4680 Structural Battery Pack explained - How does it work?

[fritter63]

what’s the deal with that bracket?

Mount/housing for the tonneau motor?

Sponsored

 

[JBee]

Mount/housing for the tonneau motor?

I think that would be on the other side of the rear bulkhead, to the left. Actually I think the motor will be inside the vault roll to save space.

I think that frame might be the framing to attach the rear foldup seats and belts.

 

[cvalue13]

Mount/housing for the tonneau motor?

It’s a good guess, but tonneau is stored behind/below the cabin - under the bed ??‍

 

[C T Rick]

As mentioned in the original post. The Pepto Bismol pink looks awful close to what we use to use in fireproofing penetrations. I believe it was a 3m product that we just added water too and mixed up. Now we have caulking that is used to fireproof wall penetrations.

You guys are in a whole different league than my 48 years in the Plumbing field.

Now when it comes to plumbing out a CT camper, I'll add my 2 cents.

Keep up the good work.

Rick

 

[RVAC]

There's 3 tables in a post of mine further up that give you the pack dimensions etc.

Yeah, shouldn't have been lazy and just extrapolated from there.

A ~92" module length would result in 50/51 cells per column, in a 6 cell wide arrangement with 4 modules that is about 1212 cells. That yields ~115kWh, which is consistent with expectations. However for the module to be 92" long the pack would have to be a bit longer than that. So either there's a bit more length available or this configuration does not appear to be viable.

For it to fit in a 92" long pack a module length of 83.3" in your third example seems reasonable, that would mean 46/45 cells per column in an arrangement that has two modules be 8 cells wide and two 6 cells wide. That would yield a total of 1274 cells which is ~121kWh. However that would be 68" wide according to your table.

 

[cvalue13]

Yeah, shouldn't have been lazy and just extrapolated from there.

A ~92" module length would result in 50/51 cells per column, in a 6 cell wide arrangement with 4 modules that is about 1212 cells. That yields ~115kWh, which is consistent with expectations. However for the module to be 92" long the pack would have to be a bit longer than that. So either there's a bit more length available or this configuration does not appear to be viable.

For it to fit in a 92" long pack a module length of 83.3" in your third example seems reasonable, that would mean 46/45 cells per column in an arrangement that has two modules be 8 cells wide and two 6 cells wide. That would yield a total of 1274 cells which is ~121kWh. However that would be 68" wide according to your table.

don't forget total pack size vs 'usable'

regardless, this is horseshoes and hand grenades

 

[RVAC]

don't forget total pack size vs 'usable'

regardless, this is horseshoes and hand grenades

Good point about usable capacity, the 2x 8 cell wide and 2x 6 cell wide module arrangement might have to be 85" long instead of 83.3” for the minimum viable option, 47/46 cells per column would be 1302 cells (124kWh nominal, 117kWh usable). Otherwise 8 across all four modules it is.

 

[PilotPete]

I may be way off here, but I remember in college there was a recent discovery of the algorithm to prove how to best box a group of oranges (spheres). The math was a mess. Trying to calculate how many round things fit into a rectangular thing is not for your HS geometry class. So, I went to my CAD program and drew 4 circles, 2x2, with the second row offset to pack the maximum density. Oddly enough, 4ea 4680 cells with room for cooling (nearly 25% of the area is for cooling), takes up 8640 sq/mm. An 88"x70" pack could then hold ~1840 cells. And that looks like somewhere around 166kWh usable.

Now, I haven't had to do real calculus in years, so I cheated (a lot). I made some assumptions on the cooling and I stretched the pack size to where I thought it "might" be able to go. But 166kWh give you (at 3.67) 450 miles or (at 3.31) 500 miles. Is this too far into dream land?

 

[scottf200]

Yes please! It would be great to see your analysis on how many cells they can actually fit. I always had a strong feeling they could get enough to push the range over 550 miles

@JBee using @cvalue13's estimated pack measurements of 92x62 in, what capacity would that yield?

There's 3 tables in a post of mine further up that give you the pack dimensions etc.

We're thinking it's around the 115-130kWh range, but it needs to run pretty efficient to get over 300miles from it.

Yeah, shouldn't have been lazy and just extrapolated from there.

A ~92" module length would result in 50/51 cells per column, in a 6 cell wide arrangement with 4 modules that is about 1212 cells. That yields ~115kWh, which is consistent with expectations. However for the module to be 92" long the pack would have to be a bit longer than that. So either there's a bit more length available or this configuration does not appear to be viable.

For it to fit in a 92" long pack a module length of 83.3" in your third example seems reasonable, that would mean 46/45 cells per column in an arrangement that has two modules be 8 cells wide and two 6 cells wide. That would yield a total of 1274 cells which is ~121kWh. However that would be 68" wide according to your table.

Good point about usable capacity, the 2x 8 cell wide and 2x 6 cell wide module arrangement might have to be 85" long instead of 83.3” for the minimum viable option, 47/46 cells per column would be 1302 cells (124kWh nominal, 117kWh usable). Otherwise 8 across all four modules it is.

AlanSubie4Life on TMC heard 1366 4680 cells (video below) but it doesn't make much sense to him.
I did give him a link to this thread.
See:

Lots of talk in videos about 1366 4680 cells, does not make sense. Am I missing something? Misheard? Not divisible properly unless I am missing something…
1366/2 =683…which is prime. Not good.

7:30 here:


Also posted on Twitter by Baglino that for 400V, pack is split in half. Must keep balance!

https://x.com/baglino/status/1730391023799386480?s=20
Background:
I don’t think anyone ever worked out the WAY underperforming 67kWh Model Y AWD 4680 pack.

Summary post: '2022 Model Y 4680 Structural Pack is "Amazing", Says Munro & Associates'
2022 Model Y 4680 Structural Pack is "Amazing", Says Munro & Associates
Apparently 92s9p, 67kWh pack capacity. Depop 1 out of 9? No one knows (pack not light enough for that?) Only 81Wh per 4680 if fully populated; terrible, or software limited. Info needed here since 81Wh is lower than expected (91Wh might be better?).

I can think of a few configurations close to 1366, but there are limitations due to bandolier structure and module structure, on what is actually possible. Have not accounted for all that below:

1376 = 2*8*86 (361V/722V (max voltages, not how they will be referred to, seems to be about 86%, so 310V/620V))
Would require 94Wh per 4680
86 not divisible by 4…no idea module count…

1344 = 2*7*96 (402V/804V aka 350V/700V) same as Model 3 voltage.
96Wh per 4680
Maybe. How would it be built though? Depopulate bandoliers?

1344 = 2*8*84

1656 - 2*9*92 - two Model Y packs using same structure. Way too big. If it had Model Y terrible density it would be close, 135kWh. But still too big.

Anyway. If anyone has seen this worked out, please post.

What is bandolier size?

 

[RVAC]

AlanSubie4Life on TMC heard 1366 4680 cells (video below) but it doesn't make much sense to him.
I did give him a link to this thread.
See:

I'm thinking Cammisa might have meant to say 1386 cells? With two modules that are 8-cell wide and two modules 6-cell wide. That would be a pack longer than 92" though, could also be 1360 cells in four modules that are 8-cell wide.

 

[scottf200]

I'm thinking Cammisa might have meant to say 1386 cells? With two modules that are 8-cell wide and two modules 6-cell wide. That would be a pack longer than 92" though, could also be 1360 cells in four modules that are 8-cell wide.

FYI from GhostAndSkaterr post:

"More info we can extrapolate from the EPA data for the Cybertruck
Pretty much confirmed 220S6S battery, or 1320 cells"

 

[JBee]

FYI from GhostAndSkaterr post:

"More info we can extrapolate from the EPA data for the Cybertruck
Pretty much confirmed 220S6S battery, or 1320 cells"

As per the other thread, I can't get those numbers to work throughout the pack parameters.

--

So I can't quite replicate those exact numbers in the EPA document in my pack modelling.

That 170Wh/kg figure must be the Specific energy density at the whole pack level. So we can work out the battery mass from that being:

123kWh / 170 = 723kg (1593lbs) for the whole pack

Which sounds a little bit high, given that the 1400cells or so to make that number only weighs 497kg (1095lbs). But given it's a structural pack, those extra 226kg (498lbs) are not a bad use of mass.

I also don't know how they are getting the 816V nominal, as I can't get cell numbers to divide evenly into a split 400V pack. Is that maybe max pack voltage? The cell chemistry dictates the voltage, so this will need to be the same as other cells with that voltage.

So you can have 55S packs giving you 203.5V nominal each. You can the have 4 of those for 814V nominal pack. That gives you 1320 cells, and is exactly as predicted in the structural pack thread. That Hagerty video also mentions 1366 cells in the pack.

The cell capacity must have increased since the 4680 teardown, because with 1320 cells I get 114.7kWh in the pack not 123kWh. But that is only a 7% increase so easily doable.

 

[scottf200]

I also don't know how they are getting the 816V nominal, as I can't get cell numbers to divide evenly into a split 400V pack. Is that maybe max pack voltage? The cell chemistry dictates the voltage, so this will need to be the same as other cells with that voltage.

Not sure if this helps in your brainstorming. LLM didn't see the values lining up either.

 

[scottf200]

I also don't know how they are getting the 816V nominal, as I can't get cell numbers to divide evenly into a split 400V pack. Is that maybe max pack voltage? The cell chemistry dictates the voltage, so this will need to be the same as other cells with that voltage.

Tweet question to GhostAndSkaterr : "Are you assuming 25Ah cell?"

We can work it backwards, assuming the cells was the wrong way, but the document gives us energy directly

220S for 816 V is 3.709 cells nominal voltage, with 1320 cells and 122.4 kWh (816 V x 150 Ah), each is 92.73 Wh

From all that, each cells is exactly 25 Ah

Or also, 150 Ah and 6 cells in parallel also gives 25 Ah each

 

[JBee]

So I had a whole post of math which I've now condensed:

I think the 220S6P (or 4 modules of 55S6P) is the closest it's going to get to all metrics we know so far.

The configuration results in:

4 modules of 55S (203.5V nom) and 6P at 150.3Ah
In a pack that is 220S (814V) and 122.4kWh. (including 7% increase)
715kg (1576lbs) structural pack and 496kg (1093lbs) battery mass.

All the numbers line up as close as possible like that, including the 400V pack split for charging, and the physical dimensions required to fit it in the structural pack area of the CT with the buffers.

I will draw up a pack in CAD to see how it looks like. ?

(Edit the above now assumes 25Ah Cells @ 261Wh/kg)

Sponsored