3.0 Battery Longevity

[ecarfan]

As an interesting contrast to the Roadster 3.0 battery data that @bolosky has been so helpfully compilingi in this thread, a group of Dutch Tesla owners have been compiling data on Model S/X battery degradation and Electrek just reported on what they have found, see Tesla battery degradation at less than 10% after over 160,000 miles, according to latest data

Quote from that article: “The data clearly shows that for the first 50,000 miles (100,000 km), most Tesla battery packs will lose about 5% of their capacity, but after the 50,000-mile mark, the capacity levels off and it looks like it could be difficult to make a pack degrade by another 5%.”

That is great news for S/X owners. I would like to see our 3.0 batteries behave similarly.

 

[bolosky]

The old Roadster batteries were also pretty good, although not that good. For instance, my (second) original battery lost maybe 12.5% of its initial CAC in over 80K miles. I suppose I could produce a set of graphs for them, too, since I have the data in the logs that I got to look at the 3.0s. That's partly why I've been so disappointed with the 3.0 cells: I got spoiled by the original ones.

 

[dpeilow]

Quote from that article: “The data clearly shows that for the first 50,000 miles (100,000 km), most Tesla battery packs will lose about 5% of their capacity, but after the 50,000-mile mark, the capacity levels off and it looks like it could be difficult to make a pack degrade by another 5%.”

That is great news for S/X owners. I would like to see our 3.0 batteries behave similarly.

They are generalising across many different types of Model S packs and chemistries. The thread here Early 75/75D pack degradation seems to give a differing view, but that has gone quiet in terms of real information recently.

In the meantime, I received a message about a week ago to say that Tesla was still looking into the contents of my letter, but no new information as yet. This was via the original route rather than the VP of Engineering route.

 

[bolosky]

I just fully range charged my car. Here's the state of charge screen. Note that the AR (amp hours remaining) is 189.54, while the CAC (calculated amp hour capacity) is 186.01. One of these is not correct .

I also noticed that it reported being at 100% charge on the SOC screen once it got to 186 AR, even though it kept charging to 189.5. So presumably the percentage is based on the CAC (but obviously capped at 100%).

I guess this is why range charging causes the CAC to move up. Of course, I really still don't know what number is accurate.

The VDS reports 309 ideal miles.

 

[dpeilow]

Something funky is going on. 309 miles is slightly higher than I would expect for that CAC drop, given that the cars started at ~350 on day 1.

 

[bolosky]

Something funky is going on. 309 miles is slightly higher than I would expect for that CAC drop, given that the cars started at ~350 on day 1.

It's only a little off. My new battery full range was 344, so it's now at 309/344 = 90%. In CAC space it was about 214 and now is 189.5, giving 189.5/214 = 89%.

Also, I didn't do the usual let the car sit after charging to see the ideal range number, so it may be somewhat inflated by that. I did a second range charge and let it sit overnight and wound up with something like 295, but the VDS also showed 96% SOC, so it had discharged somewhat while it was balancing overnight.

 

[bolosky]

Adding in three months' worth of data for my car (#670). The car spent maybe 7 weeks of that time out of service (mostly waiting for tie rod ends to come in from England, and then also for various other parts; it's having problems unrelated to being electric, but instead because it's a high mileage sports car). The big flat place on the days graph at 190Ah is when it was in the shop for five weeks.

The good news is that the current CAC (188.5) is a value that the car first hit on November 20, so it's gone six months without the CAC declining (overall; it bounced around a lot in the interim). However, I wouldn't get too excited about this just quite yet, though. My CAC climbed a bunch from a series of range charges associated with various long drives.

Still, this is the longest period without a large decline I've seen, and it's getting close to the point where you could claim that the trend line is bending some in the right direction. Another few months of this, or another car doing it at the same place (Dave?) and I might become a smidge less pessimistic.

My standard mode charge is showing just over 200 miles, and yesterday's range mode charge was at 305.

I'm not including the zoomed by-miles graph, because that hasn't changed.

 

[Timothy]

Do you want more data to add to your graph? If so, tell me how to collect it and where to send it.

 

[bolosky]

Do you want more data to add to your graph? If so, tell me how to collect it and where to send it.

Absolutely!

To collect it, put a USB jump drive with a folder called "VehicleLogs" at top level into the USB port on the Roadster (in the 2.X, it's under the center console behind the shift buttons). It will take maybe 10 minutes to download the logs. When it's done, there will be a file a couple of levels down that's named after the date plus a few digits and some other stuff (like: 201805200236.tar.gz). PM me and I'll give you instructions on how to send it to me, and I'll take it from there.

 

[Timothy]

Got the log. Sorry, but I am green at this. How do I do a private message?

 

[redan]

Got the log. Sorry, but I am green at this. How do I do a private message?

Click on the name of @bolosky to the left under the picture of their Roadster and select "Start a Conversation."

 

[Timothy]

Thanks!!

 

[dpeilow]

Last week I sent an email to my original contact on this topic asking for an update. It has been over 2 months since submitting the letter and 1 month since I had last had an email stating they were still looking into it. The response I got back was not really encouraging:

Thanks for the email. I have asked the service team to give you an update as quickly as possible.

I have just replied back with the following:

Dear X,

I didn't hear anything yet from the Service Team. Did you mean at a global level or locally? As we stated in the letter, this is an issue affecting owners across continents and feel it needs to be looked at by corporate colleagues, rather than on a case-by-case basis. I would be grateful if you could let me know if the letter went to HQ and who has looked at it there.

Personally, now that we have the warmer weather and I am using my Roadster more regularly, I am seeing the CAC drop by between 0.5 and 1 Ah each time I use the car. This was expected and is in line with the behaviour others (who use their cars all year round) have been experiencing. It is not really acceptable regarding the long-term health of the car, however.

Therefore as it is an issue experienced across the 3.0 fleet, we feel it is more of an engineering rather than service issue.

Regarding the more focused letter that was sent to the VP of Engineering at @Kerios and I spoke to him in Cambridge - I have never even had an acknowledgement of its receipt.

My CAC is now down to 208 from 212 when I last posted. Nowhere near as bad as some, but that is only after a few trips to the beach and work in the recent sunny weather here. I did one range mode charge in that time, taking it down to 40 ish miles with a small (10 mile) top up midway. I will pull the logs again and send them to @bolosky.

 

[bolosky]

Adding one new car, #1180 (thanks Timothy) and updating #33 and #209 (thanks slcasner & msjulie!)

Not a lot of good news here. #33 now holds the record for the lowest recorded CAC in the study at 183.6. It's among the oldest cars in the study, has higher mileage than most cars, and doesn't seem to have a lot of range charges, so there are three strikes against it. I suspect that several range charges would bring the CAC up some, but it's hard to say how much. Regardless, any idea that cars would stop declining, or even have a knee around 185Ah is out the window.

#209 is doing a little better than #33 when compared at either the same mileage or the same age, but that may well just be because #33 is driven more. It's pretty much a middle-of-the-pack car.

#1180 is toward the bottom both by miles and by days. Looking at it, most of the effect seems to be a drop of about 7 Ah in a few days when it was new. The owner said that he'd only taken one long trip in it, and I bet this is when. (Really, I'm sure, because you can see lots of mileage then.) This seems like the effect where if you run the battery very low you get a big drop (kind of the opposite of range charging). If it weren't for this drop, it would be middle of the pack. I don't think that the battery was damaged by the long drive/low SOC, I think it's just strangeness in the CAC algorithm.

Waiting, driving and collecting more data is all we can do. Thanks everyone for helping with this project!

 

[slcasner]

Not a lot of good news here. #33 now holds the record for the lowest recorded CAC in the study at 183.6.

What a lovely distinction!

 

[markwj]

Back in the day, there were reports of a 'bug' in the CAC calculation on the early roadsters leading to lower than expected values. I had heard Tesla recommending:

1. Select standard mode, and drive down to 60% SOC
2. Wait an hour
3. Select range mode, and charge up to 100%
4. Wait an hour
5. Drive down to 60%
6. Wait an hour

That supposedly 'resets' the calculation algorithm. It could be an old wives tale, but it was at least a Tesla Service old wife.

Anybody tried that, or want to?

 

[Rolf68]

What a lovely distinction!

See the bright side : I got a CA of ca. 108 right now with my original battery. Would gladly swap with you anytime

 

[supersnoop]

Back in the day, there were reports of a 'bug' in the CAC calculation on the early roadsters leading to lower than expected values. I had heard Tesla recommending:

1. Select standard mode, and drive down to 60% SOC
2. Wait an hour
3. Select range mode, and charge up to 100%
4. Wait an hour
5. Drive down to 60%
6. Wait an hour

That supposedly 'resets' the calculation algorithm. It could be an old wives tale, but it was at least a Tesla Service old wife.

Anybody tried that, or want to?

The current published method seems to be:
1. Range charge to full
2. Drive to below 10% in a single continuous trip (single turn of the key)
3. Charge back up

Nothing will immediately reset the CAC, but that will help the car recalculate.

 

[markwj]

The current published method seems to be

Do you know where it is published?

 

[supersnoop]

Do you know where it is published?

Anecdotally on these boards.