-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
-
On today's TMC Podcast, we ask the question "What is the Tesla Cybercab?". Join us on YouTube live at 1PM and participate in the chat.
A Model S owner hit 75,000 miles, still gets 93% of original range
- Thread starter EarlyAdopter
- Start date
I hadn't seen that anywhere before, but with this long range example plus the Roadster long term battery study I will now be answering the question of "how long is the battery good for?" with "a couple of decades if you drive the typical 12,000 miles a year."
Again, either I read it in the manual, on their web site, or Elon said it in one of his interviews. I can't recall exactly.
This language is imprecise. Can you tighten it up?
Elaborating: Isn't the battery technically "good" as long as it supports at least 1 (rated) mile of charge?
JRP3
Hyperactive Member
I wouldn't call that "good" myself. In truth the definition must be flexible, since it will have different meanings for different people. Those pushing the pack capacity when new may not feel the pack is "good" after a 20 mile range loss, others might be fine with 100 miles of range loss.
Vger
Active Member
Thanks so much for re-posting the link. I had missed that video, and found it fascinating!! Really glad to know one of the good Prof.'s grad students is working at Tesla now! Very interesting how Tesla is clearly put in a good light wrt battery knowledge, and Nissan and GM, not so much!
For as long as I remember the definition for end-of-life of a battery is when it has lost 20% of it's capacity. When manufacturers state their battery lasts for 500 or 1000 cycles, that means after that many cycles the battery can only hold 80% of it's original capacity. My Model S has 272 rated miles when charged 100%. When the official end-of-life of the battery is reached, it still has 217 miles on a full charge. That's more than enough for me. Part of the reason I got the 85 over the 60 is to have way more range than I need so I would not have to worry about it. Even a 30% loss (=190 miles) I would still be fine. Doing the math based on the 75k miles = 7% loss, I would be fine when I hit 300k miles.
omgwtfbyobbq
Active Member
Assuming degradation is linear for as long a possible, sure. I'd love to be able to take a battery in an S from 100% to whatever percent my commute can handle over several hundred thousand miles.
The problem is, and I think this is mentioned in the video JRP3 referenced, batteries generally don't degrade in a predictable manner. They'll degrade at a fairly consistent rate until they don't. If that change in degradation happens at 30% capacity, great! If that change ends up being at ~80% capacity, then the pack's lifespan is probably in the range of ~150-200k miles, which isn't bad, but also isn't great either.
The big difference in this context is obviously cost. Either way the costs work out favorably for the S, since the high end pack cost assuming you buy the $10k replacement is something like 6-7c/mile, but if the packs can maintain a consistent rate of capacity loss all the way down to ~30-50% capacity because of Tesla's BMS magic, then that's a bigger game changer since the pack cost drops to a couple cents per mile.
At that point, the cost to run an S on self-installed PV panels is something like 3c/mile, and Tesla has a huge jump in terms of IP on everyone else (similar to Toyota with their hybrid patents).
JRP3
Hyperactive Member
NCA chemistry tested beyond 2500 cycles shows no abrupt capacity loss and 70% capacity retention, which would be over 500,000 miles.
Page 20: http://www.embedded-world.eu/fileadmin/user_upload/pdf/batterie2011/Sonnemann_Panasonic.pdf
The video I linked shows that Professor Dahn's predictive methodology does in fact provide a way to reliably predict cell life cycle.
Page 20: http://www.embedded-world.eu/fileadmin/user_upload/pdf/batterie2011/Sonnemann_Panasonic.pdf
The video I linked shows that Professor Dahn's predictive methodology does in fact provide a way to reliably predict cell life cycle.
Ok. I still want to keep it short and sweet. How about:
Q: "How long is the battery good for?"
A: "Should still be good for 200+ miles of range after 20 years of typical driving."
Where "typical driving" is the national average of 12,000 miles per year.
7% degradation in 75k miles in 18 months does not tell you much about how the battery will fare on a typical driving of 75k miles in 6 years. Calendar losses and 6 summers will then become larger variables than the 400 odd cycles the pack has gone through.
All we can deduce for now is that the losses will be more than 7% in that typical scenario.
All we can deduce for now is that the losses will be more than 7% in that typical scenario.
JRP3
Hyperactive Member
Not necessarily. To put 75K on in a short time means doing deep discharges and full charges very often, which is harder on the battery than shallow cycles. It also means the cells have less time at rest to cool down internally the way a car that spreads out it's mileage over time would. We also don't know if his pack is in balance or reporting the full amount of range that is actually available. So at this point we can't really deduce too much more beyond what has been presented.
omgwtfbyobbq
Active Member
Correct if I'm wrong, but doesn't the video also shows that the cycling done in labs doesn't accurately predict real world lifespan due to the effects of storing cells at higher/lower socs and higher temperatures? Unless those NCAs were cycled under the realistic conditions described around the 5:50 mark of Dahn's video, which I don't think is viable since it would take ~7 years of real world testin to predict 7 years of real world use, or Panasonic used his high accuracy/prescision calorimetry to determine lifespan, which I'm not aware of any manufacturer doing, and something they probably didn't do in that pdf since they just reference the usual charge/discharge cycling, then we can't assume their batteries will last ~500k miles and still have 70% capacity.
Then again, I could be wrong. It could be that the S always keeps the pack at the ~77F testing temp and that cycling the NCA cells 2500 times in a short period of time will result in the same degradation as doing the same number of cycles over 7+ years, but I don't think either of those are accurate. Odds are, per Dahn's video, real world pformance will be worse than the original published results. IMO, the question at the heart of this is, how much worse will the real world performance be?
If Tesla's BMS keeps things close, great, we'll see 85kWh S sedans trundling around a decade from now on the original batteries with ~300-500k miles and ~200 miles of range. But, it's not certain how much the BMS will help battery lifespan in real world use. It will certainly help, and I doubt we'll see widespread pack replacement at ~100k miles, but will that help allow the pack to last ~200k miles, ~400k miles, or more, before it hits ~70% capacity? And what other conditions are required for that kind of lifespan (age, average temperature, average c-rates)?
Last edited:
JRP3
Hyperactive Member
It's important to understand that those test cycles were full 100% cycles, from 4.2V to 2.5V, something that Tesla never allows in the vehicle. Full charge/discharge cycles reduce cycle life, which means the referenced test results likely underestimate capacity retention in that respect. It's true we don't know the duration of that test, so on the other hand quick turnover cycling could overestimate capacity retention. Might be a wash. On the other hand that test was on three year old technology, improvements are likely since then. I'm less concerned with temperature effects since the pack is temperature controlled, and since Professor Dahn described the dramatic effects small percentages of certain additives have in reducing the effects of elevated temperatures. As for manufacturers using his predictive methods, remember that at least one of his former assistants now works at Tesla. Bottom line, even only 1000 full cycles to 70% is over 200K miles of life, longer than most cars last, so I can't see any rational for concern.
omgwtfbyobbq
Active Member
I wouldn't go so far as to say skepticism of your 500k mile claim counts as concern, but I don't disagree with most of your post.
I think the average pack lifespan driver's see, both in terms of mileage and age, will probably have an impact on Tesla's market share, especially once Tesla launches their mass market sedan, but that's just one of those things we'll figure out as time goes by.
I think the average pack lifespan driver's see, both in terms of mileage and age, will probably have an impact on Tesla's market share, especially once Tesla launches their mass market sedan, but that's just one of those things we'll figure out as time goes by.
JRP3
Hyperactive Member
Sure, but it's better to have as much accurate data as possible before hand, which helps us figure out true market share before the market figures it out. The main reason I can be so bullish about Tesla is because I know more about their technology than most people.
Similar threads
- Poll
