The balancing stops when the min and max voltages reach .02V difference, or the SOC reaches 78%. I took the time once to manually balance all the bricks to the same voltage and only gained 10 miles of range. I imagine if it was programmed to balance to a near zero difference, it would be balancing almost all the time. Personally I think the CAC is related to how the car is previously driven, like the estimated range.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
If Marius' car is at 199.000 km (AKA 124K miles) and if that distance is all on the same battery pack, then the battery is nearing the end of its useful charging cycle lifetime, no? IIRC, Tesla said 7 years or 100K miles.
I believe it was 70% after 5y or 50k miles?
I doubt what Marius is experiencing, is "normal aging". It looks more like his CAC has not been re-calculated for quite a while before what now appears to be a sudden drop..
My CAC is going up and down 3 or 4 times per week, I don't need to go below a certain SOC to trigger it.. The jumps are usually a few tenths only. Now at 153.90 and almost 40k kms.
Sudden drops... I guess I just had my first one..
Last Wednesday, I had to run the battery quite empty, and had to use Range Mode for about 10km after normal range went to zero. I arrived home with "Range uncertain" msg on the VDS. Immediately after plugging it, it appeared that the CAC dropped from 153.20 to 150.69! The day after it dropped further to 150.45. But since then it moved up a little again every day, and now I am again at 151.66. Seems like another week and I'll be back in the high 152s..
I didn't see much difference in ideal range predictions, it dropped from 293km to 291km but this morning it was back at 293.
Last Wednesday, I had to run the battery quite empty, and had to use Range Mode for about 10km after normal range went to zero. I arrived home with "Range uncertain" msg on the VDS. Immediately after plugging it, it appeared that the CAC dropped from 153.20 to 150.69! The day after it dropped further to 150.45. But since then it moved up a little again every day, and now I am again at 151.66. Seems like another week and I'll be back in the high 152s..
I didn't see much difference in ideal range predictions, it dropped from 293km to 291km but this morning it was back at 293.
dhrivnak
Active Member
Yes I have seen my CAC and ideal miles (they are strongly correlated) change over time often suddenly. It has dropped about 6 miles each winter to recover 4 in the spring. See the first graph below. Also notice ideal miles can easily jump 5 miles from one day to the next. I also notice jumps after along drive. This is likely due to giving the computer a chance to re-calibrate after longish periods of just short drives. As I normally drive under 40 miles or over 200 with few drives in-between.
What is mildly concerning is last summer I did a few range charges and after the first three I saw drops of about 2.8 miles after each range charge. Coincidence or with a 5+ year old battery has become fragile on range charges? Not enough data to say for sure especially since I am wary of doing more range charges. I have about 31,000 miles and I have to agree the battery is still within "normal" limits according the Roadster owners battery study. So here is hoping the new battery is affordable.
What is mildly concerning is last summer I did a few range charges and after the first three I saw drops of about 2.8 miles after each range charge. Coincidence or with a 5+ year old battery has become fragile on range charges? Not enough data to say for sure especially since I am wary of doing more range charges. I have about 31,000 miles and I have to agree the battery is still within "normal" limits according the Roadster owners battery study. So here is hoping the new battery is affordable.
spaceballs
Member
I've assume he's taken his ESS apart and/or BMB boards to manually do this (note voids ESS if it still under warranty).
As Spaceballs stated, you need to you need to access the inside of the pack. Not for the faint of heart or anyone with a warranty.
Mine's been jumping around a lot of late. I was around 151 when I got the car, up to 153 after my trip around Ireland last summer, dropped down to the high 148s immediately afterwards and into 149s over the winter. It was climbing again, but after a range mode charge this weekend it dropped to 146s, climbed again to 147s then back to 146.08 after another range mode charge to get home.
A drop of almost 4 points after one range charge seems a lot. I had thought it was balanced before that (I'd allowed it to balance the previous week and the balance grid was all 0s) so I think I might have to pull the logs.
A drop of almost 4 points after one range charge seems a lot. I had thought it was balanced before that (I'd allowed it to balance the previous week and the balance grid was all 0s) so I think I might have to pull the logs.
The CAC can vary a lot depending upon driving. The graph below shows the CAC of my 1.5 Roadster #33 over its 6.5-year life during which there have not been any drive-train changes (other than replacement of the line input fuses).
I've marked some events of interest. In July, 2010 there was a sharp drop on a day when I drove on the track at Laguna Seca on a hot day. However, the CAC was already falling fairly rapidly for the preceding month, perhaps as a result of a firmware update causing the calculation of CAC to change. I've also marked several increases since then that occurred each time we took the car on a long trip with several range mode charges. In particular, there was a rise from 141.5 to 148 when we went on a 12-day, 2342-mile road trip to Vancouver in April, 2014. After the CAC dropped down to 143 again in the intervening year as the car was not driven as much, in April and May we took another two trips and saw the CAC increase to 149. So, knock wood, the battery seems to be holding up pretty well as it approaches its specified 7-year lifetime.
n2mb_racing
Active Member
Interesting! It seems like the long road trips either actually improve the battery condition, or improve the cars estimate of the current battery condition. With normal commuting, it seems that the average trend is down until the battery is really exercised.
I assume that the long road trips don't actually improve the battery condition since every charging cycle causes a bit more deformation of battery structures, but taking the charge level down low lets the estimator know where the bottom of the range should be. I'm surprised, though, that the CAC estimate decreases as quickly as my car's history shows in the periods of only short commuting between trips.
With mine it seems it can go either way. I've had it jump up after a full range mode drive and I've had it jump down after one. Basically it's the algorithm re-calibrating and it could go either way.
Mark77a
Member
Mine seems to do this too - ie recalibrate up or down after a long run down to c 20% or less. But little recalibration happens in the usual upper, 60 to 90% range. I've seen CAC's of between 149 and 153 in the ESS's 3rd / 4th years of life.
I'd love to get a nice graph like slcasner ... is that from the Mac version of the log reader ? (I've been using the PC version), or are you a stats/software wizzard like many on here
The data for my graph came from Tom Saxton's VMSParser which is a command-line program that runs on both Mac OS X and Windows and generates text output. I wrote a Perl script to select and format the CAC values (identified as "brickahave" for average brick amp-hours) from the VMSParser output into the form needed for the graphing program that I use and maintain, named graf, which runs on Mac and Unix systems, and perhaps could also run on Windows within an X Window System environment. However, the output from VMSParser could also be imported to Excel as CSV data and graphed from there.
Perhaps one of the other log parser programs can display a CAC graph, too, but I am not familiar with them. I have my own log parser and one of these days I'll get around to having it directly produce the CAC graph data in the format I want.
Perhaps one of the other log parser programs can display a CAC graph, too, but I am not familiar with them. I have my own log parser and one of these days I'll get around to having it directly produce the CAC graph data in the format I want.
Mark77a
Member
Well, yes, but if you consider the overall trend line, there would be other minima below the line by a similar amount. Also, it is possible that some of the rise before that drop was not real, and therefore perhaps not all of the drop, either. There was not a similar drop when I participated in REFUEL again for the three subsequent years. Overall the loss of capacity over the 6.5 years is 10% or less, which is well within expectations.
Similar threads
