Speak EV - Electric Car Forums banner

1 - 13 of 13 Posts

·
Registered
Joined
·
59 Posts
Discussion Starter #1
I picked up a new 90d in August last year. A couple of times I charged to 100% and was getting a reading of 280 miles on a 100% charge. (Possibly a tad more, I think 284?)

After that I set the charge limit to 80% and was getting 226 miles at most when fully charged at 80%.

As the months passed and the temperature dropped, this started dropping by a couple of miles every so often. I thought it probably had something to do with the lower temperature.

Fast forward to today, having covered 21k miles since August last year (of which 95% are motorway miles), I charged the car at home overnight and the reading is showing 214 miles at 80% charge. I recently charged at a SC a couple of times to 100% (having got there with approx 50-60 miles capacity remaining) and at full charge the reading was 260 miles.

FYI, initially I would only use superchargers but only charged to 80%. This lasted a couple months but then I started charging at home and work with maybe one charge every 2 weeks at a SC. (2 phase at home and 3 phase power supply at work.)

Other than where I have said I charged to 100%, the charge limit has remained at 80%.

The car goes into Tesla WD near the end of the month and this is one of the things I've mention to the service desk. They will hopefully investigate. Are they likely to be as concerned as me and will attempt to fix the issue? Or will they take the opposite approach?

Why is the range dropping at this rate?

Has anyone else had a similar issue? What did you do?

Is there anything I can do to recover the lost capacity? Am I doing anything wrong?

I'm no expert but this amount of battery degradation seems excessive to me.
 

·
Registered
Renault Zoe 50
Joined
·
19,946 Posts
It's just an estimate of range, not actual battery capacity. There have been many posts on this in the past.

A common concern which is dealt with by displaying 'energy' rather than 'distance' in the settings. :)
 

·
Registered
Joined
·
59 Posts
Discussion Starter #3
Good to know, thanks.

My concern remains though as presumably had the reading displayed energy rather than range, the energy would have dropped too? (I'm presuming the software interprets the energy and converts to range? Therefor if energy is reduced the range displayed reduces too?)

Or am I barking up the wrong tree??
 

·
Registered
Joined
·
133 Posts
Most batteries show an early 'degradation' of a few % which then levels off to virtually zero for a number of years, which is why Nissan can afford to give an 8 year battery warranty. There are posts here regardin Tesla battery capacity .
 

·
Registered
Joined
·
38 Posts
There is some 'received wisdom' that doing a deep discharge (e.g. down to about 10% or less) from time to time can help with the perceived drift in estimated range. There is some talk of how this helps with'rebalancing the pack' which seems like mumbo jumbo to me, much more likely it's an effect of getting more accurate inferred measurements - helping the car's systems keep an accurate idea of how many kWh have gone in and come out, and how many miles that equates to.
So if you are tending to re charge to full from 40/50/60% most of the time, you might find that a deep discharge & re charge makes the estimated range look a bit closer to what you are expecting. Nothing has changed in the state of the battery - it's just the calculation that the car makes.
(Please don't flame me - you're welcome to disagree - I don't know the systems well enough to be confident that this is real - but it's consistent with my own observations as a model S owner, and with some of what I know about measurement in electrical systems ...)
If you've got the patience for it, the sticky post about battery life is very useful & interesting reading.
 

·
Registered
Joined
·
6,865 Posts
@Grimupnorth I tried this, specifically driving from 100% to 0% in one go. It made no difference to the cars predictions.

I have however seen some anecdotal comments from U.S. owners over on TMC that have traded in 85 against 90 variants, and that they indeed did see some initial degradation (which is to be expected) but it is more noticeable in the packs with the newer chemistry.

Adding silicon to the cathode was in some ways a fairly bold move, as it theoretically does have downsides wrt to lifespan. (My understanding, and I bow to any battery chemists out there, is that the silicon expands / contracts more compared to the graphite in the cathode, so micro-fractures develop and reduce the cathode's efficacy.)

I'm sure this stuff was lab tested, but lab testing is still not 100% accurate, especially calendar aging effects. Jeff Dahn, who Tesla signed up last year, was a leading expert in resolving accelerated testing procedures/validation.

Ultimately the cells in the 90 packs are still relatively young, so it will take a few more years to see if this is appreciable on a fleet wide scale in real world use, hopefully the success seen in the older / higher mileage 85 packs does continue onto the 90 packs. It would be somewhat ironic if there is a crossover point where an 85 had lower degradation rate and hence gave better range than a 90 in an older example of similar vintage/mileage.
 

·
Banned
Joined
·
4,201 Posts
Why is the range dropping at this rate?

Has anyone else had a similar issue? What did you do?

Is there anything I can do to recover the lost capacity? Am I doing anything wrong?

I'm no expert but this amount of battery degradation seems excessive to me.
Have you actually noticed a drop in usable range day to day whilst driving??

Worrying about battery degradation is a well trodden path all EVers suffer at some point.

Go on to the Leaf, Zoe etc forums and your find exactly the same question been asked :)

Clearly some batteries do fail early, but those like hens teeth. The Tesla 'rated range' predictor is about as useful as the Guessometer in the Leaf.....Pretty much useless. Unless your actually noticing a drastic drop in range affecting your real life journeys I would forget about trying to worry about changes in rated range, your EV ownership will be much less stressful :).
 

·
Registered
Joined
·
2,813 Posts
Battery balancing isn't mumbo-jumbo - even things like electric hand tools have balancing controllers in their battery packs, because for any battery with series cells having them out of balance (ie. one cell at a higher state of charge than another) loses capacity in the whole battery. You can't charge beyond the point where the highest cell reaches 100% (leaving the other cells not fully charged) and you can't discharge below the point where the lowest cell reaches 0% (leaving charge in the other cells that you can't use).

However, in the Tesla case balancing occurs primarily when doing a charge to 100% (known as 'top balancing '). This can clearly be seen in the car's behaviour, and is also easy to analyse as it's industry-standard technique.

The supposition is that discharging near to zero and then fully re-charging, rather than cycling in a narrow range, helps calibrate the state-of-charge estimation. This is almost certainly true to some extent, but since the algorithms are proprietary, secret, and have changed from one software version to another, it's very difficult to be sure of the details.

Doing such a cycle and seeing no change doesn't prove that there's no effect: if the calibration happened to be spot-on beforehand, then recalibration isn't going to help.
 

·
Chartered Engineer
Joined
·
1,635 Posts
As the others have said above the total range estimate is just that and varies depending on how energy efficient the recent driving history has been. It's better than on some EVs but it's still a rough estimate and initially is usually optimistic.

I've been tracking my car (a 90D) using the data displayed on the trip meters and over 40k miles 100% indicated SOC seems to pretty consistently equate to 75kWh, this hasn't altered so far. There seems to be buffers top and bottom so 100% indicated SOC isn't 100% actual - my guess is there's a 5kWh buffer built in at the top based on the charge levels - and there appears to be the same at the bottom plus some bricking protection (to prevent complete discharge). So far I haven't noticed any degradation and even if I switch to the miles display (I usually run on %) it dropped initially but has gradually crept back up, I suspect because my driving style has adapted to the car.
 

·
Registered
Joined
·
268 Posts
Individual readings are not super reliable. I'd track it over time and see what your average is. At 80K miles I'm at an average of about 3% loss on my S. But if I look at individual charges the number can be all over the place.
 
  • Like
Reactions: Simon Mac

·
Registered
Joined
·
2,813 Posts
As the others have said above the total range estimate is just that and varies depending on how energy efficient the recent driving history has been. It's better than on some EVs but it's still a rough estimate and initially is usually optimistic.
Unless you are referring to the 'projected range' on the energy screen, I don't think that's true. The range numbers shown on the main (speedo) display are significantly different from those shown by many other EVs.

Many EVs show you a range figure which is "how far can you go, assuming your future driving pattern matches the recent past". Tesla shows a figure which is "how far can the car go, assuming a standard driving pattern" (with a choice of two standards - rated or typical in Europe, rated or Ideal in N.America).

In a sense, the other EVs are trying to tell you what you want to know - how far can I go with my driving patterns under the conditions I am about to encounter? - but falls down because this is impossible to do accurately (predicting the future), and it is difficult for the driver to compensate for the inaccuracies since it's not obvious how much of any problem has already been factored in. Conversely, Tesla's number is a fixed translation of the amount of energy in the battery and so significantly more accurate, even if less directly useful - you need to correct it yourself for weather, driving style etc.. However, these corrections are easy to understand and apply, making the Tesla approach more useful IMO.

However, even the process of estimating the energy content of the battery (the starting point for both display approaches) isn't 100% accurate, and trying to use it to measure battery degradation is problematic as you are unlikely to be running the same software version for your 'before' and 'after' numbers, making the comparison unreliable.
 

·
Registered
Joined
·
6,865 Posts
@arg, further muddying the waters is it varies by battery/drive variant, and even firmware version.

I have to say I'm not particularly happy with Tesla and the settings on my car. For what ever reason they have set that fixed rate (in kWh/mi) artificially high (more than an equivalent launch 85, despite the launch 60 being the more efficient car), and it plays havoc with the range assurance and route planning. :rolleyes:

So now I drive in % mode, but I can't escape the pointless warnings about slowing down to reach my destination. I've learnt to ignore them but others in the car (esp. wife) notice and it just reinforces the whole range anxiety problem :(
 

·
Chartered Engineer
Joined
·
1,635 Posts
Unless you are referring to the 'projected range' on the energy screen, I don't think that's true. The range numbers shown on the main (speedo) display are significantly different from those shown by many other EVs.
True - I was referring to the energy screen estimate. As you say the one on the dash display is based off test cycle figures which is often optimistic which is why I just use %.

As it happens though as my driving style has adapted I'm getting closer to the Wh/mile figure the numbers on the dash are based on and if I'm pushing things at all I calculate out the likely energy usage to reach the next charger and ignore the warnings. So far the numbers are working out well suggesting any degradation is very small.

Incidentally I only push things if I'm heading for a Supercharger, if I'm aiming for a public one I leave a good buffer in hand because the reliability of getting a charge is lower.
 
1 - 13 of 13 Posts
Top