Tesla battery day announcement

[Hillhater]

, the older the car the more issues arise, lets say you have a million mile battery, you want a car that will also last a million miles, with minimum repair/maintenance. That I think is sustainable approach

A cheap million mile car, is even more delusional than a m’mile battery..
Think about it, . The average car does less than 25k per year, , so you expect a basic car to run for 40 years, with “minimal maintenance “ ?
Even if they could build a drive train to last that long reliably, what shape is the suspension , shocks, wheel bearings etc going to be in ?...let alone the bodywork or the interior !
If nothing else, after 10 years the whole “style” becomes dated and unfashionable, which if nothing else simply kills any sale value.
Then there is the small issue of technology changes ...what will battery or drive train systems be like in 10-20 years time ..never mind 40 yrs ?
Some think we will all be using autonimous, community,“share” cars before then, with no need for ownership cost or reliability concerns.

 

[spinningmagnets]

I currently have a 1991 Toyota 4-cyl truck, 170K miles (I am the second owner), and I have owned several older cars in my past. The tires, brakes, shocks, wheel bearings, and suspension joints will "wear out" the same on an EV compared to a gasoline car. When restoring an older classic car, the gas tank and fuel pump can be easily replaced. If they are rare and difficult to find, they can be refurbished. On an EV motor, the wear components are two shaft bearings. Easily replaced by a mechanic, or the garage enthusiast.

In my mind, the comparison comes down to a worn out battery ($14,000?), compared to a worn out engine and automatic transmission. Lets just say forty years from now, a distant relative passes away and leaves you their two classic 2021 cars. You want to restore them both. The gasoline car has 300K miles on the engine and auto trans. They need to be rebuilt, but are old enough that nobody makes the parts to rebuild them (like the bronze transmission bushings on a 1969 Norton Commando being rebuilt today).

When it comes to a worn-out battery pack, the controller is somewhat agnostic. It will work with any battery that provides the volts and amps that it demands. If there are too many handshake protocols to bypass (Rich Benoit), then perhaps you would need to replace the controller too, but...I think EV hackers will allow DIY packs to be retro-fitted to old EVs. Lets say the original pack provided 400 miles per charge, but you really only "need" 100 miles. That means the replacement pack has some room to work with.

A million-mile battery sounds great, especially if it is not more expensive than the current packs. Part of the reason the MM battery is interesting is that...retired packs are getting a second life as home power back-ups, when the original pack erodes down to 80% of its new range...

The Tesla packs are desirable, but the Nissan Leaf packs are snapped up with aggressive enthusiasm. When the EV car body has reached a dilapidated state, a true MM battery pack will still be able to be sold to somebody, even if it only has 30% of its range left. Worn batteries have been recycled to recapture the desirable chemicals, and the EV motor/controller units have already been retro-fitted to other vintage cars.

 

[hallkbrdz]

Unless Tesla improves their parts and distribution to aftermarket repair shops, that's going to be really expensive. I've thought about opening up a EV only garage, but working on Tesla's is a sketchy proposition.

??I've been using an old Tesla 15s module for 5 years now for DR. They could surely make it easier to use them, but they're usable right now.

I'm speaking about their cars in total. They had a parts list you could view (at https://epc.teslamotors.com/#/catalogs), but it had no prices and there was no easy online ordering like any other manufacturer. That's gone. Now searches show a parts site at https://epc.tesla.com, but it is at best an insecure site that my browsers won't connect to! When I checked last year you had to call a Tesla service center and explain why you need a part. Then IF they deem you worthy they will let you buy one IF it is available. Oh - you are not an owner but a 3rd party shop... click.

 

[JackFlorey]

I'm speaking about their cars in total. They had a parts list you could view (at https://epc.teslamotors.com/#/catalogs), but it had no prices and there was no easy online ordering like any other manufacturer. That's gone. Now searches show a parts site at https://epc.tesla.com, but it is at best an insecure site that my browsers won't connect to! When I checked last year you had to call a Tesla service center and explain why you need a part. Then IF they deem you worthy they will let you buy one IF it is available. Oh - you are not an owner but a 3rd party shop... click.

Ah, I see what you mean. I haven't experienced that yet. There are two shops down in the valley here that specialize in Teslas; it would be interesting to see if they have the same experience.

 

[Hillhater]

A Snip fron som of the TBD feedback reporting ..

Some researchers estimate that price parity, or the point at which electric vehicles are equal in value to internal combustion cars, is reached when battery packs cost $140 per kilowatt hour (kWh).
Tesla's battery packs cost $219 per kWh in 2019, according to electric vehicle consulting firm Cairn Energy Research Advisors.........

 

[whatever]

hey there hillhater, miles dont mean much to me, lets go with km,
its not a million mile battery its a 1609344km battery, lets just say a
1.6million km battery. My car currently has 440,000 km on it, and runs like a new car.
Just basic maintanence on it ( 1999 ford falcon), this particular model popular as a taxi,
frequently said to get 800,000km before retirement from taxi fleet.
That said a really well maintained car could get over a million km, yes you would need
replace shocks etc but not unchievable. 1.6million km well might be stretching it a bit.
40year old cars as you know alot of them are worth a damn fortune now,
and highly sort after
You honestly believe australians are going to share their cars with other people??????
and let the car drive itself?? there will be a revolution before that happens!
maybe when hell freezes over
drivetrain of tesla is really simple, keep changing the oil and those gears will last a very very
long time. Most people dont have time to maintain their cars, but a well maintained car will
pretty much keep going and going. Its a disspearing art maintaining vehicles.
did a pic of the brushed rotating pwm used in early days of motors

The loss of efficiency using brushed motor/ brushed rotating controller, would be made up for
by the increase in battery efficiencies........decent range/long life cheapish, extremely do-able.

 

[Hillhater]

Whatever,
Plenty of vehicles run happily past 1m km.(.most of them diesel trucks),..but sure some cars also (but not common)
And ..Plenty of cars last 40+ years with basic maintenance on original running gear.
..BUT, very few cars last 40 years AND cover. 1mkm, of daily use, without some serious rework.
Most high mileage cars are relatively yourng, with long distance commercial use (Taxi’s, Rep’s lease cars, couriers, etc)...it not easy to rack up a million kms just taking the kids to school and shopping !
Do you plan to keep your 20 yr old car for another 20 years ?..... If not, ask yourself why not if its “still running like new “.
And yes SOME 40 yr old cars are very valuable , but only the rare, unique, “collectable” ones.
How much is a base model 1980. Ford Falcon or Laser, or a GM Astra, etc worth now ?
Tesla’s drive train is still a multi shaft gear train with a differential assembly incorperated, and is an unknown quantity yet , only 10 yrs and few high km examplles, but remember that motor runs high rpm ( up to 18,000 rpm) which means the input gears, seals bearings , etc are also spinning at that rate,
So dont be surprised if some issues crop up on well used examples (Tesla do not report failures)
But there is much more to Tesla’s ..or any EV ..than just a battery, motor, and transmission.
There is an enormous amount of associated power electronics and control systems..many thousands of individual components, each with a finite life and probable failure rate, and some of those have already been seen
Did you catch the “memory” capacity limit issue ?
Where a particular module had a known design life that equated to less than 5 years of use in the ModS,...before causing the car to default to “limp” mode....requiring a $5000 dealer replacement unit.
ICE’s are a known quantity with predictable failures and flaws, but EV are a new game, lots of different technology and components of unknown reliability...and VERY FEW experienced service support shops, spares , etc,.. compare to ICE’s.
If you really want to minimise maintenance on any EV, you would not consider a Brushed system of any sort.

 

[neptronix]

So here are my quick takeaways:

Simplified manufacturing, simplified chemistry to end product, and form factor to support structural use. All reducing their costs (good) and allowing for faster production.

From a non-Tesla perspective however (unless they actually sell cells direct at some point), the structural use in a honeycomb pattern with a strength adhesive sounds like the cells will be near impossible to remove without destruction. Thus making them essentially useless since it is a structural member, even other EV conversions - unless you can use the entire body/battery part as is.

Time will tell...

If the cells are as great as they say, it would be worth figuring out what kind of solvent dissolves the adhesive, and then reselling individual cells.

I'm pretty impressed with what Tesla's got up their sleeves though.

 

[Hillhater]

If the cells are as great as they say, it would be worth figuring out what kind of solvent dissolves the adhesive, and then reselling individual cells......

Has this not already been figured out for the M3 pack glued cells ?
It is likely a similar material ?

I'm pretty impressed with what Tesla's got up their sleeves though.

I would be more impressed if they were making at least a little improvement in cell energy density !
And has anyone seen a figure for the new cell weight ?
My best guess is 350 gms. ?

 

[hallkbrdz]

At this point everything is a WAG! It's still in development, they don't even know yet what the final specs will be. :wink:

 

[neptronix]

I would be more impressed if they were making at least a little improvement in cell energy density !

It sounds like there's a notable improvement in cell density on the way. We just don't know by how much.
Seems there should be a bump in power density also.

Tesla is promising a 620 mile range in the new Roadster. They're not known to roll back their promises. I believe they have a cell with significantly improved density in the works if they can jam that much capacity into a car of that size.

 

[Hillhater]

It was obvious that they avoided mentioning energy density, but referred only to “Range improvements” .!
That can be achieved with more capacity in the “PACK”
They can get 5+ times the capacity in the bigger cell using existing cathode/Anode chemistry , and with the “tabless” design they should be able to increase the “Power density” due to the improved current path giving lower IR.....but you dont really need more power with the existing pack capacity’s.
The Sandy Munroe uTube vid, falsely stated that they could get almost double the capacity in the same space on a M3...that is wrong !
Energy density improvements should come with the modified Anode and cathode, but that is not due for a few years.
As for promisses.. every promised release date, delivery target, cost target, etc...has been missed, ( remember the $35k mod 3 ?) ........so i will keep an open mind on how much of what was “promised” is available and when.

 

[Pajda]

I would be more impressed if they were making at least a little improvement in cell energy density !

This is very good point. actuaIly if we look at the presented gain values for the new 4680 cell we can easily translate that "5x more energy" actually mean lower energy density than its predecessor 2170 because the volume of 4680 is 5.5x bigger. And so "6x more power" also means only an insignificant gain. :? It should be noted that similar numbers were published by Tesla when releasing 2170 format and this cell finally also have the same energy density at best compared with 18650.

From my point of view the most promising technology is tabless electrode design, where the main practical benefit is not the reducing the IR (which is "sufficiently" good already in 2170) but the significantly better heat transfer thru the longitudinal axis of the cell which finally allows "bottom cooling/heating" only. The removal of the Tesla cooling pipe snake bring significant gains in all aspects (cost and pack energy desnity). If fact I think that after a few years we can evaluate actual TM3 battery pack desing as the classical example of over engineering.

 

[whatever]

thanks hillhater for detailed reply some good points, yes aware of memory issue, too much data wore out the chip.
To be honest i'm totally in awe of tesla/spacex and elon musk. The guy is a historical figure of our time, and will be remembered
for a long time to come, us mere mortals will all disappear into oblivion.
Who would have thought to use 18650 cells in the first place to run a car, considering there were so many companies producing all sorts of cells at that time for evs, they are of course moving away from 18650, but I guess it was the most common cell size ( of quality) available at the time. The fact that they used those cells and had success is quite the achievement. Its all pretty mind boggling.
Theres going to be heaps of issues of course with maintanence, but that does of course leave a niche place for entreprenuers to fill.
Plenty of folks will spring up on youtube with 'how to fix tesla problems" type vids.
The battery design will evolve for a long time, there is so much can be done in that area. I suspect at some point the battery electrolyte itself will be used as cooling fluid.

 

[Hillhater]

Musk will certainly be remembered for many achievements.
But Remember , Tesla is not Musks brainchild, nor is the decision to use 18650 cells in their pack designs, Musk “bought” into Tesla after the company was established and those basic design decisions were made.
The rational behind the 18650 choice has been well documented by Tesla themselves..basicly, it was cost, quality and availability
Likewise, the concept and detail logic of cell improvements , manufacturing methods etc are most likely done by some of the many specialists employed by Tesla

 

[spinningmagnets]

the significantly better heat transfer thru the longitudinal axis of the cell which finally allows "bottom cooling/heating"

I recall some research a while back that compared using active cooling to the sides of cylindrical cells, and then cooling the sides of flat-pouch cells, and then cooling the tabs of flat pouch cells. I seem to remember that cooling the tabs of the cells resulted in the most even temperatures of the internal active battery materials.

It affected cell longevity a lot more that I would have guessed, because the hotter portions of the cell interior had more resistance, and the parts that were only warm provided most of the current and cycle life. The hotter portions had more decay in the chemicals. All of these factors added up to cells that did not last as long as a similar-volume cell that used tab-cooling.

Perhaps Tesla will actively cool only the tops and bottoms of these 4680 cells?

 

[Pajda]

Perhaps Tesla will actively cool only the tops and bottoms of these 4680 cells?

Actually I am very interested in presented Tesla CTV (cell to vehicle) design. They showed us that they would like to remove side cooling (cell cans will be glued together and acts as a supporting structure). But this approach leads to the combination of this three problems to solve:

1) top/bottom cell cooling
2) top/bottom cell terminal connection
3) top/bottom cell mechanical fixation to the pack case

My guess is that the top of the cells will be used for electrical connections of both cell terminals with some mechanical fixation to the case and the bottom will be used for cooling and also mechanical fixation to the pack case.

 

[whatever]

I guess another change might be going to flat pouch cells instead of spiral wound eventually, as the cylinder cells leave alot of empty space between cells, I'm guessing its faster to spiral wind them ( after viewing that video its damn fast to wind a cell).
looks like end cooling will be the way to go, with both spiral and flat cells, its a good idea.
I've always thought if your cells get hot your taking too much current from them, I would have thought even if cooling them your still taking more current than they are comfortable with, better to increase the capacity so no cooling necessary. It is interesting they still have good longevity when being actively cooled.

 

[wineboyrider]

the significantly better heat transfer thru the longitudinal axis of the cell which finally allows "bottom cooling/heating"

Perhaps Tesla will actively cool only the tops and bottoms of these 4680 cells?

Don't forget they also announced that the new battery format will be structural as well!

 

[Hillhater]

Perhaps Tesla will actively cool only the tops and bottoms of these 4680 cells?

Actually I am very interested in presented Tesla CTV (cell to vehicle) design. They showed us that they would like to remove side cooling (cell cans will be glued together and acts as a supporting structure). But this approach leads to the combination of this three problems to solve:

1) top/bottom cell cooling
2) top/bottom cell terminal connection
3) top/bottom cell mechanical fixation to the pack case

My guess is that the top of the cells will be used for electrical connections of both cell terminals with some mechanical fixation to the case and the bottom will be used for cooling and also mechanical fixation to the pack case.

In the presentation, they actually said there would be no “features between the cells” with only structural epoxy around them. Also cells would be bonded to the top and bottom of the pack to make a rigid structural init.
“Like a honeycombe sheet”
On a separate point...
Musk stated that they already have packs of these 4680 cells in cars ,...presumably under test ?
Also that the MEGA casting machine is in Freemont , already producing front and rear one piece cast sections...
But for which vehicle design ?
I cannot imagine they would tool up massive die sets without a final vehicle design ?
Is there a Mk 2 Model Y in the pipeline, or maybe the $25k Tesla is being born ?

Further, with a vehicle constructed from 2 cast sections and a battery module, ....what happens in the event of a relatively minor crash ?
For passenger protection, i assume there will still be collapsable zones, so some of that “one piece” structure will have to deform.
How do you repair or replace these major components economically ?

 

[jonescg]

My previous employer was seeking to do all the things Tesla is doing now (except Tesla has competent engineers and a vision, and my previous employer ...didn't. http://www.capebouvardtechnologies.com.au/tech The structural battery thing is kinda dumb really. It's totally unnecessary but if you were to do it, larger cells make a lot more sense.

Axially cooling the cells with a baseplate is a good move too, however with 1/6th the internal resistance (guessing here) they won't generate nearly as much heat and won't need as much cooling anyway. Indeed for most markets they'll only need to pre-heat the cells.

 

[boars]

My previous employer was seeking to do all the things Tesla is doing now (except Tesla has competent engineers and a vision, and my previous employer ...didn't. http://www.capebouvardtechnologies.com.au/tech The structural battery thing is kinda dumb really. It's totally unnecessary but if you were to do it, larger cells make a lot more sense.

Axially cooling the cells with a baseplate is a good move too, however with 1/6th the internal resistance (guessing here) they won't generate nearly as much heat and won't need as much cooling anyway. Indeed for most markets they'll only need to pre-heat the cells.

Perhaps the cooling has something to do with fast charging and the same system can be used to pre-warm the batteries?

Curious as to why the structural idea is dumb though.

 

[jonescg]

Any large box with a bit of depth and some internal bracing will be rigid AF. If the battery is built with rigidity internal, simply bolting it into the car makes it a structural member. In effect, Tesla's eventual realisation that fully encapsulating the cells is a good idea for corrosion inhibition, prevention of rubbing or fretting, and generally improved robustness makes for a rigid monolith which in turn adds rigidity to the vehicle chassis.

Liquid cooling in a battery pack is a massive headache, as the risk of a small leak damaging the whole battery is real. So any move to get away from that risk is a wise one.

 

[TheBeastie]

Here are a few articles from Arstechnica on Tesla's battery day/promises who I believe try harder than most of the clickbait crap out there...

https://arstechnica.com/cars/2020/09/how-tesla-plans-to-make-batteries-cheap-enough-for-a-25000-car/
https://arstechnica.com/cars/2020/09/heres-what-tesla-will-put-in-its-new-batteries/

And just for the sake of completeness here is Tesla's own complete stream of Batteryday on YouTube, that recap barely covers anything that went on the full video, IMO.
https://www.youtube.com/watch?v=l6T9xIeZTds
[youtube]l6T9xIeZTds[/youtube]

As you probably guess I am a bit dubious on it all, Elon promised years ago that cobalt usage would be "virtually nothing", but now years later its "we are working on it" technology...