Did Elon Confirm 350kW Charging For Cybertruck??

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,667
Reaction score
2,843
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
That's incorrect. The C rate depends upon the state of charge (and other factors).
As I have been stating clearly and as the data say for themselves - just not as much as you seem to think.

The fudging is arbitrarily picking 20%-80% when I continue to tell you I charge from 10%-60% (which is well above 1C).
It's hardly arbitrary. It is where most operate. That's the value of the ROT. There is a reason the region below 20% is marked in warning yellow on the energy display. I assume that most people respect that.

So you repeatedly tell me that you operate in the 10% to 60% region and I have always accepted that and explained the benefits of charging in that region. I know what the rate is. Can you, from your own data or experience, tell me what that rate is?

And a lot of experienced long-distance EV drivers charge like this whenever possible.
And most don't. Again you miss the whole point of this. While your personal preferences and practices are certainly of interest we mustn't assume that all drivers operate the way you do. I don't have any reason to suppose that I am atypical but, of course, I wouldn't assume that all drivers operate the way I do. FWIW the statistics on my super charging are:

Charge taken on: V_avg= 44.2727; V_sdev= 19.1942.

Starting SoC: V_avg= 29.2727; V_sdev= 12.3862; V_min= 16

Terminating Charge: V_avg= 73.5455; V_sdev= 17.7841; V_max= 90

Time to Charge: V_avg= 28.1818; V_sdev= 13.1287; V_min= 5; V_max= 43;

Charge Rate: V_avg= 0.980811; V_sdev= 0.167247

Thus the average charge I took on was 44% and the average time it took to do it was 28 minutes. Seems as if the rule of thumb is pretty good. The average rate was 0.98C. When I charge up to 90% it drops to 0.78C and when I charge to 29% (I don't know why I did that but I did) it goes up to 1.32C. Now those are respectively 2.5 standard deviations over and 0.9 standard deviations under the mean. If you "speak the language" you will understand the significance of that. Note that n = 11 which isn't huge.

Now just as I don't always charge from X% to Y% I am confident that you don't always charge from 10% to 60% so that's why I keep saying let's see your data. So let's see your data. Please.



As far as the Cybertruck, yes, thqt's why Elon has said max Supercharger power levels will continue to increase. And none of us know what the exact chemistry or charging capabilities of the 4680 cells will be. While I expect the thermal capabilities to be similar to existing cells it's possible they won't taper as soon or as much. We will have to wait and see.
There is, as I said in the previous post, little point in installing a 200 kWh battery that is capable of accepting an average 2C charge if you don't have a charger that can support that. A 250 kWh charger will not support that.

You are still dodging the questions as to how much faster a 3 will accept charge than an X and how you know that there are 30 - 40 miles range left when SoC indicates 0. I'd really like to know about those things. You also dodged the question as to whether the 3 charging data was taken by you.





Advertisement

 
Last edited:

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Messages
4,983
Reaction score
6,092
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
The last 20% of charging in a Tesla is tamped down, so it doesn't matter what size Supercharger you're at. It could take another half-hour or it could take another hour and a half - because of temperatures and cell-balancing.

I don't know why people continue to mention it.

And this argument about the Model X is dumb. It uses the older cooling architecture and so tamps down earlier than a Model 3 or Y. That doesn't mean it doesn't charge above 1C - it just means that charging faster comes with a penalty of heat you then have to shed. So you do want to use less of your battery pack if you want to charge less. Less thermal flexing.

And there's a reason that companies (like my Zero) actually don't use the top 20% of the battery pack at all.

-Crissa
 
Last edited:

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,667
Reaction score
2,843
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
Sometimes it is interesting just after looking at the 46th post in a thread to go back and check No. 1. Here that post asked if 350 kW was for real and what that might mean in terms of new batteries etc. The answer is it wouldn't mean much.

If we assume a charging profile which is constant at rC to some break point 0 < b < 1 then the average charging rate, from 0 Soc to 100% SoC is (r*b + 1)/2 so that if r = 1 and the break point is also 1 the average rate is 1. The X has a 100 kWh battery and charges at average rate of about 1 C. thus r = 2.5 and b = 0.4 which is just about where the charging curve breaks for this car. Now if we have a 200 kWh battery and a 250 kW charger we have a maximum possible charge rate of 250 kW, r = 250/200 = 1.25 and b has to be 0.8 in order to achieve overall rate of 1C. That is probably way too late for the break point at least for today's cells technology. It seems that 0.5 is more suitable and so we would have (1.25*0.5 + 1)/2 = 0.8125C as the average charging rate. If we go to 350 kW then r = 350/200 = 1.75 and (1.75*0.5 + 1)/2 = 0.9375C. Thus if this profile were used (and it won't be but it will be something that resembles it) it would take 74 minutes to fully charge (0 to 100) a TM CT from a V3 (250 kW charger) and 64 minutes to do it with a 350 kW charger. A 350 kW charger isn't sufficient to charge the TM CT at the rate of the X!. And the charge rate is only 0.93C. If the new cells can tolerate more than 1C, on average, an even bigger charger than 350 kW would be needed to take advantage of that. Another way around this would be if the new technology battery were less sensitive to charge rate at high SoC. Were no taper required or if it didn't need to start until later we could do better. If we could start the decline at 57% we could achieve 1C. If we could break at 68.6% then we could achieve 1.2C with a 350 kW charger. So we hope that it is something like this that the new cells will deliver i.e. higher tolerable average charge rate with later onset of tapering.
 
Last edited:

JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Messages
340
Reaction score
340
Location
6000
Vehicles
Cybertruck
Country flag
I do mostly agree that a 350kW SC won't have a big impact on anything other than a TM CT with 200kWh pack. That is based on the current taper and max charge C though. It would make sense that there is stiil significant SC charging capacity left if using a 250Kw charger for any smaller pack like a 75-150kWh one. We don't know that many details on the 4680 yet, but I think its reasonable to expect a 20-30% improvement in both with the new tech.

The question is how much of the taper and max C are due to thermal issues rather than battery chemistry, meaning that like in the plaid, charge rate increased over the previous model, despite using older cells, simply because of the improved HVAC system in the plaid. (Which btw is external to the pack itself) On top of this the tabless design of the 4680 not only lowers the electrical resistance, meaning less heat is generated in the first place because of the shorter path, but it also reduces the thermal resistance to get that heat out through the ends of the cell. Getting heat out of a 4680 with a "tabbed" cell design would mean the middle of the cell could "cooking" whilst the outside where its being cooled could be "freezing". The heat has to get out from the middle of the cell and is retarded by a small delta T through the rest of the cell (which is also heating electrically) as it makes its way to the outside. I'm not sure exactly what the difference could be, but I can imagine 10's of degrees being possible, which is also the main reason for the tesla staying with multi small diameter cell packs for so long with their intercell cooling method.

With tabless cells this changes in that the interior and exterior of the cell essentially have the same heat path and cool at similar rates, meaning the internal cell temperature is no longer the charge/discharge constraint. It also reduces the electrical heat generated because of a shorter path. This is a bigger deal than most believe it is and we might see up to a doubling in the charge/discharge rate, and with that a requirement for a 350kW SC even for smaller pack sizes.

I've always been a bit amused that the tabless battery actually has heaps more "tabs"!
Although I'd be happy with the TM CT performance as discussed before, I won't say no to even faster charging! :giggle:
 
Last edited:

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,667
Reaction score
2,843
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
TruckElectric was kind enough to post some charging data on the new Plain and of course I thought this wonderful as it is a new data set for me. In contemplating it I looked for a different way of presenting the data in order to facilitate comparison between vehicles. Like it or not, here it comes. I started by fitting polynomials to the SoC vs time data reported by the car. Using this data I can prepare a picture such as this one.
Array.jpg


What this picture shows is the number of minutes it takes to charge my X LR+ between SoC1 and SoC2 at a V3 charger assuming that all charges follow the pattern that the car/charger worked out on the day I did the charge from which this picture is derived. That's a bit of a tall assumption, of course, and why an ensemble is really more telling.

This picture is interesting, of course, but not susceptible of easy interpretation (I don't think) so I rewickered what it says into the following form:
Vehicles.jpg


What this picture shows is the fraction of an hour required to charge a fraction of the battery's capacity. IOW looking at the column of dots at 0.3 on the X axis we see that 30% charge can be added to the battery can take anywhere from 0.13hr (8 minutes representing an average charge rate of 2.25C) to 0.31 hr (representing an average charge rate of 0.96C) and where one is in this column of dots naturally depends on the level at which you start to take your 30%. If you start at 10% you will see the highest rate and if you start at 50% the lowest. The solid line of the same color of the dots is the linear least squares fit to all the dots. It's slope is an overall representation of the X and a V3 charger. It suggests an average charging rate for this combination of 1.34 C meaning that on average you ought to plan on about 0.45 minutes for each % of charge taken on.

The other solid lines represent the fits for other Teslas. The dots are not shown for these as it would make the picture too busy. Look at the one labeled X LR V2. That shows what to expect from this car and a V2 charger. It suggests average rate of 1.12C meaning 0.54 minutes per %. I had originally stated the ROT as 1C with 0.6 min because 1C is more convenient but it looks based on both data sets that should either distinguish between V2 and V3 chargers or split the difference and use 0.5 minutes per percent recognizing that this will increase the error for both but after all it is a ROT, not a precise estimator.

This leaves the bottom line. Actually it is two lines that overlie one another representing the Plaid data from the article posted by TruckElectric from Motor Trend and LR 3 based on data posted by HaulingAss earlier in this thread. He won't say where he got it. In any case based on these data sets the average charging rates of the Plain and the LR3 are identical. They show average rates higher than the X at 1.53C (they differ by 1 in the 3rd digit to the right of the decimal). This implies charging time of about 0.4 min/%.

Note that the finding that the Plaid and 3 are about the same is contrary to the findings of the Motor Trend article. They found the Plaid charging appreciably better than the 3. In comparison to HaulingAss data that conclusion is not supported by MotorTrends published data.
 

JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Messages
340
Reaction score
340
Location
6000
Vehicles
Cybertruck
Country flag
Thanks Adjelange for the detailed post.

So given that we know that the Plaid has the original 18650 pack (but with better chemistry but not tabless) I think its safe to assume that most of the charge performance increase is from the improved thermal management of the pack, along with stricter cell quality control and tolerances?

I expect the 4680 thermals and charge rates to improve by that much again, so say 1.8C or 033min/%, making the 200kWh TM pack a very ICE comparable travel experience, with breaks primarily dictated by driver health rather than vehicle energy capacity.

Can you add that to your graph for comparison?
 

Crissa

Well-known member
First Name
Crissa
Joined
Jul 8, 2020
Messages
4,983
Reaction score
6,092
Location
Santa Cruz
Vehicles
2014 Zero S, 2013 Mazda 3
Country flag
The 4680 tabless merely gets the cells to the same point as the 2170 cells. Thermal management might be easier, but we really don't know.

Chemistry makes a big difference in charge rates, but since they're using the same pattern as the Model 3, either they're not taking advantage of that, or are saving it for a point where they've proven the reliability.

-Crissa
 

JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Messages
340
Reaction score
340
Location
6000
Vehicles
Cybertruck
Country flag
The 4680 tabless merely gets the cells to the same point as the 2170 cells. Thermal management might be easier, but we really don't know.

Chemistry makes a big difference in charge rates, but since they're using the same pattern as the Model 3, either they're not taking advantage of that, or are saving it for a point where they've proven the reliability.

-Crissa
As per my previous longer post I'm confident the tabless 4680 design will be much better electrically and thermally. Here some info why:

 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,667
Reaction score
2,843
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
I expect the 4680 thermals and charge rates to improve by that much again, so say 1.8C or 033min/%, making the 200kWh TM pack a very ICE comparable travel experience, with breaks primarily dictated by driver health rather than vehicle energy capacity.

Can you add that to your graph for comparison?
We need to take note that loading battery at 1.8C implies that it can be fully charged in 1/1.8 hr. If the battery is a 200 kWh battery that means the charger must deliver 200 kwh/(1/1.8) = 200*1.8 = 360 kW and it must do it continuously - no taper possible. Tesla has no charger that big.

No problem putting the line on the graph though.

Vehicles2.jpg
 

JBee

Well-known member
First Name
JB
Joined
Nov 22, 2019
Messages
340
Reaction score
340
Location
6000
Vehicles
Cybertruck
Country flag
We need to take note that loading battery at 1.8C implies that it can be fully charged in 1/1.8 hr. If the battery is a 200 kWh battery that means the charger must deliver 200 kwh/(1/1.8) = 200*1.8 = 360 kW and it must do it continuously - no taper possible. Tesla has no charger that big.

No problem putting the line on the graph though.

Vehicles2.jpg
So I suppose with a 1.8C 4680 pack (being the upper limit) even a DM CT with 150kWh will need a 350kW SC with taper. But anything under that still can do it with a 250kWh SC. I suppose we could do a graph the other way around that shows at what C and capacity a pack requires a 250kW or 350kW SC.

Model wise the MY is meant to get 4680s, and probably the M2, but the MY probably can't fit more than 120kW (of the 4680s) and the M2 maybe 75kWh? So the majority of cars globally will not have large high-C 4680 packs and only need 250kW SC. That is assuming CT sales don't takeoff after release and are seen on road, and also because of the value proposition compared to a MY. (6 adults and 3cbm bed vs 5+2kids and no boot)
 

ajdelange

Well-known member
First Name
A. J.
Joined
Dec 8, 2019
Messages
2,667
Reaction score
2,843
Location
Virginia/Quebec
Vehicles
Tesla X LR+, Lexus SUV, Toyota SR5, Toyota Landcruiser
Occupation
EE (Retired)
Country flag
Remember that battery_pack_kwh*r where the average charge rate is rC is the average power required of the charger. If taper is necessary there will have to be parts of the charge where the rate is higher than rC so that other parts can be run at less than rC.
 

Advertisement





 


Advertisement
Top