SOC data: 281-based, New-Bars

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.

garygid

Well-known member
Joined
Apr 21, 2010
Messages
12,469
Location
Laguna Hills, Orange Co, CA
Now that we have some working hardware for displaying the SOC (as a percentage, using 281 as 100%) as we drive, post and discuss the data readings here.

ONLY data for the LEAFs with the updated firmware, where the "fuel" indicator "bars" are affectionately called "New-Bars".

Comparisons to Old-Bars may be made if helpful to the new-Bars discussion, but clearly distinguish the Old-Bars data.

I think my Old-Bars SOC read 98.9 after charging to 100% overnight. It "should be" the same SOC for old-Bars and new-Bars.

Also, we "think" that the SOC for Low Battery warning, Very Low Battery Warning, and Turtle Mode should be the same for both the "original" and "updated" LEAF firmware.

It is currently presumed that only the SOC values that trigger the Bar-On and Bar-Off display changes are now different, with each Bar now representing a bit less capacity. So, now there is apparently a "hidden" reserve, after all the Bars disappear, that you can no longer see.

The SOC-Meter gives us some way of examining that "hidden" capacity. If we CANNOT use it, we have effectively "lost" that capacity.
 
For those that want to record the SOC when bars are gained or lost, we now have 3-buss CAN message recorder to capture all the data during charging, discharging, or driving.

It requires a PC with Windows to run our CAN-Do program and our 3-port CAN-Capture hardware.

This hardware costs about $250 plus tax and shipping, and requires adding 3 end connectors to one OBD cable. See the "AVR-CAN hardware for CAN-data Capture" thread for more details.
 
I have one of the early hardware/software setups from Gary's group so that we can do some testing on a car that has the Nissan firmware update, i.e. displays "new bars" on its dash.

I did some testing down near the bottom of the SOC range.

Bars..Percent SOC...miles showing...warnings
..2......29%.............19.................none
..1......22%.............13.................none
..1......17%.............8 (flashing)....Low battery
..0........8%.............3 (flashing)...zero bars showing
..0.....8% ~7%.........---..............Very Low battery (reading was 8% when VLB warning appeared, but reading changed to 7% almost immediately)

LEAF parked and turned off in garage, connected to charge at L2....5% SOC display.

Miles/kWh display on the LEAF main dashboard was 3.7 for this driving period.

Full recharge to 100% took 24.1 kWh measured by TED. (Note that TED showed 26.1 kWh for full charge after I took this car to "turtle" mode once. Perhaps we can use the 5% display and the 2kWh differential for charging to infer battery capacity?)

SOC display with 100% charge and 12 bars on dash was 100%, SOC "number" 281.
 
Am I correct in assuming that this SOC gauge will also allow one to detect the minute loss of battery pack capacity over time? How about eventually the SOC for each of the 48 modules (or possibly finer) constituting the battery pack, which might allow one to detect a defective module whose SOC is significantly less than the others?
 
New report on a freeway drive today, total distance 58.9 miles, ran the "tank" down to 2 bars. I ran at normal freeway speeds, sometimes up to 74 mph, got on and off the freeway for safe data recording at times, up and down some hills, A/C on the whole time, ambient temps around 74 F. I took SOC readings immediately as the bars dropped to the indicated bar level.

Main dash energy economy for this trip = 3.5 mi/kWh

Odo miles...Bars...SOC%...SOC #...Guessometer miles
....0...........12.....100%.....280........102
...3.3..........11.....92%......257........92
...9.3..........10.....84%......237........68
...14.4..........9.....78%......220........67
...23.0..........8.....71%......199........68
...27.2..........7.....64%......180........50
...31.7..........6.....57%......162........47
...39.0..........5.....51%......141........43
...43.0..........4.....44%......123........30
...49.1..........3.....36%......103........23
...57.3..........2.....29%.......83.........19
Car Parked, trip end
...58.9..........2.....26%.......74..........15

Edit: After several hours of sitting,

...58.9...........1....25%.......70..........14
 
Bars...SOC%...SOC #...Units burned
12.....100%.....280........N/A
11......92%......257........23
10......84%......237........20
..9......78%......220........17
..8......71%......199........21
..7......64%......180........19
..6......57%......162........18
..5......51%......141........21
..4......44%......123........18
..3......36%......103........20
..2......29%.......83.........20
Guess...
..1......20%.......55.........28
..0......10%.......28.........28
 
Boomer23 said:
Main dash energy economy for this trip = 3.5 mi/kWh

Odo miles...Bars...SOC%...SOC #...Guessometer miles
....0...........12.....100%.....280........102
...
...58.9..........2.....26%.......74..........15
Cool.

KWh used = 58.9/3.5 = 16.8 Kwh
KWh / SOC Unit = 16.8/(280-74) = 0.082
100% SOC = 22.87 KWh.
 
mwalsh said:
evnow said:
100% SOC = 22.87 KWh.

.....on the new firmware.
.....if you can trust the miles per KWh display to be accurate. Plus, even if it is, what if they round it? The real number could be anywhere from 3.45 to 3.54 giving capacity numbers ranging from 23.21KW to 22.62KW. So what exactly do we know? :D
 
Boomer23 said:
Main dash energy economy for this trip = 3.5 mi/kWh

Odo miles...Bars...SOC%...SOC #...Guessometer miles
....0...........12.....100%.....280........102
...3.3..........11.....92%......257........92
...9.3..........10.....84%......237........68
...14.4..........9.....78%......220........67
...
...58.9..........2.....26%.......74..........15
There is something wrong with the 12th bar.

12th to 11th bar : 3.3 miles for 23 SOC units or 1.8 m/kwh.
11th to 10th bar : 6.0 miles for 20 SOC units or 3.7 m/kwh.
10th to 9th bar : 5.1 miles for 17 SOC units or 3.7 m/kwh.
etc
etc

So, looks like that 280 number goes down rather fast in the first few miles.

Why ?
 
This should make it clear.

boomer23 can probably tell us whether the m/kwh of each segment (or the m/kwh cumulative) is what he actually experienced during the run.

LeafSOC-u.png


In anycase, the 1st bar as we have heard from so many people, goes off too fast - even though we expect the initial driving to be slower (near the home) - and if we go by the SOC-units shown here, it has 23 units compared to others which are between 21 & 17. So, the 1st bar sould have got higher miles, not lower. The first m/kwh is a good give-away. 1.8 m/kwh is just too low.

So, whats happening ? Here is my guess.

- 280 number at the top is a fake (i.e. when it crosses say 270, it is shown as 280). We can confirm this by looking at the charging soc-u that gary has a log of.
- The top 23 units are worth only about 10 to 15 i.e. at the top, the units are not linear.
 
Actually, the SOC # earlier in the morning was showing 281 at 100% charge. I logged the 280 number because that was the display when I actually started driving. So evnow's supposition about 280 being a default number for anything over 270 is incorrect. As I recall, though I was driving and couldn't monitor the SOC mini-display closely, the SOC number dropped smoothly from 280 downward.

Regarding my experience of the drive matching the variability per segment that your chart shows, I'd say yes. I did some rapid accelerating at times on the freeway, I got off the freeway sometimes and climbed some short hills and returned down the same hills, and I did some fairly efficient freeway and road cruising toward the end.

However, regarding the segment from 12 to 11 bars remaining, I don't think that I drove aggressively enough to only get 1.8 mi/kWh. That first segment is a mystery to me. I always thought that Nissan had left a little "head space" so that when we charge to 100%, we're really only charging to 98%, maybe it was to save the battery life by reducing charging heat damage at the end of a 100% charge. In fact, mwalsh reported once that he seemed to get a few more miles out of the top bar when he followed a 100% charge with another plug-in. But the SOC number 281 seems to be real and it seems to correspond to 100%, according to Gary, et al.
 
Boomer23 said:
But the SOC number 281 seems to be real and it seems to correspond to 100%, according to Gary, et al.
That means
- Leaf is very inefficient when we first start it (no regen !) or
- Top bar SOC units are not linear and have only half the KWh of rest of the SOC units.
 
evnow said:
- Leaf is very inefficient when we first start it (no regen !) or
- Top bar SOC units are not linear and have only half the KWh of rest of the SOC units.


I see no reason for the fuel bars to have linear values, and the SOC units seem to bare that out. We don't even know if the SOC units are linear !!!
Obviously, more data points will help.

We need to base the power consumption on raw data. Drive the vehicle on as level and steady state speed as possible, where we can guess how many miles per SOC units (as opposed to trying to guess what that is in kw, since we truly do not now the relationship to SOC units).

If we do this test, and have a somewhat linear consumption of SOC units per mile, we can at least guess that the SOC units are linear.

Or... that the battery cannot deliver power at the same efficiency based on SOC, as you suggest. Or, ???
 
If the SOC "estimate" is based on pack or cell(s) estimated "resting" voltage, that voltage does taper off more quickly at the top end of a charged battery's use, before it settles into a long decline.

Showing SOC vs. Time with a constant load (like High-Beam lights, or maybe a Full-On Heater) would most likely show a faster drop in "SOC" at the very start, then a long decline, and finally a faster fall near the end.

Not so strangely, that matches very well what we are seeing!

This would mean that these "SOC units" represent (or "hold") less energy at the beginning, AND less toward the end.
 
TonyWilliams said:
evnow said:
- Leaf is very inefficient when we first start it (no regen !) or
- Top bar SOC units are not linear and have only half the KWh of rest of the SOC units.


I see no reason for the fuel bars to have linear values, and the SOC units seem to bare that out. We don't even know if the SOC units are linear !!!
Obviously, more data points will help.

We need to base the power consumption on raw data. Drive the vehicle on as level and steady state speed as possible, where we can guess how many miles per SOC units (as opposed to trying to guess what that is in kw, since we truly do not now the relationship to SOC units).

If we do this test, and have a somewhat linear consumption of SOC units per mile, we can at least guess that the SOC units are linear.

Or... that the battery cannot deliver power at the same efficiency based on SOC, as you suggest. Or, ???

I'll throw my LEAF on a flatbed to the salt flats immediately. :lol:

Seriously, though, if we get some agreement on Tony's suggestion, we could try some relatively level and steady speeds. But I doubt that we could do it for very far or long, unless we put it on an oval track, like Edmunds did. There's nowhere I can think of around here that is level, flat and long enough to be useful, unless it's on the tarmac at the old El Toro base.
 
garygid said:
This would mean that these "SOC units" represent (or "hold") less energy at the beginning, AND less toward the end.

It sounds like the 281 number is just pack voltage, not kwh like we thought..perhaps more linear towards the end if Nissan never lets you go past the voltage "knee" on the low side.
 
Boomer23 said:
nowhere I can think of around here that is level, flat and long enough to be useful, unless it's on the tarmac at the old El Toro base.

The closest tracks I can think of are the Ford and Chrysler tracks in western Arizona:

Ford Track

Chrysler Track

The 5 freeway, north of the Grapevine, is relatively flat for a long way. There are dry lake beds in the upper desert, between Victorville and Palmdale.

BUT... the easy answer is a chassis dyno.
 
garygid said:
If the SOC "estimate" is based on pack or cell(s) estimated "resting" voltage, that voltage does taper off more quickly at the top end of a charged battery's use, before it settles into a long decline.
I'm sure Nissan's soc mapping tables take care of that. They would also be tracking current draw.
 
Back
Top