Pics/Info: Inside the Tesla 100 kWh Battery Pack

[smac]

Thanks. Had not heard that before.

That having been said, I think my initial point that the Model S was designed from the ground up (as opposed to a repurposed ICE platform), holds true.

It surprised me too when I first found out about it.

I agree, the benefit of not simply trying to re-purpose an existing ICE design (like say the Golf E) has huge benefits. And the skateboard idea of the Model S was a great solution.

Giving it a "ICE length" hood was probably aethetics as well as hedging their bets on if they would go REX or not early on in the design.

With the BEV and only BEV designs like the 3, Merc EQ, IPace, it's also clear the stretched skateboard/ cab forward style seems to be where we are heading to. So my point is really about how manufacturers even with less dense chemistries, may be able to squeeze more kWh in to their cars in this way.

(FWIW I am expecting the Model Y to continue this trend even further).

 

[Cloxxki]

Less dense batteries woudn't worry me too much, especially when it's mostly volume.
Allow me to over think this.

In Model S/X, we know there is some space left, the 2170 cells are to be placed into the same chassis, same upright positions, 5mm longer.
Imagine merely sacrificing 2cm of car height to get a fatter battery in there. So you can fit in stead of 65mm, 90mm long cells. Even if we pretend the added juice would be proportionate to that (in reality a bit better), that's +38% in battery space. And yes I know about the double stacked module and blow right past it for ease of example.

+38% from the now relatively old 18650's in present Teslas (degradation issues with repeated fast DC charging) would fit 141,8kWh if it were the same old chemistry. With thicker cells, say 21x90mm, that could be around 155kWh or so.
A lower density 2017 battery tech will likely find enough space to make it a decent range. And the 2cm higher car body isn't going to ruin aero completely, it's mainly the roof line.
This is all using the same floor surface as the S/X battery. These cars could get a heck of a lot of batteries in the back (hidden compartment), let alone frunk.
If there was a less dense but totally awesome chemistry, be it for charge speed, cost or reliability, a long range range could be built around it.

Off topic, I am very disappointed Model 3 is kept as non-premium as possible 74kWh big pack supposedly has been leaked? Not exactly the most inspired effort by Tesla to make the car has as long a range as possible. In a Model S/X, this would take 60% the space of the 100 pack. 300+ miles is enough, sure, but it's not OMG the ICE car is dead long. And with their self promoted cheapest batteries on the planet, offering the ICE killer car could have been done from $45k, ea-si-ly. 100kWh should fit in a Model 3, if they want it hard enough. That'd be a bit over 400 miles. But of course, milking the fanbase with more modest cars at less modest markup is priority. I get it, but don't cheer.

 

[smac]

@Cloxxki It's interesting the approach the IPace, EQ, and Etron are taking. Which is exactly what you are proposing, make the pack taller, raising the roof line of the vehicle to still give decent interior height, under the guise of making it an "SUV". They will still be low roofed for the SUV class due to aero, but taller than a sedan / sports wagon. Think Subaru Outback.

I'll pick the IPace here as it's the closest to release and Jaguar have released provisional specs. It is 1560mm tall. Which is 100mm shorter than their closest ICE equivalent the F Pace, but 100mm taller than the Model S. The rumour I am hearing is that the pack itself is 100mm deep, so overall the interior head room should be more than the S.

It's speculation on my part, but I can't help think the recent announcement that the Y will be on a whole new platform, is probably for similar reasons.

 

[Cloxxki]

@Cloxxki It's interesting the approach the IPace, EQ, and Etron are taking. Which is exactly what you are proposing, make the pack taller, raising the roof line of the vehicle to still give decent interior height, under the guise of making it an "SUV". They will still be low roofed for the SUV class due to aero, but taller than a sedan / sports wagon. Think Subaru Outback.

I'll pick the IPace here as it's the closest to release and Jaguar have released provisional specs. It is 1560mm tall. Which is 100mm shorter than their closest ICE equivalent the F Pace, but 100mm taller than the Model S. The rumour I am hearing is that the pack itself is 100mm deep, so overall the interior head room should be more than the S.

It's speculation on my part, but I can't help think the recent announcement that the Y will be on a whole new platform, is probably for similar reasons.

If the ride height / ground clearance is greater though, that is deducted from the extra head room of course. And I could even imagine lower sections in a pack, say 6cm in stead of 10cm, to accomodate leg room spaces, get the same comfort in a lower car.

Cars are huge. if you design a BEV (literally) ground-up, there is a lot of space for fluffy batteries.

Which kinds of technologies are attractive in thise realm, and how do they perform for energy and power density and cost?

 

[Offgrid]

Just pulled this out of a 2015 s70
Have you Ever seen this contactor cover anywhere but the 100 kwh packs
I think it was a replacement pack

 

[SurferChick]

S

I originally posted this on my projects site, but will also post here.

All rights reserved. Images may not be copied or used without my express permission.

Pics and Info: Inside the Tesla 100kWh Battery Pack | wk057's SkieNET

Jason Hughes on Twitter

---

So I recently purchased the first known salvage auction Tesla with the new 100kWh battery pack, a P100D VIN 159k.

There have been tons of rumors flying around how what changes Tesla had made to increase the capacity so drastically. Rumors of new cooling patents, increased voltage, new cell double bond wiring, incompatible with older cars, and all sorts of things.



Personally, I figured Tesla wouldn't reinvent the wheel just yet and go with a whole new pack design, but who knows. Not me, so that had to be corrected.

Without further ado, here is a shot of a module from the 100 kWh pack.







So, immediately with one picture we have a lot of answers.

• They're 18650s
• There are more of them (12 more per group to be exact)
• Single cell fuse/bond wires
• Still six cell groups (no voltage change)

Awesome. 516 cells per module. That's 8,256 cells per pack, a ~16% increase vs the 85/90 packs.

A comparison with an older module makes the added cells more clear:



Basically they crammed a couple more rows of cells into the module.



But what about the rumors around cooling? Well, they did modify the cooling, but not in any exotic way. The new modules simply have two shorter and thinner cooling loops per module. This way the coolant doesnt have to run past so many cells before exiting.







What about compatibility with other/older cars? Well, the pack itself has the same high-voltage connection, the same low voltages connectors, and the same cooling connector. However, the pack has the newer ring around the high-voltage connector. So, it's plug-and-play (for the most part, firmware and config changes needed) on the Model X and refreshed Model S, however it would require a different spacer ring on the high voltage connector. Tesla even has a part number for it, so it should be pretty simple to put into any Model S/X.

The six screws on the orange HV connector can be removed to replace the ring with one compatible with older Model S.

As for real capacity, the BMS reports usable capacity at a whopping 98.4 kWh. It also reports a 4 kWh unusable bottom charge, so that's 102.4 kWh total pack capacity! Congratulations, Tesla. A high capacity pack that meets its nameplate rating!



Also, the BMS reports a max discharge current of 1,760A for the P100D Ludicrous pack. The normal Ludicrous P90D/P85D packs report 1,520A. This makes perfect sense with scaling, since it's 74 vs 86 cells per group.



Additionally, the pyro fuse in the 100 kWh is labeled differently. I couldn't find any physical differences between the normal ones, but it is labeled "DISCONNECT,PYROTECHNIC,HYBRID,100kWh", which suggests it has a different setting for the higher current. It also still has the two non-rechargeable lithium ion cells in the case, suggesting it will need replacement at some point like the rest of them.



It's great to put rumors about this pack to rest. No exotic cooling, is compatible with older cars (physically), no new cell type, etc. Just a new module design that packs more cells into each one.



Enjoy.



-wk

Super-interesting thread. Who knew that Tesla-nuts were such clever people! Very cool to see what's going on under the "hood"...

 

[CSFTN]

That's one of many reasons. Even if the cells could handle the power, the charge port, charge connector, and associated wiring simply can not.

Let's look at it this way: Let's assume Tesla bumps up the supercharger side and can pack in 250kW supercharging power into the pack at say, 40% SoC. Well, that'd work out to something like ~750A. Well, the cables from the charge port to the fast charge contactors would be hot enough to boil water within about one minute. The cable from the car to the supercharger would be just as hot in about the same time. The pins in the charge port along with other termination points would likely be roughly double those temperatures in the same amount of time, melting the plastic around them pretty quickly, eventually causing serious problems (ie: fire). Assuming we can get around all of that, the pack cooling at 100% efficiency can maybe sap away a few kW worth of heat. Well, assuming the cells could handle it without damage, and maintained a reasonable IR, the IR of the best Tesla cells I've measured would mean there would be about 20kW more heat generated beyond the thermal system capability inside the pack. For perspective, 20kW of heat is the output of a 70,000 BTU heater (6 ton HVAC system). So, if after the first minute or so the wiring involved didn't burst into flames, and the cells don't explode, then within the next couple of minutes the pack would be hot enough to fry and egg on and likely be on fire.

Nothing against Musk and his desire to bring faster charging to his vehicles, but it's just not happening with the current cars. Him giving the impression of anything else, IMO, is just wrong.



The MPH charge rate thing is a bit of a joke, honestly. It's some kind of running average of power input over some unknown amount of time. The result is an artificially high number.

Anyway, I wouldn't expect anything but a minimal increase in supercharger power any time soon, and only on the largest packs. I'd guess the 100 packs could be charged at a faster peak rate with a more aggressive taper, but that's about it. I'd bet it's at least two years before they even have a prototype car capable of even double speed (~250 kW) supercharging, and probably double that or more before any customer has there hands on one, and probably even longer before the stations exist in enough quantity to be useful.

Any desire to update this set of statements given Supercharger V3 specs are now known?
(and note that wk057s prediction of 2 years is spot-on)

 

[Electroman]

So I am confused now. Who was more correct? wk07 or Musk? Reading through that wk07's post from more than 2 years ago, that all seems reasonable.

So maybe SpC V3 is less capable than what Musk originally predicted? Is the reality somewhere in the middle?

 

[Cloxxki]

So I am confused now. Who was more correct? wk07 or Musk? Reading through that wk07's post from more than 2 years ago, that all seems reasonable.

So maybe SpC V3 is less capable than what Musk originally predicted? Is the reality somewhere in the middle?

wk07 may not have calculated with water cooled plugs. Tesla's not the first to implement this, of course.

 

[tga]

And the cars that support 250kW are reported to have heavier wiring from the plug, etc. @wk057 was commenting (correctly, IMHO) on the hardware in play at the time of his post.

 

[wk057]

Any desire to update this set of statements given Supercharger V3 specs are now known?
(and note that wk057s prediction of 2 years is spot-on)

So, since my post quote above (which appears to have held up pretty well), not a lot has changed really. I've found that the laws of physics tend to stay pretty well founded over time.

That said, the liquid cooled cables can alter some of the underlying issues on the charger side of things. The amount of waste heat involved hasn't changed, but assuming they're able to sap away sufficient heat from the external cabling that could solve that issue for running longer duration over-current situations. Assuming they managed to pull this off, that'd cover the external side of things I suppose. (We'll ignore that to do this they're having to waste many kW's of energy in exchange for speed).

However, the charge port and car side, at least on S/X, hasn't changed. I've seen the 3 chargeport as well, and I'm definitely skeptical. Going to have to do some testing there once I can access a v3 supercharger myself.

I still don't see speeds increasing significantly for S/X. The 100, maybe a little. But I've done a lot of testing over the years on the 18650 types... and we're not pumping 250kW into them safely. A few percent more, maybe, but not ~700A.

The 3 cells seem to have a lower IR than the 18650s (expected, but they did a better than expected job), and the cooling on the 3 packs gets more contact with each cell, making sucking away heat more efficient. The chemistry seems capable, at least. My worry remains the longevity of the charge port and related in-car components when we're talking about a 5x over-current situation for an extended period (minutes vs seconds).

Let's put it this way, a worn charge port or connector will make enough heat to melt it at 80A... and we're talking 700A+ for SpCv3. Also, the gen1 (manual) charge ports didn't have robust thermal monitoring, so there's no way to really do this safely there, either.

Summary, my stances haven't really changed much with regard to the car side. If we see any significant bump on the S/X side it'll be to the 100, and it will be modest, no where near what they're doing to the Model 3. Everyone else is probably SOL.

 

[Speedr117]

I think I saw 145 vs. the current 120 for the 100 pack. Does this seem doable (what you mentioned as a few percentage more)? Also, maybe less of a taper and no more sharing to make a meaningful decrease in time at a Supercharger V3?

 

[ucmndd]

Let's put it this way, a worn charge port or connector will make enough heat to melt it at 80A... and we're talking 700A+ for SpCv3.

Don't disagree with your conclusions, but isn't the max amperage we're talking about on v3 ~625, not 700+?

400v * 625A = 250kw

Not that it matters much to the outcome, I'm just trying to understand.

 

[MP3Mike]

Don't disagree with your conclusions, but isn't the max amperage we're talking about on v3 ~625, not 700+?

400v * 625A = 250kw

Not that it matters much to the outcome, I'm just trying to understand.

The 250kW charge rate is not when the battery is full, it is when the battery is empty and the voltage is lower.

I think the voltage at the low end of the SoC is ~350v which would mean that you have to have 714 amps to get 250kW.

 

[ucmndd]

The 250kW charge rate is not when the battery is full, it is when the battery is empty and the voltage is lower.

I think the voltage at the low end of the SoC is ~350v which would mean that you have to have 714 amps to get 250kW.

Right. Thanks.

 

[wdolson]

So, since my post quote above (which appears to have held up pretty well), not a lot has changed really. I've found that the laws of physics tend to stay pretty well founded over time.

That said, the liquid cooled cables can alter some of the underlying issues on the charger side of things. The amount of waste heat involved hasn't changed, but assuming they're able to sap away sufficient heat from the external cabling that could solve that issue for running longer duration over-current situations. Assuming they managed to pull this off, that'd cover the external side of things I suppose. (We'll ignore that to do this they're having to waste many kW's of energy in exchange for speed).

However, the charge port and car side, at least on S/X, hasn't changed. I've seen the 3 chargeport as well, and I'm definitely skeptical. Going to have to do some testing there once I can access a v3 supercharger myself.

I still don't see speeds increasing significantly for S/X. The 100, maybe a little. But I've done a lot of testing over the years on the 18650 types... and we're not pumping 250kW into them safely. A few percent more, maybe, but not ~700A.

The 3 cells seem to have a lower IR than the 18650s (expected, but they did a better than expected job), and the cooling on the 3 packs gets more contact with each cell, making sucking away heat more efficient. The chemistry seems capable, at least. My worry remains the longevity of the charge port and related in-car components when we're talking about a 5x over-current situation for an extended period (minutes vs seconds).

Let's put it this way, a worn charge port or connector will make enough heat to melt it at 80A... and we're talking 700A+ for SpCv3. Also, the gen1 (manual) charge ports didn't have robust thermal monitoring, so there's no way to really do this safely there, either.

Summary, my stances haven't really changed much with regard to the car side. If we see any significant bump on the S/X side it'll be to the 100, and it will be modest, no where near what they're doing to the Model 3. Everyone else is probably SOL.

Rumors have been flying about an S/X refresh and they might be coming out with a pack built around the 2170 sometimes soon that will be able to take V3 currents.

I agree with you about wear. When new everything may be able to take the high currents with V3, but if anything develops a little extra resistance, that part is going to get very hot, very fast.

Back in 2016 I met a Tesla tech at a California supercharger who was replacing all the cables from the cabinets to the cars as well as some of the internals in the cabinets. He said that the combination of hot summers in the San Joaquin Valley was wearing out superchargers much faster than Tesla had thought and it was resulting in high losses from heat in the equipment.

He said they had a fix coming, and I haven't heard as much about slow rates at CA superchargers, so it sounds like they mostly solved the problem in California but will they have problems with the new higher power superchargers?

Too bad we don't have room temperature superconductors. That would solve the problem.

 

[Electroman]

Everyone else is probably SOL.

There was never a promise, nor is it reasonable to expect that older models can benefit from advances in technology. SD TV sets won’t get any sharper with HD transmission.

SpC V3 is not a software update.

So I don’t understand your SOL comment.

 

[wk057]

There was never a promise, nor is it reasonable to expect that older models can benefit from advances in technology. SD TV sets won’t get any sharper with HD transmission.

SpC V3 is not a software update.

So I don’t understand your SOL comment.

Because they said S/X would get improvements. They didn't specify any specific years or future... they said and implied that all S/X would get improved rates.

We will increase Model S and X charging speeds via software updates in the coming months.

Introducing V3 Supercharging

 

[supratachophobia]

Rumors have been flying about an S/X refresh and they might be coming out with a pack built around the 2170 sometimes soon that will be able to take V3 currents.

I agree with you about wear. When new everything may be able to take the high currents with V3, but if anything develops a little extra resistance, that part is going to get very hot, very fast.

Back in 2016 I met a Tesla tech at a California supercharger who was replacing all the cables from the cabinets to the cars as well as some of the internals in the cabinets. He said that the combination of hot summers in the San Joaquin Valley was wearing out superchargers much faster than Tesla had thought and it was resulting in high losses from heat in the equipment.

He said they had a fix coming, and I haven't heard as much about slow rates at CA superchargers, so it sounds like they mostly solved the problem in California but will they have problems with the new higher power superchargers?

Too bad we don't have room temperature superconductors. That would solve the problem.

I still think we are a ways off from 2170's in S/X. You have to remember how much Panasonic has tooled for the 18650, and how much the 2170's are needed for 3 and powerwall. This isn't a lightswitch type of thing. *Maybe* it is a thing where new sections of the gigafactory will slowly transition their area to 2170 production, but I don't believe it will just happen overnight. Panasonic would be left holding a huge bag.... of 18650's without a use.

 

[supratachophobia]

Because they said S/X would get improvements. They didn't specify any specific years or future... they said and implied that all S/X would get improved rates.


Introducing V3 Supercharging

I'll never forget Elon telling us back in 2011 that Tesla was going to come out with a car that could go 300 miles and a car that would cost 50k. The S could go 300 miles with an 85kwh pack (at 55mph....), but certainly not at the price-point of the 40kwh Model S. That guy has been feeding us half-truths at best for almost a decade. And at worst, lies. Yeah, telling people things that aren't true, that's called lying.