-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
First charge curve data I've seen.
- Thread starter Daniel in SD
- Start date
If the CT splits the battery pack in half for 400v charging, you would think that at the SC 250kW output, each half would see 125kW and thus have a flatter curve for longer? Sounds like even though was a V3 charger, the site was full so could still be power sharing. Look forward to seeing a more thorough charging video by out Of spec or such.
Most likely this is just how fast the current generation of 4680's charge. The Model 3 did get a huge upgrade to the charge rate early on so I suppose it's possible it could get better.
It does raise the question why they bothered having 800V capability. I guess it will charge faster at CCS stations than it would if it were 400V only.
It does raise the question why they bothered having 800V capability. I guess it will charge faster at CCS stations than it would if it were 400V only.
That’s basically what we’re seeing with this data.
If we consider that half of the CT 122 kWh pack is 90% of that 68 kWh Model Y pack, then the “125kW” equivalent is 139 kW’s into the Model Y. That 139kW point on the MY curve is around 20% SOC. If you go back to the CT data, it’s level until 20% then starts tapering. So with this charge curve, each half of the CT pack can take half of the 250kW until about 20%.
Projecting to a 800v charge configuration, and assuming this same MY 4680 curve, the CT would take 350kW only up to about 10% SOC, which is very low. I would expect (hope!) they can release a more robust charging curve to better use the capabilities of the V4 stations.
The battery bricks and cells don’t know/don’t care if the HV bus is at 400 or 800v. The benefit of the 800v configuration is reducing powertrain losses as well as reducing charging losses between the charger cabinet, stall/connector and the pack.
Last edited:
With a reasonable extrapolation, this charging curve meets Tesla’s stated charge rate spec of 138 mi in 15 min for the DM CT.
136 mi is 40% of the 340 mi total range. It should be able to charge from 0 to 40% in 15 minutes.
With a reasonable ramp up to 250 kW from 0%, maintaining 250kW until 20% and tapering to 150kW at 40% SOC should take a little under 15 minutes. It might even be able to go from 5 to 45% in 15 minutes.
136 mi is 40% of the 340 mi total range. It should be able to charge from 0 to 40% in 15 minutes.
With a reasonable ramp up to 250 kW from 0%, maintaining 250kW until 20% and tapering to 150kW at 40% SOC should take a little under 15 minutes. It might even be able to go from 5 to 45% in 15 minutes.
He goes from 14% to 54% in 17:30 so that sounds plausible.
Aero is horrible but it does look cool.
I do wonder what the plan was to get 500 miles of range.
scottf200
Well-Known Member
A Google sheet data geek put this doc together that shows the Y, CT, Lightning, & R1T - 'optimal' charging and c-rates
docs.google.com
CYBERTRUCK
EPA Range Model Y(optimal?),Cybertruck 400v,Ford F150 Lightning,Rivian R1T 75,4.4,330,123,2.8,340,131,2.4,320,135,2.4,328 75,123,131,135 kW,time,mile,min,mile/min,kW,time,mile,min,mile/min,kW,time,mile,min,mile/min,kW,time,mile,min 0% 1%,104,3,0.4,7.6,130,3,0.6,5.3,185,3,0.4 2%,108,7,0.8,7.8,157...
I created a charging model (overlayed gray dashed line below) based on the demonstrated Model Y 4680 charging power from Branden Flasch and extrapolated that to the CT pack size. Results are below.
Obviously I'm speculating, but until Tesla improves the charging curve, don't expect significantly faster charging than what's already been shown at either 400v or 800v stations.
Obviously I'm speculating, but until Tesla improves the charging curve, don't expect significantly faster charging than what's already been shown at either 400v or 800v stations.
So same C rate as MYAWD battery.
I don’t know if I’d really expect more out of these. Teslas been running those batteries for over a year in the MYAWD, and they’ve been having the cybertruck candidates rolling around for a while now. If there were improvements to be had, I think they would have found them by now
I don’t know if I’d really expect more out of these. Teslas been running those batteries for over a year in the MYAWD, and they’ve been having the cybertruck candidates rolling around for a while now. If there were improvements to be had, I think they would have found them by now
GhostSkater
Member
Keep in mind we have Lars on record saying it will do 15-85% in 18 to 20 minutes, while the current charge curve don’t inspire confidence on that, I don’t thing they would say it so specific like that if it wasn’t the case
For that to happen the average charging power in that window has to be 258 kW / 2.1 C or better
100% agree! Averaging >250kW over that window would be immense and is the far extreme from what we’ve seen thus far. This is a major disconnect and should give us hope.
If this first charge dataset isn’t a fluke and represents a typical curve now, the next critical information we need is the battery pack temps while charging. If it’s really capable of handling the thermal load of 2C charging as Lars stated, which I too want to believe, then the cell temps while charging now should be relatively low and controllable. What that means in practice is cell temps probably between 40 and 50C throughout the charge session and robust control over those temps in both cold and hot weather.
I think a good indicator of that robust control would be repeatable charging session performance, ignoring any impacts from stall or station power sharing. If we see a lot of charging data that overlaps, i.e. charge power at a given SoC that is similar to other sessions, then that is a good sign the thermal management system is just loafing along and sized for much higher loads.
GhostSkater
Member
Agree on all
The 4680 V1 really struggle to keep cool, even with charging power going down it remains at 60°C
I went deeper in other posts, but there is a fundamental thermal design difference between 4680 V1 vs V2, the first all the heat that flows through the electrode foils and the shingles ends up having to travel through the thin copper flower that is then crimped to the cell can
On V2, the copper flower is no more and the shingles and directly laser welded to the cap, which results in significantly lower thermal resistance, I did some estimation of the thermal resistance and it goes down by almost half if side cooled and almost 10x if bottom cooled
IIRC the CT will only split the battery pack in half when using the 800v V4 DC chargers. In the video above the guy was using a V3 DC charger and went from 14%-90% and that took 50 minutes. It took 40 minutes to go from 14-80%. It took 45 minutes to go from 14-85%. Is a 350kw V4 DC charger really going to more than cut that in half for 15-85% as Lars claims? I sure hope so but I'm not going to hold my breath LOL.
It is very odd that he said that. Maybe they’ll make fast charging 4680s by the time they start shipping to real customers. Maybe they just havent updated the charge profile for V2. I’m sure the original plan was to make it charge as fast as the other cell types.
One thing I just found out is that newer 2170s don’t charge as fast as the old ones. It seems that they aren’t prioritizing fast charging as much as before.
AlanSubie4Life
Efficiency Obsessed Member
Panasonic 2170L -> I think these charge just as well as the NC Panasonic 2170, though I guess I am not sure. Just have not heard complaints about them (there were some initial issues I think but they went away once heat pump was ironed out).
LG NMC 2710 -> These seem to have lower capacity and perform worse in EPA testing (which has not yet been published). No idea on their Supercharging behavior though Europeans would know.
I think they're prioritizing cost, probably. Hopefully the IRA restrictions will lead to quality batteries being brought back.
Anyway, I am looking forward to the CT getting 390kW up to 25%. And with the range extender installed, I expect it to charge at ~500kW up to 25% on the V4. That would put it roughly on par with Model 3 as required. Just need the software update then this thing will be a true Beast.
Going to need those V4 chargers rolled out everywhere really quickly. Hopefully in about a year or two they'll be as numerous as v3. And they had better be capable of 500kW; doesn't take much back-of-the-envelope calculations to see that is required!!!
GhostSkater
Member
The LG cells barely hit 250 kW in sub 10% SoC and immediately starts dropping steadily, no 250 kW plateau
You can find plenty of examples on Björn Nyland videos
