Discuss data from the LEAF Battery app, and Comparisons

[LEAFfan]

Incidentally, the Leaf (2013 SV) doesn't seem to connect while I'm charging. (It connects to the ELM but says "waiting for leaf") If I press a key fob button, or door lock, I get data for a few seconds, and it stops again. Has anyone else seen this?

Mine always connects while charging in the 'On' position (two button pushes)

So now, when I'm in my house (the bluetooth reaches my living room, dining room, and kitchen) and I want to know my charging status, I run the battery app, and press the charging door unlock button on my fob (which apparently "wakes up" something in the car). The app gets 2 or 3 updates and loses connection again. But it's faster than carwings.

 

[Staque]

What's the difference between the calculation of the KWh displayed and the Wh "trip meter"? (the one with the reset button) Is the Wh counter using a coulomb counter or something to get such precise measurement (down to the single watt-hour)?

I've noticed on my 2013 SV I start with 22.7 KWh (yes, 284 GIDs), and reset my Wh counter so it reads zero. After I drive for a while, it shows 22.0KWh and 700Wh (so far so good), then 20.8 and 1800 (uh-oh, 100Wh missing), eventually to 16.3KWh and 6000Wh, indicating a max capacity of 22.3KWh instead of 22.7KWh. I have to do more measurement to lower SOCs, but if I extrapolate what I have so far, I will have a discrepancy of 1.4KWh between the two meters, which makes me wonder if I'll be able to go those last 5 miles or not.

I've been using the watt-hour "trip meter" to gauge how much energy it takes to get to different places under different conditions, but if I can't compare those values to the KWh remaining, it's slightly less useful than I imagined it being. I can deal with a 6% error, but I'm wondering if there's any reason for that error existing, understanding that all calculations of a battery's capacity and SOC is a sophisticated form of guessing.

By the end of the day yesterday, the readings were 10.2KWh and 11500Wh spent. That's 1.0KWh off from my starting point.

 

[91040]

Have you considered the inefficiency of regeneration?

By the end of the day yesterday, the readings were 10.2KWh and 11500Wh spent. That's 1.0KWh off from my starting point.

 

[TonyWilliams]

Have you considered the inefficiency of regeneration?

By the end of the day yesterday, the readings were 10.2KWh and 11500Wh spent. That's 1.0KWh off from my starting point.

The whole concept is a bit skewed because the 22.7kWh STORED (284 * 80) is not what is available to the user EVER.

Of course the data won't match up. Which is why I asked Jim to change the DTE reading to be based on USABLE capacity with temperature adjustments. I believe that is in the latest version.

If you change that 284 Gids to 75 wattHours per unit, you'll get 21.3kWh useable, with about 0.3-0.5kWh not available at the bottom of the discharge. That leaves about 21-ish kWh useable, not the 22.7kWh that is STORED. Now when you driving down the highway at 65mph, busting out 4 miles per kWh on the dash gauge, your DTE should very, very closely match the actual miles (when the DTE is set on 4 miles per kWh).

This has been known for a LOooooong time, well before this app arrived.

 

[garygid]

Maybe one count is using 80 Wh per GID and
the other is using 75 Wh per GID?

Or, the Regen energy is not handled correctly?
Usually the energy used is more than the initial
recoverable energy in the battery, since
Regen "charges" the battery. However, the
net energy used is less than the initial plus regen
in, since the in-to-out process loses some energy in
going through the battery, like 10 in and 9 (or 9-1/2)
out.

However, if the two numbers are both based on
the same GIDs, they would have a constant sum.

 

[Staque]

Tony, Gary, I'm still not up to speed on why there is a difference between these two numbers. I'm only 3 months into the Leaf thing. I understand that the KWh number is directly calculated from GIDs, which is the car's estimate of remaining energy. I do not know where the resettable watt-hour counter comes from.

I know that most lithium BMS, such as I find on my cell phone, use coulomb counting for their percent-meters, because most lithium chemistries are nearly 100% efficient as electrons-in vs. electrons-out goes. Where you get the losses are the higher voltage during charging means more energy used to get the same number of electrons out which produces less energy when the pack voltage is sagging under load. Ok, cool. So if the watt-hour meter is simply an ammeter that is being multiplied by some nominal voltage, I can totally understand why it would count up more slowly than the GIDs drop. Or maybe I've got that backwards...

At any rate, I was just asking what device or technology the number comes from. Like if it is a meter that is super-precise (it must be if it can count single watt-hours), but not super-accurate, maybe it could use GIDs to compensate. So if it's ticking up 1, 2, 3, 15, 25, 65, 75 and then a GID drops, it can just re-adjust the counter to 80, and continue counting from there. That way, over the long run it will count up in the same way that GIDs count down, which is how I expect it to work. So if I normally use 5500Wh to drive home from work (according to the trip gauge), and I see 5.5KWh remaining, then I can be somewhat confident that I'll barely make it home. As it is, there is a 5% to 10% error that I need to account for in my head.

 

[Turbo3]

Resetable Wh comes from the high resolution SOC% value. It gives an approximate indication of energy into/out of battery. It is the only value available from the Leaf with the kind of resolution necessary for 1 Wh resolution counting.

 

[Staque]

Resetable Wh comes from the high resolution SOC% value. It gives an approximate indication of energy into/out of battery. It is the only value available from the Leaf with the kind of resolution necessary for 1 Wh resolution counting.

Do you know what about it makes it approximate? Given that it matches my dashboard mi/KWh almost exactly, it seems like potentially the most useful measure.

Does the Leaf give you the number as %SOC? How do you then convert that to watt-hours? (That is, what does the Leaf consider 100% to be? It seems to be always a moving target.)

 

[davewill]

Does the Leaf give you the number as %SOC? How do you then convert that to watt-hours? (That is, what does the Leaf consider 100% to be? It seems to be always a moving target.)

He uses the CAP Wh number at the top of the cell detail screen.

 

[TickTock]

... what does the Leaf consider 100% to be? It seems to be always a moving target.)

The LEAF considers 100% to be 65.6Ah. Always. Not sure how many volts to multiply with to get useable kWh. probably changes with time.
[Edit] Although, I was reading 100% SOH with 65.6 Ah, after gathering more data, 66.04 seems to work better. Only 2 out of 100 measurements are off by 1% (one of which is the 100% datapoint) using 66.04.[/Edit]

 

[FalconFour]

Okay, I've got a bit of a concern about health.

This thread's become a bit daunting, and I've dug back about 6 pages from the current to see if anything's been mentioned about this recently. I stopped when I saw a screenshot showing about 93% health.

Mine's 68% and dropping steadily each day. :shock: :?

It's a bit unnerving. I understand internal resistance, but don't understand what causes it and how to reverse damage. I've been changing my driving habits since I noticed this low health figure - and my efficiency has been pretty impressively improving as well. What information is health derived from? Is there something I can do to help fix this? I'm usually running between 6 and 7 temperature bars around here lately with 90-100 degree weather, but I've been laying off the lead-footing lately, and there are no fast chargers in my area to even consider that as an issue (I wish we had QCs...).

Here's how it looks right now as I'm writing this, sitting below me charging in the driveway (yay Bluetooth!):


Eek?

 

[TonyWilliams]

At any rate, I was just asking what device or technology the number comes from. Like if it is a meter that is super-precise (it must be if it can count single watt-hours), but not super-accurate, maybe it could use GIDs to compensate. So if it's ticking up 1, 2, 3, 15, 25, 65, 75 and then a GID drops, it can just re-adjust the counter to 80, and continue counting from there. That way, over the long run it will count up in the same way that GIDs count down, which is how I expect it to work. So if I normally use 5500Wh to drive home from work (according to the trip gauge), and I see 5.5KWh remaining, then I can be somewhat confident that I'll barely make it home. As it is, there is a 5% to 10% error that I need to account for in my head.

I can't comment on the wattHour readout, but the DTE is based on coulomb counting that is battery temperature adjusted in the app (and by the car), then multiplied by your input of economy (I think the app defaults at 4 miles/kWh). Coulomb counting will also get adjustments from the car based on Open Circuit Voltage.

The app should have a remaining energy at zero DTE of 0.48kWh from the coulomb counting. You can also adjust "zero DTE" on the app to give yourself a reserve, if you like.

I think I would just use the app's DTE and not try and account for stuff in your head !!!

 

[charge]

Just got the app and ODBII a few days ago...

I have a question about gids vs. AH capacity. First, some background. Have a 2012, Got it 1/30/12. About 22k miles. I'm in CT, so high temps are not much of an issue here. Never QC'd. I'm using the latest version of the app v.26.

From the battery app - AH=60.44(91.22%), Health = 90.35%. Yesterday I charged to 80% and had a 213 gid reading.

If I understand the readings from the app properly, I'm down almost 9% capacity. However, if I compare to gids, a new batter should have about 281 gids when charged to 100% and about 225 when charged to 80%. When I compare my 80% charge gid (213) to the expected gid for for a new battery charged to 80% (225), then per gids I'm down about 5%.

Am I understanding this correctly? If so, why the discrepancy between AH cap and gids? Also, which is more accurate?

Thanks in advance for your help! This forum is great.

 

[LEAFfan]

Just got the app and ODBII a few days ago...

I have a question about gids vs. AH capacity. First, some background. Have a 2012, Got it 1/30/12. About 22k miles. I'm in CT, so high temps are not much of an issue here. Never QC'd. I'm using the latest version of the app v.26.

From the battery app - AH=60.44(91.22%), Health = 90.35%. Yesterday I charged to 80% and had a 213 gid reading.

If I understand the readings from the app properly, I'm down almost 9% capacity. However, if I compare to gids, a new batter should have about 281 gids when charged to 100% and about 225 when charged to 80%. When I compare my 80% charge gid (213) to the expected gid for for a new battery charged to 80% (225), then per gids I'm down about 5%.

Am I understanding this correctly? If so, why the discrepancy between AH cap and gids? Also, which is more accurate?

Thanks in advance for your help! This forum is great.

Charge to 100% and that Gid reading will give you the best indicator for battery capacity.

 

[TickTock]

I wouldn't sweat this. "Health" is a misnomer - noone knows what this represents. This was suggested as a health candidate before the real health signal was known (msgId:5b3,2nd byte) but the name stuck. I suggest calling it Hfactor until we know what it is and unless you are logging this value to help try to figure out what it is, I would recommend ignoring it. With the LeafApp, you should just focus on the AHr for your health metric (AHr/65.6*100 to be precise so the LEAF in the screengrab has a health of 83.7%). Some have theorized that maybe Hfactor/hlth is related to series resistance but it is just a guess at this point.

Okay, I've got a bit of a concern about health.

This thread's become a bit daunting, and I've dug back about 6 pages from the current to see if anything's been mentioned about this recently. I stopped when I saw a screenshot showing about 93% health.

Mine's 68% and dropping steadily each day. :shock: :?

It's a bit unnerving. I understand internal resistance, but don't understand what causes it and how to reverse damage. I've been changing my driving habits since I noticed this low health figure - and my efficiency has been pretty impressively improving as well. What information is health derived from? Is there something I can do to help fix this? I'm usually running between 6 and 7 temperature bars around here lately with 90-100 degree weather, but I've been laying off the lead-footing lately, and there are no fast chargers in my area to even consider that as an issue (I wish we had QCs...).

Here's how it looks right now as I'm writing this, sitting below me charging in the driveway (yay Bluetooth!):


Eek?

 

[mwalsh]

I downloaded the "Health" beta yesterday (or was in Wednesday) and got it up and running this morning.

Readings are AHr=53.30 CAP=80.45% Hlth=76.57%

AHr and % readings since first getting the app up and going on June 22nd:

6/22: AHr=54.07 CAP=81.61%
6/24: AHr=54.01 CAP=81.52%
6/25: AHr=54.12 CAP=81.68% (weird that it went up!)
7/11: AHr=53.16 CAP=80.23%
7/12: AHr=53.30 CAP=80.45% Hlth=76.57% (went up again)

I'm not using the app and phone every day - finding it too difficult to read the screen with polarized sunglasses on, so sticking with the tried and true Gidometer for daily use.

Took another reading today. Only going for once a month (trying not to obsess over it):

8/26: AHr=52.54 CAP=79.30% Hlth=75.16%

 

[ttweed]

I wouldn't sweat this. "Health" is a misnomer - no one knows what this represents. This was suggested as a health candidate before the real health signal was known (msgId:5b3,2nd byte) but the name stuck.

I saw your post in the CANbus decoding thread about the "real" health signal looking like CAN msgId 5B32 = SOH% times 2. Is that definitely known now? If so, why hasn't it been incorporated in the LBA yet (or is it in one of the betas)? I haven't messed with any of the beta releases and am just using v0.26c, which only reports "Hfactor." It seems like it would be better to report that "real" number as "Health" than a value that represents an "unknown" parameter. Is Jim working on that?

TT

 

[mwalsh]

I downloaded the "Health" beta yesterday (or was in Wednesday) and got it up and running this morning.

Readings are AHr=53.30 CAP=80.45% Hlth=76.57%

AHr and % readings since first getting the app up and going on June 22nd:

6/22: AHr=54.07 CAP=81.61%
6/24: AHr=54.01 CAP=81.52%
6/25: AHr=54.12 CAP=81.68% (weird that it went up!)
7/11: AHr=53.16 CAP=80.23%
7/12: AHr=53.30 CAP=80.45% Hlth=76.57% (went up again)

I'm not using the app and phone every day - finding it too difficult to read the screen with polarized sunglasses on, so sticking with the tried and true Gidometer for daily use.

Took another reading today. Only going for once a month (trying not to obsess over it):

8/26: AHr=52.54 CAP=79.30% Hlth=75.16%

And another reading today, in light of a rather unexpected 14 Gid pop from yesterday (from 208 yesterday to 222 today):

8/27: AHr=52.73 CAP=79.58% Hlth=74.97%

AHr=up CAP=up Hlth=down

 

[TomT]

Yeah, I have the same concern. It has gotten to the point where I fast-forward over that page on my Lincomatic and ignore it on the Battery App...
I tried to get more info on the Health/XFactor/Whatever value from Craig at our Phoenix meeting but didn't glean anything that really shed any more light on it...

Okay, I've got a bit of a concern about health.
Mine's 68% and dropping steadily each day. :shock: :?

 

[DaveEV]

And another reading today, in light of a rather unexpected 14 Gid pop from yesterday (from 208 yesterday to 222 today)

Did you happen to QC yesterday? I have found small temporary bumps in Ah/GID readings after a QC - I don't think I've been the first to notice this.