SOC-Meter Improvement Suggestions

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.
1. Instead of a box make the SOC percentage replace the useless miles remaining guess on the dash.
 
DougMarker said:
1. Instead of a box make the SOC percentage replace the useless miles remaining guess on the dash.

I think that is much easier said that done. You might have to rewrite firmware inside some different dashboard controller.
Probably magnitudes more effort for such a thing.
 
Related other thread:
http://www.mynissanleaf.com/viewtopic.php?f=44&t=5113" onclick="window.open(this.href);return false;
 
garygid said:
SOC-Meter Improvement Suggestions Summary:

1. ...
Allow manual input of a "miles per XXX" figure; combine it with the current "SoC" figure to optionally display "remaining miles". It'd be up to the user to acquire her own "miles per XXX" figure: by guesswork, or by using historical data, by using the car's "miles per kWh" figure(s), etc.
 
Before that simple "calculator" (one multiply) function would be meaningful,
a MUCH BETTER indication of the remaining usable energy is needed.

You put in 4.3 mi/kWh, and the "guesSOC" value is 10% (28 raw).
Does that dependably mean that there is 2.1 kWh usable energy left?
NO! :( :( :(

So, doing the multiply of 2.1 x 4.3 = 9.03 ...
or even just mental math of 2 x 4 = 8-something ...
can be VERY misleading if only 1.2 usable kWh remains!!!
 
garygid said:
Before that simple "calculator" (one multiply) function would be meaningful,
a MUCH BETTER indication of the remaining usable energy is needed.

You put in 4.3 mi/kWh, and the "guesSOC" value is 10% (28 raw).
Does that dependably mean that there is 2.1 kWh usable energy left?
NO! :( :( :(

So, doing the multiply of 2.1 x 4.3 = 9.03 ...
or even just mental math of 2 x 4 = 8-something ...
can be VERY misleading if only 1.2 usable kWh remains!!!
You're quite right.

What I tried to get at (but failed, obviously) is this: IF the user has enough historical data to derive a reliable "average miles per 'tankful'" ("miles possible between 12 bars and turtle") average, and the "SoC number" of 281 - 0 does correspond to "12 bars - turtle", then by allowing the user to enter the "average miles per 'tankful'" (not "miles per kWh" as I originally suggested), we can display the remaining miles, which is very useful and is vastly more accurate than the Guess-O-Meter number.
 
No, it appears that the "SOC" number is not linear, or even dependable, especially at the low end.

So, such a "miles remaining" calculation would also be undependable, in my opinion.
 
garygid said:
No, it appears that the "SOC" number is not linear, or even dependable, especially at the low end.
But I keep seeing people posting the CAN bus SoC number along with a percentage. Does the percentage number also come off of the CAN bus or is it calculated as a percentage of 281 (in which case it's not very useful since the number is not linear)? (Sorry if this has been discussed already; I have not been following the "SoC meter" threads very closely.)
 
Yes, the calculated "SOC-Percentage" is just a convenience for folks that prefer it, somewhat like using the Bars, which we believe are also directly derived from the SOC-Raw value.

However, it is the apparent failure of the LEAF-generated SOC-Raw value to dependably represent the Battery's remaining Usable Energy that remains as a fundamental problem for those trying to use almost all of the LEAF's available "range" (Energy).

The amount of Energy available in the "Reserve" seems to be (just occasionally) significantly lower than "usual".

Technically, predicting the battery's low-end "capacity" is difficult, especially if one rarely ever "goes" there.
 
DougMarker said:
1. Instead of a box make the SOC percentage replace the useless miles remaining guess on the dash.
2. Integrate the kW used by the traction motor to report the total kWh used for a given trip. I can get an estimate by dividing the miles_travelled by the reported efficiency, but the efficiency also includes AC and other. I'd like to know the energy just going to (and from) the motor. Would need some mechanism to reset the total (like holding the black button for 3 seconds). Maybe keep two running accumulators - one for drive and the other for regen.
 
It appears that we probably have a value for power (or maybe amps):
1. going into the inverter (to drive the inverter and motor)
2. the net coming out of the inverter (net being generated by the inverter and motor). This net power is then used to power the car's HV bus and possibly charge the battery.

It is not clear what is being measured, or the scale factors. Perhaps the Service Manual contains information about the sensor, which might tell us what is actually being measured.

Although the "integration" of approximate power use to get energy used is easy in theory ... it is more difficult in practice with limited computational resources (extended precision and floating point instructions must be done as subroutines by a math library) and no real micro-second "clock". Of course, a timer and counter can be used to make an interval "clock", it just takes more instructions and interrupt handling.

What would you use the "Drive/Regen Energy" for?

Would it need to be written to flash before you Power-Off the SOC-Meter?

When would one zero the Energy value?

Even a Total Energy (in/out of Battery Pack) does not cover:
1. Self-Discharge of Cells
2. Equilization losses
3. Battery charging losses
4. Energy recovery (from battery) losses
5. Charger Losses

For now, I think that Battery Pack Temperature is more useful in managing one's use of the LEAF (when to charge).
 
So the application is just to help figure out where all the power is going. I thought of this because I consistantly use 1.3kWh more going home than going to work. The altitude drop from home to work is only 250ft so my math says the difference should be half this. One (good) theory is the AC is working harder on the return trip since it is much hotter on my way home. Hasn't been cool enough yet for me to endure shutting off the AC. But being able to log the energy use of just the traction would eliminate other variables in the data and get a better metric on the efficiency of the motor inverter, etc.

Another feature idea:
3. Indicate if, or better yet, how much friction braking is being applied. Not sure if there is an entry for this but given the car is drive by wire, I'd be surprised if there wasn't.
 
It appears that the SOC-Meter gets many messages containing Battery Voltage and Current. To integrate the Amp-Hours (or Power) over time, we need to know something more about the timing of these messages, to properly incorporate a "delta-T" into the calculation.

If the V&I messages are being sent at a constant rate, that would help. But, I have not yet looked at the timing.
 
4. Allow 100% calibration. When in % mode, press and hold the black button to establish the current setting as 100% (use n gids from now on instead of 281 where n is the current reading). This can be done after a 100% charge. Does the AVR have non-volatile (flash) memory available to the code (I know it does for code storage - but can the sw store to NV memory)? Otherwise I will probably have to keep the AVR powered (maybe I will change the switch to only shut off the display with the vehicle when that position is selected).
 
although i would want SOC displayed a half dozen ways; too much information leads to too much mis-interpretation which makes it easy to understand Nissan's thinking.

there are simply too many people in this world now who get into trouble with preconceived notions of the world in general, to wit; weekly reports of someone running out of gas on Priuschat because why?? 500 mile tank range is not enough? gauge not working? gas station closed?.

nope, nada, none of those reasons. now they all provide a "reason" as to why they cant make it to the gas station more often than once every 2 weeks and one day, but i find them to be more applicable to the "lame excuse" category.

after that comes the "blame Toyota for their snide ways of making the person look like an idiot" rant. its all rather predictable.

Nissan understands this and despite me wanting to know how much power is left in Kw, that will never happen. its like looking at gas gauge and thinking, well i know in my 1980 such and such i was able to go 20 miles below the "E" and to make sense of that, it was a car and so is this one, so i am ok...

so to suggest that Nissan give us this i think is a waste of time. i would definitely voice my wishes but doing so with little hope of getting what i want which is all the information. what we will probably get is the same info now, formatted differently which will just have us right back here in 2014 whining about the new gauges

so if you want more info, get it yourself. let me know, because i am interested
 
The uP on the AVR-CAN board in our SOC-Meter does have NV storage.

Although I do not yet know how to access/use it, the setting of "calibration" or "personalization" values should be possible.

However, before we "cover up" your strange "soc" value with a custom "100%-SOC" value, I would rather find out what makes your LEAF special.

I suspect that there is a value in the data that we should be using as the "Max-Soc" value. Otherwise, how (in your car) does the VCM compute the Fuel-Bars?

This answer might lead us to how the Capacity-Bars are computed, and that might lead us towards discovering a really useful "Useful-Capacity" value.

Being "special", your LEAF could help lead us there. Get some good extended EV-Logs of various charging sessions.

But, if I get time, I will try to get some NV-values working.
 
Back
Top