As slow as the 4680 is progressing it's likely true. Gawd knows the will is there but production is sometimes a swing and a miss. Hope it's just a tweak of the manufacturing process/equipment but hiring battery engineers sounds like they could be looking at design improvements.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Model Y - Gigafactory Texas Production
- Thread starter Ron J.
- Start date
As slow as the 4680 is progressing it's likely true. Gawd knows the will is there but production is sometimes a swing and a miss. Hope it's just a tweak of the manufacturing process/equipment but hiring battery engineers sounds like they could be looking at design improvements.
ZenRockGarden
Active Member
I agree the cell chemistry and size is already well known and the challenges have to be in manufacturing and pack-level performance.
If I had to guess, I'd say they have thermal and structural challenges getting the 4680's to work well as a pack. Maybe cooling those big thick cells is trickier than they thought despite the efficiency gains from the tabless design. Maybe impact safety concerns were found (note that the pack they have shipped basically took the original 4680 design and deleted cells down both sides, replacing them with what appears to be foam padding). I'd guess the troubles are more in the structural side, since we know they didn't get the weight savings expected - despite running fewer cells than planed. If you're having trouble with structural loads, the short term answer is to add bracing till you figure it out... which adds weight where you thought you'd be saving it.
Will be fun to see how it evolves over time. Given it's Tesla, I expect they'll sort it out, and we'll have LR structural packs next year ish
SageBrush
REJECT Fascism
Because you say so ?
I'm not sure about the fraction of unused volume difference between 2170 and 4680 cells but the height difference in 80 vs 70, or about 15%. If the wasted space is the same, the pack mass is ~ 15% closer to the centerline of the car. Elon has said that the difference is felt by the driver and I don't doubt his statement.
ZenRockGarden
Active Member
The point is that the MYAWD weighs within 1% of the MYLR, the only thing that's different is the battery pack, which is already mounted way down low in the center of the car... you're just not going to have a meaningful impact on weight distribution by moving 1% of the mass an inch or two. The end.
I thought the MYAWD was going to come with the privacy cover thingy?
Mrbrock
Active Member
So why wasn’t this happening from day 1? If it’s just a structural box that’s welding some steel together to make a case the same size as the 4680 pack. There’s a reason it’s happening “later this year”. To make sure the cars are identical to Fremont they need to make a new production line. The current bodies being produced have no floors. Adding a floor to the 2170 pack amd then the needed structural components to properly replace the 4680 pack will add weight. Adding weight means less range and then the Austin 2170 won’t be the same as the Fremont 2170. To make them the same you need a new line with new castings (compared to 4680) and bodies to match Fremont.
If it was as simple as you say it is, why hasn’t Fremont been upgraded to to this type of construction? Answer, it’s not as simple as making the 2170 structural and swapping between the two formats at will. There’s different connections and cooling and all sorts of things why they aren’t interchangeable.
I was hesitant to even post it here knowing how some act with this type of stuff. You can believe it or not, doesn’t really matter to me
No worries just found it questionable having been to both Tesla facilities and Rivian at best you will get a hello greeting in the front entry but beyond that an NDA must be signed. At least that was my experience.I was hesitant to even post it here knowing how some act with this type of stuff. You can believe it or not, doesn’t really matter to me
If the employees really like a prospect and want to work with the prospect they can become very personable and talk off the record especially when 1:1.
Today's Reuters news is even more interesting with Elon talking about a super bad feeling about the market and proposing a 10% Tesla layoff.
Last edited:
It seems entirely reasonable to speculate that the 4680 is the bottleneck for Austin right now. It also seems like this is important enough to have more than 1 group of people working on resolving this issue.
insideevs.com
Panasonic Ships 4680-Type Battery Cell Samples To Tesla
Panasonic confirms that it has sent 4680-type cylindrical battery cell samples to Tesla, ahead of series production.
insideevs.com
MontyFloyd
Member
MontyFloyd
Member
This is what I am thinking, a new battery box for Austin that does not use the battery to take part of the load.
Modularity is a tradeoff of build efficiency vs build flexibility.
From what I have seen and heard the battery box is self contained unit, and if not then very close to it.
Fremont is an old style car production line, it cannot be changed to sled type without costing a huge sum, and months of shut down to do rebuild.
The canary in the coalmine for the 4680 from a performance perspective is the mere existence of the MS and MX Plaids. If there was any tangible performance benefit, then I can 100% assure you that they would be doing everything in their power to shoehorn these in to those platforms ASAP.
Will that change in the future? Maybe... probably. But to even begin the discussion of improved performance and to a lesser extent, efficiency coming as a result of the 4680 is just silly at this time.
Will that change in the future? Maybe... probably. But to even begin the discussion of improved performance and to a lesser extent, efficiency coming as a result of the 4680 is just silly at this time.
There is a reason that the Plaids are using 18650's and not 2170's . Smaller cells, better performance , less heat and easier to dissapate heat.
The reason they didn't go to the 4680's are the same reason they didn't move the LR or P to the 4680 "SUPPLY". Everything may be 4680 in 5 years but today they can't even support the SR AWD.
Pansonic isn't due for mass production untill 2023 so I don't think we know anything about the 4680 or its performancc because of where it isn't being used.
Oh agreed. My point though is that if there was a significant performance advantage to using the 4680 right now, I have to believe that they would use whatever limited supply that they have on their halo models. This is all conjecture though at the end of the day.
What kind of sense does this make for a company that has a long line waiting for everything they can make? Unlikely that demand will dry up, so the only way it would make sense is if there will not be enough material to make what they currently have capacity for, which would be a very negative turn of events.
Edit: just read this, so makes more sense now. Elon Musk says Tesla will reduce salaried headcount by 10%, while increasing hourly employees
4680 was always a cost-cutting/saving measure that Tesla hoped would also provide some performance/range/efficiency benefits. The cost savings are easy to account for, but the latter aren't... I think it is pretty apparent that the performance aspects have not yet been realized and even for all of the cost savings, Tesla cannot afford the PR hit of releasing new LR and Ps that are "worse" than their predecessors.
We won't hear about 4680s again in those models until they figure this out.
We won't hear about 4680s again in those models until they figure this out.
ZenRockGarden
Active Member
4680 was about reducing production/manufacturing/assembly costs, not about max power.
Fewer bigger cells are cheaper to make and assemble.
For Plaid where cost isn't a priority? Huge racks of zillions of skinny cells are actually fine for max power and cooling surface area. Just not cheap to create.
My first thought was in a recession FSD take rate could crash when people aren't willing/able to add $12k to their order. FSD is a large contributor to their margins and cash flow. I don't think demand or cancelations will cause the backlog to dry up, but rather go from 12 to maybe 6 months?
Similar threads
