Is zero % range loss after 27 months a red flag?

[bcsteeve]

New EV buyer here. I put down a deposit on a used Model S today... I won't be taking delivery for a few weeks though. Its from a private dealer.

I was thinking about battery degradation and I asked for a photo of the monitor showing the expected range when fully charged. He did so and I can see it is 407km. So then I looked up what the rated range for that year/version was (2015 P85D) and it was... exactly 407km.

At first, I was like, "bonus, no degradation!"... but then I started to wonder how that's possible. It has over 55,000km and is over 2 years old, so shouldn't it be showing some degradation?

In the ICE world, this sort of thing (if there was a proper analog for it) would have me wondering if there was some ODO manipulation going on. Is there anything to be worried or suspicious about? Or should I just be happy that the battery is going strong?

 

[dennis]

New EV buyer here. I put down a deposit on a used Model S today... I won't be taking delivery for a few weeks though. Its from a private dealer.

I was thinking about battery degradation and I asked for a photo of the monitor showing the expected range when fully charged. He did so and I can see it is 407km. So then I looked up what the rated range for that year/version was (2015 P85D) and it was... exactly 407km.

At first, I was like, "bonus, no degradation!"... but then I started to wonder how that's possible. It has over 55,000km and is over 2 years old, so shouldn't it be showing some degradation?

In the ICE world, this sort of thing (if there was a proper analog for it) would have me wondering if there was some ODO manipulation going on. Is there anything to be worried or suspicious about? Or should I just be happy that the battery is going strong?

My 2 1/2 year old 2015 P85D with 27K miles has exactly 1% degradation - 253 to 250 range at 100% charge. Not to worry.

 

[Shock-On-T]

New EV buyer here. I put down a deposit on a used Model S today... I won't be taking delivery for a few weeks though. Its from a private dealer.

I was thinking about battery degradation and I asked for a photo of the monitor showing the expected range when fully charged. He did so and I can see it is 407km. So then I looked up what the rated range for that year/version was (2015 P85D) and it was... exactly 407km.

At first, I was like, "bonus, no degradation!"... but then I started to wonder how that's possible. It has over 55,000km and is over 2 years old, so shouldn't it be showing some degradation?

In the ICE world, this sort of thing (if there was a proper analog for it) would have me wondering if there was some ODO manipulation going on. Is there anything to be worried or suspicious about? Or should I just be happy that the battery is going strong?

I don't know of any hacking stories, so I reckon it's genuine.

 

[mal_tsla]

This value is estimated anyway. It's not a good way to determine degradation. Think of it like an iPhone battery. It charges to 100% even after a few years, but that 100% sure seems to drain faster than it used to. The car is making some guesstimates as to how the battery will behave when it is discharged, but the only way to know for certain is to actually discharge it (drive it 400+km).

That said, there probably isn't any degradation at that age of battery.

 

[bcsteeve]

Thanks. I was looking at a different car that was only a couple of months older and same mileage... and it came up with a range of 394km. I guess that's the only reason I was raising an eyebrow over this.

But clearly the consensus is nothing to worry about. Good enough, thanks!

 

[Tiger]

Thanks. I was looking at a different car that was only a couple of months older and same mileage... and it came up with a range of 394km. I guess that's the only reason I was raising an eyebrow over this.

You can find out actual range only by driving, because same as a mobile phone battery indicator, it will show full when full, but an old battery will drain faster. You can only trust what the odometer shows after driving from full to empty battery.

 

[DFibRL8R]

You can find out actual range only by driving, because same as a mobile phone battery indicator, it will show full when full, but an old battery will drain faster. You can only trust what the odometer shows after driving from full to empty battery.

This is not entirely correct. The Model S (just like a phone) can display percent charge and can charge to 100% even with a degraded battery. But the OP said the photo was of the expected range (rated range) which is a number the car calculates based on the available energy in the battery and typical consumption rates with average driving (something around 300 Wh/mile IIRC). So indeed it would suggest the battery he is buying has had little to no degradation despite the age/miles. This is very impressive since most of the data I have seen shows about a 5-8% loss of range in the first year then minimal degradation after that. Great if Tesla has improved this issue but the car the OP is looking at could just be an outlier. Unless of course the battery had to be replaced and it had fewer miles than the rest of the car.

 

[tccartier]

New EV buyer here. I put down a deposit on a used Model S today... I won't be taking delivery for a few weeks though. Its from a private dealer.

I was thinking about battery degradation and I asked for a photo of the monitor showing the expected range when fully charged. He did so and I can see it is 407km. So then I looked up what the rated range for that year/version was (2015 P85D) and it was... exactly 407km.

At first, I was like, "bonus, no degradation!"... but then I started to wonder how that's possible. It has over 55,000km and is over 2 years old, so shouldn't it be showing some degradation?

In the ICE world, this sort of thing (if there was a proper analog for it) would have me wondering if there was some ODO manipulation going on. Is there anything to be worried or suspicious about? Or should I just be happy that the battery is going strong?

Can you right click on the image and then scroll down to properties and see when the picture was taken.

 

[bcsteeve]

Unless of course the battery had to be replaced and it had fewer miles than the rest of the car.

I gave the VIN to the Vancouver centre and he confirmed no work has been done on the car, so I assume that would include battery replacement.

Can you right click on the image and then scroll down to properties and see when the picture was taken.

I hadn't thought of that! Did it just now, and it says "Date Taken" is August 23, 2017 at 9:30am.

 

[Musterion]

I was thinking about battery degradation and I asked for a photo of the monitor showing the expected range when fully charged. He did so and I can see it is 407km. So then I looked up what the rated range for that year/version was (2015 P85D) and it was... exactly 407km.

So here is a thought. Which screen did he take a photo of? If smaller screen, maybe you can get a photo that includes the odometer. And also I just checked something on my car: if you switch the settings to display to "ideal" miles rather than "rated" miles, both the small and large screen battery indicators increase dramatically and there is no indication on the screens themselves about which setting is being applied (in my case it increased the indicator from 266mi to 332mi, about 25%!). Last question is whether the photo was of the Consumption/Trip graph (on either small or large screen) which also displays some other mileage indicators but these are different from rated and ideal miles and reflected average and instant projections. Based on my experience and what you can read here, it is highly unlikely there is truly 0% degradation, so be careful. But it would be helpful to see the actual photo to give better advice. Good luck.

 

[eye.surgeon]

Perhaps it's because the displayed range on full charge is in no way representative of the health of the battery and is not nearly precise enough to estimate degradation. This myth has been disproved multiple times over but still persists.

 

[bcsteeve]

@Musterion... it is rated. He also showed me ideal, which was far higher (I think that was over 600km).

Anyway, I consider this closed. I'm satisfied they aren't pulling anything over on me and like a couple have said, it isn't something I can really go by anyway. Is what it is

 

[Musterion]

That's cool. Congratulations and I'm sure your next post is going to be along the lines of "Woohoo! Why the heck did I not do this sooner???"

 

[tomas]

Many software updates have 'reset' the rated range estimator. So, right after software update, it can look 'like new', then after a couple of weeks, it settles back to reflect degradation. Could be photo was after recent update.

 

[mal_tsla]

The car does not know the available energy in the battery. It estimates it, based on how the battery responds while charging and discharging.

The % full is based on the voltages of the cells. The miles is based on recent observations on how the voltage dropped while driving vs the current voltage.

In the end the only way to know for sure is to drive the car to a low state of charge. Just because the battery exhibited an expected voltage drop rate on a recent trip from 90% charge to 70% charge does not mean it will behave that way down to 10%.

That difference is, in essence, what degradation is, and this is why you cannot measure it based on the remaining miles displayed.

This is not entirely correct. The Model S (just like a phone) can display percent charge and can charge to 100% even with a degraded battery. But the OP said the photo was of the expected range (rated range) which is a number the car calculates based on the available energy in the battery and typical consumption rates with average driving (something around 300 Wh/mile IIRC). So indeed it would suggest the battery he is buying has had little to no degradation despite the age/miles. This is very impressive since most of the data I have seen shows about a 5-8% loss of range in the first year then minimal degradation after that. Great if Tesla has improved this issue but the car the OP is looking at could just be an outlier. Unless of course the battery had to be replaced and it had fewer miles than the rest of the car.

 

[DFibRL8R]

The car does not know the available energy in the battery. It estimates it, based on how the battery responds while charging and discharging.

The % full is based on the voltages of the cells. The miles is based on recent observations on how the voltage dropped while driving vs the current voltage.

In the end the only way to know for sure is to drive the car to a low state of charge. Just because the battery exhibited an expected voltage drop rate on a recent trip from 90% charge to 70% charge does not mean it will behave that way down to 10%.

That difference is, in essence, what degradation is, and this is why you cannot measure it based on the remaining miles displayed.

In my experience in driving 99k miles, I have seen very little short term variability in the rated range estimates when I charge to 100% (or 90% etc). Might vary plus or minus 2 miles estimated but not enough to conceal significant degradation based on a falsely elevated displayed estimate (which was the OPs concern). By your description the car would show very inaccurate estimates due to excess fluctuation based on the most recent trip. That would be annoying and I can't imagine why Tesla would build an algorithm that gave useless data like that. I have not observed that behavior. Have you?

 

[bcsteeve]

I've been doing some reading on estimating battery capacity, including an MIT dissertation, which may or may not be of value but it was interesting. The above author(s) who say it can't be accurate are correct that there are no practical method for precisely *measuring* battery capacity. However, there are proven methods for *estimating* with reasonable accuracy. I suspect Tesla employs those methods. Day 1's estimate would be the least accurate in one way, theoretically (statistically insignificant data) but the most accurate practically (new battery has predictable characteristics). As time goes on, practical assumption becomes less accurate but statistical assumptions become more accurate.

From what I read, if you had a robot drive the car in a closed environment, the estimated range should be extremely accurate through the entire lifespan of the battery. In the real world, it wouldn't be accurate even if they could directly measure capacity, because there's no way to know for sure what conditions the car and driver will encounter.

But that should mean (and this is my own interpretation here) that the indicator is *relatively* accurate and it should be a reasonable tool for indicating degradation, if not what it is actually there for (range).

I'd be interested to read a science based opinion and explanation for why it wouldn't be practically useful for that purpose.

 

[mal_tsla]

I'm not following how a recent trip could introduce "excess fluctuation" with a healthy battery. I'm not suggesting the car uses your wh/mi efficiency in this calculation.

What I AM suggesting is that the car tracks how much power is observed to flow from the battery as the charge level decreases (voltage). If the battery discharges less kilowatts than expected for the estimated charge level drop (%), that's indicative of battery degradation and should show up in the display in the car. However, the battery is only assumed to behave the same throughout the discharge cycle. Unless it has been actually observed to do so, it's an unknown to both driver and car.

We have all seen the smartphones turn off with 20-30% indicated charge left. The BMS in the phone sees a sudden voltage drop and declares the battery is "dead" regardless of previously estimating it was at 20%. Similar situation here.

Bottom line - if u want to know how many miles the battery will drive, you need to drive it that many miles.

In my experience in driving 99k miles, I have seen very little short term variability in the rated range estimates when I charge to 100% (or 90% etc). Might vary plus or minus 2 miles estimated but not enough to conceal significant degradation based on a falsely elevated displayed estimate (which was the OPs concern). By your description the car would show very inaccurate estimates due to excess fluctuation based on the most recent trip. That would be annoying and I can't imagine why Tesla would build an algorithm that gave useless data like that. I have not observed that behavior. Have you?

 

[mal_tsla]

The displayed miles is range at the EPA consumption rate. For our X that's 346wh/mile. Alternatively it can be set to display ideal miles. Both are very optimistic vs. real world consumption, but good reference points nonetheless.

We know for sure that with lithium ion batteries, it's not possible to "measure" the capacity except by discharging and tracking how many electrons flowed. We can only estimate.

For that display, the car is probably estimating, based on charge level (voltage) and capacity history (secret algorithm, but must factor in observed capacity during drives to be at all accurate) how many miles it would get at a fixed average consumption rate as stated above.

The voltage is absolutely measurable, so that's a constant. The chosen hypothetical wh/mi efficiency is also a constant. The only variable is the Actual Capacity, and that's unknown unless you actually try to discharge. So instead we are using the capacity estimate. Thus, degradation is not likely to show up in that display unless the driver routinely drives from high SOC to low SOC, which would boost the confidence of that estimate.

I've been doing some reading on estimating battery capacity, including an MIT dissertation, which may or may not be of value but it was interesting. The above author(s) who say it can't be accurate are correct that there are no practical method for precisely *measuring* battery capacity. However, there are proven methods for *estimating* with reasonable accuracy. I suspect Tesla employs those methods. Day 1's estimate would be the least accurate in one way, theoretically (statistically insignificant data) but the most accurate practically (new battery has predictable characteristics). As time goes on, practical assumption becomes less accurate but statistical assumptions become more accurate.

From what I read, if you had a robot drive the car in a closed environment, the estimated range should be extremely accurate through the entire lifespan of the battery. In the real world, it wouldn't be accurate even if they could directly measure capacity, because there's no way to know for sure what conditions the car and driver will encounter.

But that should mean (and this is my own interpretation here) that the indicator is *relatively* accurate and it should be a reasonable tool for indicating degradation, if not what it is actually there for (range).

I'd be interested to read a science based opinion and explanation for why it wouldn't be practically useful for that purpose.

 

[tomas]

You know, most posts in this thread have devolved into technical argument about the practicality of estimating range. That is not helping OP. @bcsteeve , I think it is a red flag. I've had model S since 2012. The rates range may or not be accurate... but it is consistent. My car new showed 272 miles rated range at 100% charge. It was advertised at 265. Now, almost 5 years later, it is around 250. It is amazingly consistent. If I were to charge 100% 3 times and look immediately after completion, it would end at 250 +/- 2. As I posted before, the only time I've seen aberration is after installing new software updates, when it would go back to 265. Within 10 days, it would be back to what it was before update. Not all software updates do this. But a few must have messed with the internal formula and reset all the parameters.

I think you need to check this in person and get an explanation from seller. It is not normal for a car this age to show no degradation. I have no idea how seller could possibly alter the number, but they could maybe have given you inaccurate photo.