TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

Interesting SuperCharging data & charts on E-pack battery (P85D)

Discussion in 'Model S: Battery & Charging' started by MarcG, Feb 2, 2015.

  1. MarcG

    MarcG Active Member

    Joined:
    Oct 29, 2014
    Messages:
    1,668
    Location:
    San Francisco
    Hi guys,

    I've had my P85D for almost two months now, and over 4,000 miles later I'm collecting more and more data to understand how the beast works :biggrin:

    Today, I present you SuperCharging data that I collected over the weekend. A few points to note before diving into it:

    1. This is data from the Roseville, CA supercharger where there are 7 stalls
    2. I used the 4A stall (there is no 4B, unlike the other three A/B pairs) which probably means it has its own unshared cabinet
    3. Ambient temperature was 15ºC, or 59ºF
    4. I started charging with 46 rated miles left (about 18% SoC)
    5. The goal was to reach 100% SoC since I was driving far with no charging options
    6. Data was collected through VisibleTesla

    With that out of the way, here is the Voltage/Current/Range/SoC/Rate of Charge graph!

    Charging graph.png


    At the beginning of the charge, I got about 115.4 kW of power (363V x 318A):

    IMG_6538.jpg


    A little while later I hit the peak charging "speed" of 361 miles per hour of charge! Although power was a lower 112.5 kW:

    IMG_6552.jpg


    So here are some interesting findings:

    A. Battery current peaked very early (around 320A) then tapered off in an inverse log fashion
    B. Voltage started around 370V and kept going up steadily to just over 400 V, where it stayed constant from 90-100% SoC
    C. Rate of charge peaked early, after about 7 minutes, then tapered off steadily in a linear fashion
    D. State of Charge (and hence Range) increased rapidly until about 50%, slowed a little until 80%, then slowed a lot after 90%
    E. SOC and range stayed at the same level for the last ~10 minutes of charge
    F. The P85D is rated at 253 miles of range, but I was able to get 254
    G. The charging "speed" was lower (354 mph) at 115.4 kW than it was (361 mph) at 112.5 kW, so I'm not sure how that's calculated
    H. Tesla advertises a charging speed of 170 miles per half-hour of charge, which means 340 mph - not sure if that's peak or average

    Enjoy!
    -Marc
     
  2. SteveWest

    SteveWest 2012 Roadster 2.5, 2014 P85+, 2015 P85D

    Joined:
    Jul 15, 2012
    Messages:
    175
    Location:
    New Zealand
    Oh that's easy, the "speed" calculation is calculated from the power level, but is heavily filtered / smoothed which introduces a delay. You'll notice on an AC charge the power level is constant almost immediately but the speed ramps up very slowly over several minutes.

    Steve
     
  3. Cosmacelf

    Cosmacelf Active Member

    Joined:
    Mar 6, 2013
    Messages:
    3,399
    Location:
    San Diego
    For some insane reason (ie. the programmer didn't have a clue what he was doing), every single number on that charge info is an instantaneous number EXCEPT for mi/hr charging speed which is an AVERAGE since the session started.

    It makes zero sense, and it constantly causes confusion. I wish Tesla would change this behavior.
     
  4. TexasEV

    TexasEV Active Member

    Joined:
    Jun 5, 2013
    Messages:
    3,784
    Location:
    Austin, TX
    F. The display gives an approximation based on an algorithm. Showing 1 mile more does not mean your range is actually 1 mile more.
    H. is not true. The rate of charge tapers as the battery fills. The 170 miles in "as little as" 30 minutes is when starting with a battery that is almost empty. You can't extrapolate that to the next half hour to get an average miles/hr, even if you are starting with an empty battery, and the 170 is best case scenario. Most sessions start with more than 10% state of charge and would get less than 170 miles added in 30 minutes.
     
  5. apacheguy

    apacheguy Sig 255, VIN 320

    Joined:
    Oct 21, 2012
    Messages:
    3,718
    Location:
    So Cal
    That's great stuff. Now can you plot charger power vs SOC?
     
  6. MarcG

    MarcG Active Member

    Joined:
    Oct 29, 2014
    Messages:
    1,668
    Location:
    San Francisco

    I looked back at all my supercharging data since I got my P85D, and the three events where I charged around 170 Rated Miles (RM) are:

    LocationAvg TempStart RMEnd RMCharged RMCharge TimeAvg Power (kW)
    Detroit Lake2ºC / 36ºF1018917939 minutes82.2
    Mount Shasta0ºC / 32ºF1217916738 minutes78.7
    Roseville22ºC / 72ºF2920217341 minutes74.9
    So it doesn't look like it's achievable to get 170 Rated Miles in 30 minutes, at least not in the P85D...



    VT unfortunately doesn't sample power as often as SoC (or Voltage, Battery Current, Range, etc.) but here you go anyway:

    Power-SoC.png

    If I could manipulate the source of the data, I would just multiply Voltage by Battery Current and plot Power accordingly.
     
  7. apacheguy

    apacheguy Sig 255, VIN 320

    Joined:
    Oct 21, 2012
    Messages:
    3,718
    Location:
    So Cal
    @jpasqua - Any interest in addressing this issue in a subsequent release?

    @MarcG - Here's what a taper curve looks like for us "classics" in case you're interested:

    Older Teslas limited to 90kW Supercharging - Page 160
     
  8. MarcG

    MarcG Active Member

    Joined:
    Oct 29, 2014
    Messages:
    1,668
    Location:
    San Francisco
    Thanks! I've read something similar before.

    @Joe, I'm assuming that the grey Power curve you're reporting in VT is what Tesla sends, not one you're computing within VT.
    So I think the suggestion would be to have another power curve labeled "Charge Power" that is your own computation of Voltage vs Current (kind of like how you have Current vs. Battery Current).
     
  9. Benjamin Brooks

    Joined:
    Jan 21, 2013
    Messages:
    444
    Location:
    San Jose, California, United States
    It's simply the "rolling average mph" (derived from total kWh delivered over time) from the start of charge until present. (I guess "heavily filtered / smoothed which introduces a delay" is a semantic over-complication IMHO)

     

Share This Page