Tesla confirms Model 3 will have less than 60kWh battery option

[Topher]

I think zenmaster is correct here.

Feel free to argue with physicists.

Are we thinking, because the Model 3 is more aerodynamic, we'll see lower losses?

Yes, the aerodynamic portion of the power consumption equation is:
1/2 * ρ * Cd * A * v^3
where ρ is the density of air, Cd is the drag coefficient, and A is the frontal area, and v is the velocity. Power consumption drops in direct proportion to lower drag coefficient (or frontal area). Note that as the frontal area and drag go down, the other portions of the power calculation will be a higher proportion of the total. Eventually we need to dispense with the simplification that the power is dominated by aerodynamic losses.

Thank you kindly.

 

[SageBrush]

Feel free to argue with physicists.

They might enjoy the break from arguing with the AGW denialists

 

[Red Sage]

35% from what ? So if you start with a speculative price and then say 35% reduction, the reduced figure is still a number out of thin air.

We know it was previously stated to be less than $190/kWh at the pack level so we know that 125/kWh would be an upper bound.

Back in 2014 some guy claimed it would be 'impossible' to see anything less than around $178 per kWh prior to roughly 2022 to 2025 or so. He seemed to believe that Tesla's internal cost was around $240 to $250 per kWh at the time. Someone asked Elon Musk about this, and he only commented that he would be 'very disappointed' if it took ten years to reach $178 per kWh. Not too long after, it was being speculated by way of an unknown source that Tesla was already at $180 per kWh during 2014. More recently, someone at Tesla called in to a radio show and said their internal cost was already below $190 per kWh. So, taking 35% from each of those yields...

$250 » $162.50
$240 » $156.00
$190 » $123.50
$180 » $117.00​

Some noted that neither Elon nor the Tesla representative ever mentioned if this was battery cell or battery pack level pricing. I say that either way, Tesla is going to be doing a lot better than Naysayers have claimed a lot sooner than anyone hoped. But I do remember an article that said that perhaps 20%-to-22% of pack cost would be over and above the battery cell price. So, if those reduced amounts are pack ratings then battery cell cost per kWh could be as low as...

$162.50 » $126.75
$156.00 » $137.28
$123.50 » $108.68
$117.00 » $102.96​

And if those reduced amounts were for battery cell ratings, then battery pack cost per kWh would be as high as...

$162.50 » $184.66
$156.00 » $177.27
$123.50 » $140.34
$117.00 » $132.95​

Take your pick.

 

[alseTrick]

We know it was previously stated to be less than $190/kWh at the pack level so we know that 125/kWh would be an upper bound.

Not necessarily true. Do we know for certain they are saying 30-35% improvement based on CURRENT packs, or is that 30-35% improvement from the original S packs, which has been speculated by some as to what Tesla has meant?

Rather more autopilot hardware will be in the base model. This will allow Tesla to get to the full autonomy faster.

But I'm still thinking that full autonomy hardware pack will not be in the base price (way too expensive, at least now).

This seems like a contradiction.

Regardless, Tesla probably needs the hardware in every vehicle so they can collect data from every vehicle and improve their system's learning.

 

[SageBrush]

Another WAG:

1. Tesla was at least cost competitive with LG in 2016
2. The $145 LG price per kWh to GM for the Bolt was per cell and not an at-loss price
3. Tesla spends another 1.22x for packaging
4. The GF will reduce cell charges 35%

This sums up to 145*0.65*1.22 = $115 per kWh in the pack

 

[SageBrush]

I'd like to hear informed estimates of the Tesla motor and inverter costs compared to ICE car transmissions.

 

[Red Sage]

This is a 320 kW Inverter for a Model S:

Tesla Model S Motor display:

Drive units awaiting installation:

I really doubt these cost very much at all in actual materials... The real expense to the manufacturer is that of research, development, and software control systems. And those combined costs are likely higher for now than would be on an 'ordinary' transmission. Once you know they are going to have to build several hundred thousand of them per year, it seems that economies of scale would make them much less expensive.

 

[SageBrush]

I sooooo have to tour the Tesla factories. Those photos are wonderful.
My international family is visiting their daughter in San Jose this summer, so I suggested we meet in Lake Tahoe: Beautiful area, blah blah blah ... and pretty close to Reno.

 

[Model 3]

Reduction of 35% whereas Elon had expected only 30% costs

Incorrect (yes, I see that you got that from the article, so that one is in error). J.B. and Elon expected at least 30% cost reduction, and was hoping for up to 50% when GF-I was fully build and in production (not from the "get go"). So this is not more then expected, just a confirmation that they now see that they will do it better then the earlier lowest estimate.

Back in 2014 some guy claimed it would be 'impossible' to see anything less than around $178 per kWh prior to roughly 2022 to 2025 or so. He seemed to believe that Tesla's internal cost was around $240 to $250 per kWh at the time. Someone asked Elon Musk about this, and he only commented that he would be 'very disappointed' if it took ten years to reach $178 per kWh.

I can't say if this is correct or not, but what I remember was that Elon would be 'very disappointed' if it took ten years to reach $100 per kWh (cell cost). Not sure if that was in 2014 or 2015. It was at the time well known that the latest contract with Panasonic had a $180/kWh cell cost. My take on this is that they are targeting well under $100/kWh cell cost by the time GF-I is fully build out and operating at full speed. Originally that would be in 2020, but with the accelerated plans, they would be at that production level in 2018. Not sure everything will be in place by then, but I guesstimate close to $100/kWh cell cost to Tesla at least by the end of next year, and under ($85-90/kWh?) by the end of 2020.

More recently, someone at Tesla called in to a radio show and said their internal cost was already below $190 per kWh.

Some noted that neither Elon nor the Tesla representative ever mentioned if this was battery cell or battery pack level pricing.

That *was* battery pack level. It had been $180/kWh or under on the cell level for at lest 3 year at that time, so no need to confirm that.

Edit: Btw, at the time where GM confirmed $145/kWh cell cost on the Bolt, Elon - at a event in Japan I think it was - said they was under this price. It was unclear if he meant "already" or "when the Bolts starts production". Anyway Bolt is in production now, so we know - if we can trust Elon - that the price to Tesla at the cell level should be (well?) under $145/kWh by now - before GF-I starts to produce cells for the automotive part of Tesla and reduces the prices by (at least) 35%.

 

[LectrikPower]

Anyone see this yet?

Tesla is now claiming 35% battery cost reduction at ‘Gigafactory 1’ – hinting at breakthrough cost below $125/kWh

Woops sorry, didn't go back far enough.

 

[JeffK]

Incorrect (yes, I see that you got that from the article, so that one is in error). J.B. and Elon expected at least 30% cost reduction, and was hoping for up to 50% when GF-I was fully build and in production (not from the "get go"). So this is not more then expected, just a confirmation that they now see that they will do it better then the earlier lowest estimate.

Q2 2014

So I mean I think that's generally true that a vertical integration and doing things at large scale results in cost reduction. I feel very confident with the 30% cost reduction per unit of energy. We're obviously going to target something higher than that.

He later said:
Q3 2014

Yes, to be precise about our prediction was that we felt comfortable with at least a 30% improvement in cost or reduction in cost just based on the location and economies of scale.

My point exactly was that when coming up with an initial price point for Model 3 you'd normally take into account the highest possible costs. If you reduce costs from there, then you can either include more for the same price point or simply collect higher profits.

 

[scaesare]

This is a 320 kW Inverter for a Model S:

That's actually a Model S inverter. The full assembly is on the right. The unit on the left is one of the 3-phases of the inverter separated, and with the overlying PCB's removed from the yoke so you can glimpse the power electronics underneath. Note the EXIF data on this similar pic is dated May 2014:

While we've seen a front Model S inverters, and have had leaked info about the model 3, no actual pics of it I've seen.

 

[Red Sage]

I can't say if this is correct or not, but what I remember was that Elon would be 'very disappointed' if it took ten years to reach $100 per kWh (cell cost).

I stand corrected. I do believe that is exactly what he said. And yes, it was in 2014.

 

[zenmaster]

IAre we thinking, because the Model 3 is more aerodynamic, we'll see lower losses?

You mean less power required to overcome drag? Of course. Also less power is required to move less weight.

 

[dsvick]

If that is true (more price reduction than Elon expected) I still don't expect Model 3 base price will drop.
Rather more autopilot hardware will be in the base model. This will allow Tesla to get to the full autonomy faster.

But I'm still thinking that full autonomy hardware pack will not be in the base price (way too expensive, at least now).
But I do expect the wiring harness to be ready so retrofitting will be possible.

Either front radar or front camera plus tailgate camera will definitely be standard due to safety.
Pretty soon both of those will be mandatory.


35% from current Panasonic price most likely. Part of it transportation cost, part is profit for Panasonic. And also more simple design (less cells per pack) results in cheaper pack assembly.

I agree, they won't change the base price, they'll take a bit more in profits or add a few kWh to the base battery.

The model 3 is already getting all of the hardware for full self drive so there won't be any hardware that they can add.

 

[ikjadoon]

You mean less power required to overcome drag? Of course. Also less power is required to move less weight.

The power difference at highway speeds is not well correlated in the Nissan Leaf with an additional 1000lbs.

Your estimates for the Model S were essentially perfect. For the Model S, it's a loss of ~18 to 21 miles per 5mph increase.

Feel free to argue with physicists.

You misread my post. I was only corroborating the quote about the Model S and it is correct. If you have better information, please inform Tesla so they can correct their website.

It's roughly 20 miles of range lost per 5mph at highway speeds on the Model S.

Yes, the aerodynamic portion of the power consumption equation is:
1/2 * ρ * Cd * A * v^3
where ρ is the density of air, Cd is the drag coefficient, and A is the frontal area, and v is the velocity. Power consumption drops in direct proportion to lower drag coefficient (or frontal area). Note that as the frontal area and drag go down, the other portions of the power calculation will be a higher proportion of the total. Eventually we need to dispense with the simplification that the power is dominated by aerodynamic losses.

Thank you kindly.

Thank you for the equation. It seems at highway speeds, aerodynamic losses are still predominant.

See this Nissan Leaf comparison with 1000lbs added weight vs overall efficiency at highway speeds. The Leaf is approximately 25% less aerodynamic (drag coefficient x frontal area) than the Model S, however.

 

[SageBrush]

It seems at highway speeds, aerodynamic losses are still predominant.

I'm pretty sure that Topher is looking forward to Model 3 level Cd and better. I know I am!

 

[Topher]

It's roughly 20 miles of range lost per 5mph at highway speeds on the Model S.

Just because it is within a small range for a small range of speeds does not mean one should make the implication that the relationship is linear. That is a sure way to confuse people into thinking that the relation applies outside of that small range. Which it does not.

Thank you kindly.

 

[zenmaster]

Thank you for the equation. It seems at highway speeds, aerodynamic losses are still predominant.

I'd have to do some math, but I'm almost positive that the huge weight difference will be a greater resistive force (rolling resistance) than drag, even at highway speeds. In areas like mine, the road gradients amplify this power burden.

 

[Topher]

I'd have to do some math, but I'm almost positive that the huge weight difference will be a greater resistive force (rolling resistance) than drag, even at highway speeds. In areas like mine, the road gradients amplify this power burden.

Seems unlikely. I can push a car by myself. I can't push one to highway speeds.

Don't let that stop you from doing the math.

Thank you kindly.