me68
Posts: 94
Joined: Fri May 24, 2013 5:53 am
Contact: Website

Re: Canion Battery Readings

Sat Jul 12, 2014 9:59 pm

Barbagris wrote:I have made a preliminary test on about 10% of the battery, and gives me a capacity of 14.5 kWh, with 26,000 km. But I still dislike the method.

What I commented on the strong voltage variations, which can make false measurement in wh, and (for me) be preferred in ah. If voltage is not estable, measures in wh are false, or inadecuate for our purpose, I think:

Image


Hello!

That's not true. caniOn calcs Watts = Volts * Amps with current values at this moment via looking for PID 373, which is sent with 100Hz.

Code: Select all

            Case("373")         ' 100 / sec
               B_Amp = (BB(2) * 256 + BB(3) - Bigv )/100
               B_Volt = ((BB(4) * 256 + BB(5))/10)
               B_W = B_Amp * B_Volt
               If (B_W > 0) Then
                  S_vasec_pos = S_vasec_pos + B_W
                  S_vakm_pos = S_vakm_pos + B_W
                  S_vaTrip_pos = S_vaTrip_pos + B_W
               Else
                  S_vasec_neg = S_vasec_neg + B_W
                  S_vakm_neg = S_vakm_neg + B_W
                  S_vaTrip_neg = S_vaTrip_neg + B_W
               End If


There are three summary values for three different intervals: seconds, km & trip.

In case of seconds there is a timer with 1 second intervall, where the collected data are written to table SECLOG and the second-values are resetted afterwards.

Code: Select all

Sub scanner_t1Sec_tick
   whkm = (S_vasec_pos + S_vasec_neg) / (360000 * S_dist / 1000)
   WriteTable_SECLOG
   S_vasec_neg = 0
   S_vasec_pos = 0
   S_dist = 0
   FpS = 0
End Sub


So Wh out are true values. I see imprecision only with Wh reg. This amount of energy will be loaded to the battery but with degree of efficiency. So on calculating Wh = Wh out - Wh reg, there is an imprecision. Measurement of Wh out only is right.

Martin
... en route with a Citroen C-Zero, Bj. 8/2012 (http://www.spritmonitor.de/en/detail/532466.html)
... our website: http://www.facebook.com/emobility.at

BlueLightning
Posts: 53
Joined: Sun Oct 14, 2012 11:59 am
Location: Normandy France
Contact: Website

Re: Canion Battery Readings

Thu Jul 31, 2014 3:07 pm

Battery Capacity Checking with Car Heater, Canion "Wh out" and SoC

Martin wrote: So Wh out are true values. I see imprecision only with Wh reg. This amount of energy will be loaded to the battery but with degree of efficiency. So on calculating Wh = Wh out - Wh reg, there is an imprecision. Measurement of Wh out only is right.

Yes Martin, you are right "Measurement of Wh out only is right" and thanks to you I got an easy way to measure my battery capacity.
As with Wh reg there is an imprecision, I parked the car put the heater on and got a record of kWh out with CaniOn.
Then Thank to Seclog, the sum at each line of B_VOLT x B_AMP x 1 / 1000 / 3600 gave me the kWh out.
The curve's slope is 0.1624 kWh per % SoC, or 16.24 kWh for 100 % SoC wich is the Battery capacity
SoC resolution is of 0.5 %, that's why this curve is showing steps.
If you are more at ease to use the Wh out given by CaniOn on the right side of the screen to draw this curve,
be sure you pick up the value only when there is a change of SoC otherwise the curve slope will be wrong.

Image
Capacity Battery measurement

Image
iOn Range
Last edited by BlueLightning on Sun Aug 03, 2014 11:27 am, edited 4 times in total.

Barbagris
Posts: 237
Joined: Wed Feb 20, 2013 8:34 am
Location: Bilbao, Spain

Re: Canion Battery Readings

Thu Jul 31, 2014 11:48 pm

Martin, I know your app performs calculations perfectly; what I mean is that I believe are not sufficiently valid (for me).

Sorry I can not explain better, I have limitations with the language.

Take an example. My car yesterday:

Image

We see that when I push the pedal, voltage drops, and when regenerate, climb. Normal.

If I push the pedal and extract 1 Ah of the battery, as the voltage drops to -let's say- 340 volts. Regenerate the same 1 Ah, and the measured voltage up to 360 volts. The "Trip Timer" screen indicate "Wh Out: 340 watts; Wh reg: 360 watts." We returned the same as we had taken (1 Ah) and we have more energy? That can not be true.

I think the only valid data for this task is Ah extracted or regenerated. Or, if you want to work with watts, OK, we can use a voltage "standard" (ie, 88 cells, 88 * 3.7 = 326 volts) and not the volts at each moment informs the bus (yes, that is valid and must be informed in the left)

Image

So to try to calculate the degradation of my battery, I do not trust the info at "trip timer" about watts*hour in one or other direction, and I'm doing calculations with amps recorded in the trip log.

I hope I explained better.
...learning...

priusfan
Posts: 194
Joined: Mon Oct 08, 2012 7:39 am
Location: France

Re: Canion Battery Readings

Fri Aug 01, 2014 7:30 am

@BarbeGrise:

I do not agree with your calcs....

You should not think AH but Amps and Volts and then calculate Watts and, after summing, WH.

The way Martin explained it is the right way to calculate WH....

You should compare the graphs for Volts & Amps (for the same period) and for each point calculate the Watts and proceed to the integration...

This is what is done with accuracy (100 samples / sec) in the formulas above.

Xavier

BlueLightning
Posts: 53
Joined: Sun Oct 14, 2012 11:59 am
Location: Normandy France
Contact: Website

Re: Canion Battery Readings

Fri Aug 01, 2014 1:24 pm

Seclog data base with CaniOn give Voltage and amperage every Second
The product of Volt by Amperage divided by 3600 give a result in Watt-h for each Seclog line.
Addition of each Watt-h value line after line give the energy amount.
I did the math and there was no differences with the values of CaniOn "Trip Timer"

Below is a quick easy way to get my Battery capacity after each Battery recharge with caniOn, an average should be done after several charges.

Battery total capacity = 60 / 50 x "Wh reg" x 100 / (100 - "SoC% initial")

"SoC% initial" : State of Charge before a full Battery charge with the help of CaniOn "Trip Timer", SoC% have to be above 15%
"Wh reg" .....: from CaniOn "Trip Timer", value in kWh after a full Recharge a "Reset" have to be done before charge

Exemple, SoC : 30%
After reset and charge, Wh Reg in kWh : 9.495

Battery total capacity = 60 / 50 x "9.495" x 100 / (100 - "30") = 60 / 50 x "9.495" x 100 / 70 = 16.28 kWh

Using Wh Out with the heater as I show above is more precise, and this is probably how Peugeot test our Batteries, but the math is not as easy.

BlueLightning
Posts: 53
Joined: Sun Oct 14, 2012 11:59 am
Location: Normandy France
Contact: Website

Re: Canion Battery Readings

Sun Aug 03, 2014 6:57 am

Battery Calibration

Here are some graphs not given, (for the moment?), by the CaniOn application.
Below on left, the green line is showing the energy given from the wall by "Canion SecLog" versus SoC. Like "Wh Reg", "wall energy" by Canion SecLog must be considered as "Untrue". Later I will show a comparison between a kill a Watt meter and "wall energy" by Canion SecLog. The Blue line is showing, the "Wh Reg" everybody see on screen and me with SecLog Data base. As said in other posts, those values are also "Untrue". For curiosity, the ratio of the slopes http://en.wikipedia.org/wiki/Slope, of those blue and green lines give the charge efficiency.
Charge efficiency = 0.1296 / 0.1618 = 80.1%, which is probably true, with math you can get true results from wrong numbers if they are wrong in the same way.

Let's go for the real subject now, Battery Auto Calibration.

If you follow the blue line, at 94 % SoC or State of Charge, you will see that the Car Battery Calculator realised that at 100% SoC or 11.4 kWh, (red dashed line), there will be a Battery voltage of only 360 Volt, (See the graph on right), which can let room for more charge till 361.2 Volt.
Smartly 3.5 % of SoC were added by the Calculator, and the charge restarted at 90.5 % SoC, in other words 100 % SoC after correction would have been 103.5 % in the former SoC scale.
See below those two graphs a green curve of the cell balance during this charge and a blue curve of the Battery voltage discontinuity at 94 % SoC before calibration.

Image
Battery Charge and Auto Calibration

On the graph below, see the Battery voltage discontinuity at 94 % SoC before calibration and see the battery Cell Balance given by Umax less Umin or the difference between the cell with the highest Voltage and the Cell with the lowest voltage during the charge, green curve.
At 50% SoC the balance is good and there is hardly any difference.

Image
Battery Cells Balance

offgridQLD
Posts: 101
Joined: Wed Jul 17, 2013 4:13 am

Re: Canion Battery Readings

Wed Aug 06, 2014 5:50 pm

Nice data.

Yes I have to agree that whr reg while charging didn't look correct and whr reg while driving was the same. The efficiency at different charge rates is different to . 2200w measured from the wall on our 2012 Jap spec Imiev. 3300w would be slightly more efficient. Perhaps some efficiency % could be tweaked into Canion to give a more true whr recharge number. I usually see 1600w - 1700w fluctuating when charging on canion screen. 80% efficiency would give me 1760w so its close but a little low reading.

The Auto recalibration is very interesting. So its using a max voltage to determine a full cell/battery pack and it can readjust its SOC scale to ensure 100% SOC alined with max cell/pack voltage.

Perhaps some times it gets this wrong and that's why some people have had there battery's only charge to 96 or 98% SOC on canion but the charger stops. Only to have it charge to 100% again on future recharges.

I wonder if there is the same corrections being made to the SOC scale at the bottom of SOC say when the pack gets down to 20 % soc as some of us have noted different battery voltages (some times significantly higher /lower) for the some SOC%.

I only have Canion and the dash gauge to tell me my SOC at the end of a long trip. At times I wonder if the SOC i am reading on the gauge and Canion SOC% is the true reading particularly around the 15 -25% SOC range. Just based on the cell voltage at the time .


Battery SOC meters are not simple if you want accuracy.

Kurt

Malm
Posts: 449
Joined: Mon Jul 22, 2013 12:16 pm
Location: Tábua - Portugal

Re: Canion Battery Readings

Wed Aug 06, 2014 7:18 pm

That corrections, I saw them so many times...

In the video of the temperatures, for example. https://www.youtube.com/watch?v=RkB2TG-gsvU. Goes directly from 82,5% to 90,5% - minute 2:45.

Or in this graph of canion:

Image

My car do it when charging, at high SoC and can sometimes do if I stop for some time. The I-MiEV DOESN'T recalculate SoC in movement (there is one exception, maybe another thing that nobody notice until now, only me).
Riding an i-Miev since 4/2011 in Portugal, 100.000 kms.

bobakka
Posts: 36
Joined: Mon Nov 03, 2014 8:03 am
Location: Oslo, Norway

Re: Canion Battery Readings

Sun Nov 30, 2014 2:23 pm

BlueLightning wrote:Below is a quick easy way to get my Battery capacity after each Battery recharge with caniOn, an average should be done after several charges.

Battery total capacity = 60 / 50 x "Wh reg" x 100 / (100 - "SoC% initial")



I am just starting to use Canion and have read this thread with a lot of interest.
My first question is why do you use the factor 60/50 to calculate Battery total capacity?
I would think the correct formula should be "Wh reg" x 100 / (100 - "SoC% initial")
You must have a reason to adjust number by 20%?
iMiev; New Feb 2011; 119000 km January 2017

BlueLightning
Posts: 53
Joined: Sun Oct 14, 2012 11:59 am
Location: Normandy France
Contact: Website

Re: Canion Battery Readings

Mon Dec 01, 2014 12:28 pm

The Wh Out given by the on board computer to canion are correct but the Wh Reg given by the on board computer to Canion are unreal and far too pessimistic.

With the help from Canion's Seclog data, below, see a graph Energy in kW-h versus State of Charge in %
Check the slope of the Charge line, 0.1344 kWh per % SoC or 13.44 kW-h for 100% SoC, this number in fact should be around 16 kW-h for 100% SoC, value of the Blue Lightning's battery capacity.

This is where my coefficient come from, 13.44 x 1.2 = 16 kw-h
A ratio of 60/50 or a coefficient of 1.2 is the best match for my iOn electric car which have a 16kWh battery capacity. In other words, this coefficient seems to give a proper result, but for my car only...

While driving, I get a higher slope, 0.1527 kWh per % SoC, but this number is also too low because those kW-h are a mixture of Wh-out when I press the accelerator and Wh Reg when I slow down.
No luck, the above coefficient is not good to correct in a real value those Wh Reg when I drive.

To deal with the Wh out only, the best way is to park the car and put the heat on to make a graph Wh out versus Soc, then me, and some others too, get a line with a slope of 16 kw-h for 100% SoC, Hallelujah..... or less....

No matter about no coefficients, your equation will give a number, if by the time your battery capacity decreases, this number will also decrease.....
Why 60/50 ? Energy is a function of power and time, I was wondering if Mitsubishi was not using the AC current frequency as a timer. AC current frequency in some countries is 50 Herz, in some others, 60 Herz....

ImageEnergy versus SoC

Return to “Batteries and Battery Management”