Capacity vs Hx (Health)

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

GregH

Well-known member
Joined
Jul 1, 2011
Messages
860
Location
Irvine, CA
I should put both of those in quotes.. by "capacity" I mean the Ah capacity we discovered in the Group 1 active CAN data on line 6.. 3rd, 4th and 5th bytes. Displayed on the LeafDD page 2 "C" and at the top of the ELM voltages page. By "health" I mean a 16 bit number I've been playing with... ("H" on the LeafDD page 2) Also in Group 1 active data but on line 5 on the 3rd and 4th bytes. Like the Ah capacity, this seems to track fairly well with actual usable capacity only this seems to be in the form of a percentage of original capacity.

This may end up being a more useful indicator of actual capacity (and possibly the number used to display the capacity bars.. which of course factors into the warranty). What I find interesting is that on the two 2013 cars I've tested so far the health percentage was over 110!! Could it be the bar of expectations has been lowered on the 2013s so they have to dip a little lower before the warranty kicks in? Will this number change if we update the software on the 2011/12 cars as Nissan intends us to do? Perhaps we could display this in the ELM software as well and get more readings. We'll have a dozen or so new LeafDD users by this time next week as well.

Here is some data from other cars (some 2013s) as well as multiple readings from my car..

Cap vs health
Other cars
2013A 67.3620 111.58%
2013B 67.3620 115.86%
Jim 62.0904 93.45%
Gary 61.9985 93.27%
Robert 61.2179 91.81%
Luke 49.3443 68.56%

My car
Early 67.5680 104.46%
Early 67.5680 104.73%
??? 66.0198 100.79%
??? 65.1211 99.11%
??? 65.1342 99.14%
??? 63.5532 96.18%
??? 63.5860 96.24%
6/02 63.0022 95.15%
6/11 62.8316 94.78%
 
GregH said:
What I find interesting is that on the two 2013 cars I've tested so far the health percentage was over 110!! Could it be the bar of expectations has been lowered on the 2013s so they have to dip a little lower before the warranty kicks in? Will this number change if we update the software on the 2011/12 cars as Nissan intends us to do?
I have to admit that thought has crossed my mind and it scares the hell out of me...

GregH said:
Perhaps we could display this in the ELM software as well and get more readings. We'll have a dozen or so new LeafDD users by this time next week as well.
Here's hoping I'm one of those LeafDD users!
 
GregH said:
I should put both of those in quotes.. by "capacity" I mean the Ah capacity we discovered in the Group 1 active CAN data on line 6.. 3rd, 4th and 5th bytes. Displayed on the LeafDD page 2 "C" and at the top of the ELM voltages page. By "health" I mean a 16 bit number I've been playing with... ("H" on the LeafDD page 2) Also in Group 1 active data but on line 5 on the 3rd and 4th bytes. Like the Ah capacity, this seems to track fairly well with actual usable capacity only this seems to be in the form of a percentage of original capacity.

This may end up being a more useful indicator of actual capacity (and possibly the number used to display the capacity bars.. which of course factors into the warranty). What I find interesting is that on the two 2013 cars I've tested so far the health percentage was over 110!! Could it be the bar of expectations has been lowered on the 2013s so they have to dip a little lower before the warranty kicks in? Will this number change if we update the software on the 2011/12 cars as Nissan intends us to do?
Very interesting information, but remember that even Gids under-predicted range in the Arizona range test. For cars with a Gid Percentage under 90% on full charge, the average difference between range predicted by Gids and actual range was 11.9%. For example, based on the linear regression line a car with 75% Gids on full charge got 85% of the range of a new Leaf.

http://www.mynissanleaf.com/wiki/index.php?title=Battery_Capacity_Loss#Range_Test_on_Cars_with_Battery_Capacity_Loss" onclick="window.open(this.href);return false;

My wild guess is that the new formula put a new Leaf at 110% plus so that:

1) They could retain backward compatibility with the rest of the software AND
2) When the capacity starts degrading the gauge will more accurately show the remaining capacity.
 
Here is some data from my 2012 leaf with approx 12K miles. I took this data with a wifi clone elm device talking to my iphone. Pair delta is difference between highest cellpair and lowest. max t is the highest of 4 temps.

I've been surprised at the variation of the the CAP value.

Date, SOC, CAP, HEALTH, KWnom, pair delta, maxt
4/27/13, 76.9365, 65.7312, 100.19, 18.2057, 30, 13
5/2/13, 60.9825, 66.3872, 101.57, 14.5744, 20, 21
5/3/13, 74.2031, 66.5184, 101.78, 17.7691, 22, 16
5/6/13, 73.8305, 66.4790, 101.67, 17.6694, 27, 14
5/7/13, 78.3671, 66.3216, 101.36, 18.7108, 23, 18
5/9/13, 78.9369, 66.2363, 101.19, 18.8226, 21, 18
5/9/13, 71.1871, 66.4134, 101.52, 17.0200, 21, 18
5/10/13, 81.8851, 66.8004, 102.35, 19.6918, 26, 20
5/13/1, 78.2961, 66.4396, 101.57, 18.7271, 27, 17
5/14/13, 70.6824, 66.4790, 101.78, 16.9160, 17, 15
5/16/13, 92.7587, 66.8595, 102.35, 22.3265, 21, 16
5/19/13, 76.9518, 66.7873, 102.34, 18.5019, 31, 19
5/20/13, 90.3895, 67.0563, 102.72, 21.8203, 21, 17
5/20/13, 65.5228, 67.0563, 102.89, 15.8174, 21, 22
5/20/13, 64.4883, 67.1481, 102.89, 15.5890, 21, 24
5/20/13, 39.8447, 67.0366, 101.77, 9.6158, 20, 27
5/22/13, 72.0988, 66.3084, 101.35, 17.2107, 21, 17
5/23/13, 68.8397, 66.2166, 101.07, 16.4100, 26, 23
5/23/13, 61.4825, 66.1379, 100.99, 14.6388, 20, 23
5/24/13, 78.4341, 66.2428, 101.18, 18.7045, 31, 21
5/27/13, 77.4607, 66.2428, 101.07, 18.4724, 27, 16
5/29/13, 78.2033, 66.2953, 101.33, 18.6642, 28, 20
5/30/13, 80.2422, 66.4331, 101.57, 19.1907, 27, 22
5/30/13, 79.3837, 66.4396, 101.57, 18.9872, 23, 25
5/31/13, 79.3872, 65.7377, 100.26, 18.7874, 22, 22
5/31/13, 78.1753, 65.5868, 99.87, 18.4582, 31, 26
6/3/13, 76.2771, 64.7865, 98.54, 17.7902, 32, 22
6/3/13, 72.4685, 64.9243, 98.82, 16.9379, 25, 23
6/4/13, 90.5800, 64.9636, 98.82, 21.1839, 21, 21
6/4/13, 84.2846, 65.0752, 99.01, 19.7454, 21, 19
6/5/13, 76.2326, 65.1276, 99.19, 17.8734, 30, 18
6/5/13, 56.1890, 65.5672, 100.03, 13.2630, 24, 22
6/5/13, 47.6342, 65.6000, 100.04, 11.2493, 20, 23
6/6/13, 70.5646, 65.5540, 100.05, 16.6529, 26, 21
6/6/13, 79.0188, 65.7574, 100.30, 18.7059, 28, 23
6/6/13, 77.9348, 65.8033, 100.45, 18.4621, 35, 23
6/7/13, 76.9895, 65.8296, 100.67, 18.2455, 27, 20
6/7/13, 66.7311, 65.9542, 100.77, 15.8443, 20, 20
6/8/13, 63.8094, 66.4856, 101.66, 15.2727, 21, 18
 
adspguy said:
Here is some data from my 2012 leaf with approx 12K miles. I took this data with a wifi clone elm device talking to my iphone. Pair delta is difference between highest cellpair and lowest. max t is the highest of 4 temps.

I've been surprised at the variation of the the CAP value.

Date, SOC, CAP, HEALTH, KWnom, pair delta, maxt
4/27/13, 76.9365, 65.7312, 100.19, 18.2057, 30, 13
5/2/13, 60.9825, 66.3872, 101.57, 14.5744, 20, 21
5/3/13, 74.2031, 66.5184, 101.78, 17.7691, 22, 16
5/6/13, 73.8305, 66.4790, 101.67, 17.6694, 27, 14
5/7/13, 78.3671, 66.3216, 101.36, 18.7108, 23, 18
5/9/13, 78.9369, 66.2363, 101.19, 18.8226, 21, 18
5/9/13, 71.1871, 66.4134, 101.52, 17.0200, 21, 18
5/10/13, 81.8851, 66.8004, 102.35, 19.6918, 26, 20
5/13/1, 78.2961, 66.4396, 101.57, 18.7271, 27, 17
5/14/13, 70.6824, 66.4790, 101.78, 16.9160, 17, 15
5/16/13, 92.7587, 66.8595, 102.35, 22.3265, 21, 16
5/19/13, 76.9518, 66.7873, 102.34, 18.5019, 31, 19
5/20/13, 90.3895, 67.0563, 102.72, 21.8203, 21, 17
5/20/13, 65.5228, 67.0563, 102.89, 15.8174, 21, 22
5/20/13, 64.4883, 67.1481, 102.89, 15.5890, 21, 24
5/20/13, 39.8447, 67.0366, 101.77, 9.6158, 20, 27
5/22/13, 72.0988, 66.3084, 101.35, 17.2107, 21, 17
5/23/13, 68.8397, 66.2166, 101.07, 16.4100, 26, 23
5/23/13, 61.4825, 66.1379, 100.99, 14.6388, 20, 23
5/24/13, 78.4341, 66.2428, 101.18, 18.7045, 31, 21
5/27/13, 77.4607, 66.2428, 101.07, 18.4724, 27, 16
5/29/13, 78.2033, 66.2953, 101.33, 18.6642, 28, 20
5/30/13, 80.2422, 66.4331, 101.57, 19.1907, 27, 22
5/30/13, 79.3837, 66.4396, 101.57, 18.9872, 23, 25
5/31/13, 79.3872, 65.7377, 100.26, 18.7874, 22, 22
5/31/13, 78.1753, 65.5868, 99.87, 18.4582, 31, 26
6/3/13, 76.2771, 64.7865, 98.54, 17.7902, 32, 22
6/3/13, 72.4685, 64.9243, 98.82, 16.9379, 25, 23
6/4/13, 90.5800, 64.9636, 98.82, 21.1839, 21, 21
6/4/13, 84.2846, 65.0752, 99.01, 19.7454, 21, 19
6/5/13, 76.2326, 65.1276, 99.19, 17.8734, 30, 18
6/5/13, 56.1890, 65.5672, 100.03, 13.2630, 24, 22
6/5/13, 47.6342, 65.6000, 100.04, 11.2493, 20, 23
6/6/13, 70.5646, 65.5540, 100.05, 16.6529, 26, 21
6/6/13, 79.0188, 65.7574, 100.30, 18.7059, 28, 23
6/6/13, 77.9348, 65.8033, 100.45, 18.4621, 35, 23
6/7/13, 76.9895, 65.8296, 100.67, 18.2455, 27, 20
6/7/13, 66.7311, 65.9542, 100.77, 15.8443, 20, 20
6/8/13, 63.8094, 66.4856, 101.66, 15.2727, 21, 18

Great data set thanks! When/how often do you see the Ah CAP and Health numbers get updated? In this data it looks like there are times when CAP changes but Health does not, and other times when Health does but CAP does not.. Mostly it looks like they're both changing a lot more frequently then I would have expected...
 
GregH said:
Great data set thanks! When/how often do you see the Ah CAP and Health numbers get updated? In this data it looks like there are times when CAP changes but Health does not, and other times when Health does but CAP does not.. Mostly it looks like they're both changing a lot more frequently then I would have expected...
I get the data when I feel inspired and have a moment. Often this relates to when I am stuck waiting in the car for my kids when I pick them up. It is a bit of a manual effort to get the data. I've been keeping the raw data from the elm device for all the 210x groups, and using a c program I wrote to generate a report from raw data. I was most curious about if there was a relationship between max cell pair delta and CAP etc. When I saw your "health" post last night I added it and re-ran the program. I have more data that hasn't been entered since I only download it once in a while.

It is great that you have been able to figure out the meanings of the various data fields, I have all the raw data for when you figure out more or to maybe help. The cap doesn't seem to directly track either the max temp or max cell pair delta. I have found that after 100% charging it does seem to improve the cap some short time after and also the max cell pair delta.

I'd be happy to share more data with you, let me know.

My end plan is to try to make an iphone app like the android leafbattery app.
 
here is another group

Date, SOC, CAP, HEALTH, pair delta, maxt
6/8/13, 63.8094, 66.4856, 101.66, 21, 18
6/10/13, 76.1225, 66.5971, 101.9, 28, 17
6/10/13, 64.6072, 66.5971, 101.88, 18, 20
6/10/13, 84.6918, 66.866, 102.37, 17, 19
6/11/13, 75.3436, 67.0169, 102.6, 25, 19
6/11/13, 23.9343, 67.3384, 103.4, 22, 23
6/12/13, 78.7696, 67.4171, 103.36, 28, 20

this one with new record high CAP and Health numbers
 
adspguy said:
here is another group

Date, SOC, CAP, HEALTH, pair delta, maxt
6/8/13, 63.8094, 66.4856, 101.66, 21, 18
6/10/13, 76.1225, 66.5971, 101.9, 28, 17
6/10/13, 64.6072, 66.5971, 101.88, 18, 20
6/10/13, 84.6918, 66.866, 102.37, 17, 19
6/11/13, 75.3436, 67.0169, 102.6, 25, 19
6/11/13, 23.9343, 67.3384, 103.4, 22, 23
6/12/13, 78.7696, 67.4171, 103.36, 28, 20

this one with new record high CAP and Health numbers

Wow.. Nice to see the numbers going UP for a change! What's the occasion? Any idea why it dipped a few weeks ago then surges in the last few days? Ambient temps? Deeper or shallower discharges? Partial or full charges? Interesting. An iPhone app would be cool.. I'd try it. Although I already have the Dash Display ;)
 
Cap doesn't seem to follow temp or pair delta directly. The right axis is both temp in C and pair delta mV
ecxk6.jpg


Cap vs health:
dbo6g.jpg


These are plots of the previously posted data values

I usually charge to 80% but charged to 100% on index values 12,14,30 and 44
 
So, other than being scaled differently, the CAP and Health
seem to be basically the same thing, right?

What is the scale factor?
Is it a constant?
Perhaps the scale factor changes with... age?
 
garygid said:
So, other than being scaled differently, the CAP and Health
seem to be basically the same thing, right?

What is the scale factor?
Is it a constant?
Perhaps the scale factor changes with... age?
Great graph btw.. Gary the Ah CAP and Health seem to be very similar but it's hard to say which (if either) the car is using to determine capacity loss. need. more. data.
 
klapauzius said:
The scale factor is 1.8 and a linear fit yields:

health (%)=1.8 *cap -19

Yes! and that -19 offset is interesting.. Looks like you need to knock about 11Ah off the Ah capacity in order for it scale linearly with the health. Notice on the graph the health highs and lows are higher and lower than the Ah capacity..
adspguy said:
Cap vs health:
dbo6g.jpg

Perhaps health is a percentage of original USEABLE capacity? Since we can't really touch the top or bottom few percent SOC... Perhaps this is a better mark of capacity loss as it might better reflect loss in range?
 
These images aren't coming through (or they've been removed), but I'm following this thread with great interest. I look at that second page of LeafDD periodically.
 
I tried to fit a generalized linear model to capacity, using the data posted here. So basically

cap in ah = a* SOC+b*max temp + c

But it does not fit well... There is a very tentative inverse relationship of cap vs. temperature, but I suspect it is not linear....Maybe some exponential model could work, i.e.

ah=A*exp(-b*max_temp)
but the data look quite noisy, so this fit does not produce great results either...
 
And the cap/health roller coaster continues......
z4p7q.png

and
u4g1d.png


I've recently changed to 100% charging daily due to a longer commute during the week. Also, some days I take multiple data points, like when I arrive at work, and then just before I leave 9 hours later.
 
I think we need to wait to get more data...
At the moment it appears that there is just a tentative connection....Am I hallucinating or is the drop in cap lagging behind the temperature increases?
 
It seems like before the update the health was about the same as the SOC %. Now it is much lower.
This is the readings after 100%.

 
What are the numbers in the first photo (other than Vpack)? If the 2nd photo is Ah Capacity and Health that seems odd.. 85.67% Health sounds really low for a 61.35Ah pack. Just the other day I was at 93.57% Health with 62.1166Ah Capacity in my car.
 
Back
Top