I want my 281!

[LEAFfan]

Why not try the meter on another LEAF to see if its 80 or 100% charge reads low as well?

Mine never reaches 80% SOC at 80% charging. But that seems right because I lose the tenth bar really fast most of the time. Only once did it go 5 miles and I have no idea how it did that since it was less than 80% SOC. But after I lose the eighth bar, the SOC shows about 5% higher than CarWings.

 

[ebill3]

My last four L1 80% charges were all spot on 82.2% and 231 gids (or whatever).
Voltage was either 388.0 or 388.5.

Bill

 

[garygid]

I take it back, TickTock,
... but your Battery Voltages seem ABOUT right, but low. The approximate 2-volt difference might come from an out-of-balance condition, where one cell-pair is 0.02 volts (or so) higher than the rest. So, since charging is (most likely) stopped by the highest cell, the rest are 0.02 (or 0.04) volts low, and (96 - 1) x 0.02 (or 0.04) is about 2 (or 4) volts low for the pack.

Maybe, the way you charge, there is never time to do a/enough cell-balance, so out-of-balance might be at least part of the problem.

At the dealer, in addition to (or as part of) the annual battery check, they should be able to see/read/print ALL 96 of the cell-pair voltages.

Also, they might be able to force some "equalization" cycles, if needed.

Get it checked, and keep us posted ... please.

 

[Stoaty]

Just measured my battery at 129 gids with my new SOC meter (shows 5 bars, so about right). Thanks, Gary!

 

[mwalsh]

So much for lower numbers as battery balance gradually degrades - today's number was 91.8% / 258 Gids. So it's really got to be down to SOC at the beginning charging, and the trigger point has to be somewhere between 30 and 40%.

 

[vegastar]

I think the BMS may not be abble to know the exact charge state of the battery. I track the energy at the wall needed to charge the car from a knowm initial state (like when the 3rd bar disappears) and it varies a lot (more than 1kW.h). Only in the lower states of charge (LBW and VLBW) the wall comsuption is more consistent (less than 0.5kWh spread). I also verified that sometimes in the lower part of the charge state for the same driving conditions I get more km than the estimated by the guessometer in a 2 to 1 ratio. Yesterday I made 45 km (28 miles) between 2 bars and VLBW for the same driving conditions of the other 106 km (66m) in 10 bars.

 

[mwalsh]

This morning I got........100% / 281 Gids! YAAY!

So my remaining SOC theory officially goes out of the window now, because I got home last night with only 27% remaining.

However, since I was at a very low 91.8 the night before last (one of the lowest numbers I've seen too), I reckon the BMS was thinking it needed to do a balancing. But the question is why, after just a couple of days.....I typically see a week of gradual decline before a balancing event.

 

[surfingslovak]

This morning I got........100% / 281 Gids! YAAY!

Congrats! Are you guys recording this by any chance? It would be great to have a large data pool from several vehicles that would show the level of charge (80 o 100%), the number of gids prior and after the charge and ambient temp. This would be especially useful if you had a regular charge pattern.

 

[mwalsh]

This morning I got........100% / 281 Gids! YAAY!

Congrats! Are you guys recording this by any chance? It would be great to have a large data pool from several vehicles that would show the level of charge (80 o 100%), the number of gids prior and after the charge and ambient temp. This would be especially useful if you had a regular charge pattern.

No, I'm really not. I do start out collecting data obsessively, and that works well for a while. Then it just sorta goes off the rails because I have so much else going on.

 

[Boomer23]

This morning I got........100% / 281 Gids! YAAY!

Congrats! Are you guys recording this by any chance? It would be great to have a large data pool from several vehicles that would show the level of charge (80 o 100%), the number of gids prior and after the charge and ambient temp. This would be especially useful if you had a regular charge pattern.

No, I'm really not. I do start out collecting data obsessively, and that works well for a while. Then it just sorta goes off the rails because I have so much else going on.

I got the same readings this morning as well!

I keep a paper calendar in the car and I record several data points every night before I set the car to charge, and the first time I drive the car in a day.
I started recording my SOC and gids readings a couple weeks ago because of this thread. I'll have to check out google docs and set up a reporting spreadsheet.

 

[GregH]

Well, cooler temps didn't do anything for me. Wake up yesterday at 80% was 5 bars on temperature so after a short drive, let it sit for the day and night outside in the rain and cool temps (around 60% SOC) then plugged in this morning for a full charge.. 270#

 

[garygid]

TickTock,
Thanks for the Log file, of charging to 80%.

You expected your car to charge to "80%"
(which is normally 82% of 281 on our "SOC").

This is a 3 hour and 30-minute Log,
goes to about 388 volts and shows 10 of 12 bars.
These would be correct for an 80% charge.
So, from those, it appears that your LEAF is working OK

HOWEVER, it reads only 198 gids, which would be 80% of 250,
but is only 70.5% of our assumed 281 SOC.

So, there is something going on that we do not understand.

Maybe there is another "full-SOC" value that is near 281 many times,
but down around 250 in your car (for some reason).

Maybe there is new firmware in YOUR car that has re-scaled
the value that we are using for SOC to about 250 max???

Or, maybe something totally ... unknown.

So, something is indeed strange about "our" SOC value?
I wonder if others with later SN LEAFs have this "strangeness"?

If you can log a charge to 100%, I would like to see it.

Also, please try your SOC-Meter on another LEAF.
Cheers, Gary

 

[GregH]

We need to perform some tests to verify if one person's "gid" is the same as someone else's.. I believe the raw # is a remaining energy counter.. Perhaps not amp-hours (which is most common in other EVs) but at the least I would expect it to be an estimate by the car regarding remaining usable energy and I would hope it was consistent from car to car.

With that in mind, I think it's wrong for us to use gid/2.81 as a representation of our percentage SOC. Many of us may NEVER see 281# and for us 100% may be 270# or less. This of course complicates the task of making a universal tool.. But not too much..

The EV1 kept track of SOC relative to a separate pack capacity number.. Nominal was 48Ah, but for whatever reason, a person may only have a 42Ah or 43Ah capable car. Charge completion was of course based on cell (or in that case module) voltages, but the total Ah capacity would change based on a number of factors (temperature, age, cycles etc) and 100% was always 100% be it 39Ah or 48Ah.

I think the most valuable thing for us to have on Gary's SOC meter now would be a true Ah integrator.. So we could see if these gids are at all consistent between vehicles.

 

[garygid]

Assuming that one had an "approximate" Ah "counter",
when would you "zero" (or "set") the counter, and what would 100% (or "full") be set to?

I prefer to have a measure of available energy, which would be lower on EVs with a less-capable battery pack.

100% of a variable number is not very helpful: like knowing that the gas tank is full, but of variable (and unknown) size (capacity).

 

[TonyWilliams]

Assuming that one had an "approximate" Ah "counter",
when would you "zero" (or "set") the counter, and what would 100% (or "full") be set to?

I prefer to have a measure of available energy, which would be lower on EVs with a less-capable battery pack.

100% of a variable number is not very helpful: like knowing that the gas tank is full, but of variable (and unknown) size (capacity).

Which is exactly what the 12 fuel bars will do when the battery has 87% capacity... they'll show 12 bars, and even smile for you if you like.

 

[garygid]

Yep, useless as *its on a *oar *og. :lol:

 

[TickTock]

We should be able to compute this from the logs of a charge since we monitor voltage, charging current and gids. We just need to see if we get a better fit when plotted against V*I*t or against just I*t to determine if it is energy or charge (or neither). Based on my last data, I get about 70 Wh/gid (239 gids=100% for me) and this isn't too far off from Stoaty's computation of 65wh (281 gids=100%) suggesting that a gid may be constant car to car even if total capacity isn't.

 

[garygid]

TickTock,
Apparently LEAFfan, also in AZ, gets "low-SOC" values after charging to 80% (though not as low as your SOC-readings).

Maybe it's the geomagnetic psycho-resonance there in AZ? (joking)

It sure is strange to me.
If you log another charge from low to 100%, please send me the EV-Log.
Thanks.

 

[lincomatic]

Car sitting outside all night.. 55F this morning... 100% charge 277# 392.5V from 2 bars last night. I usually get 393.5V. This car has never seen anything higher that 277.

 

[garygid]

Under "normal" circumstances (and we are not sure what those are),
a charge:

1. to 80% reads APPROXIMATELY 231 "gids" (82.2% of 281).
Occasionally a few "gids" more or less is observed, and still considered "normal".
One person reports 70%, another about 75%, but both are unexplained.

2. to 100% reads about 278 "gids" (98.9% of 281).
Occasionally a number around 275 to 281 is observed, and still considered "normal".

No, these variations are not understood, but temperature and the
charge-from point have been considered.

Battery Pack equalization might be an important factor.