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

Battery charge gauge linearity...

Discussion in 'Model S: Battery & Charging' started by ThosEM, Feb 25, 2014.

Tags:
  1. ThosEM

    ThosEM Space Weatherman

    Joined:
    Dec 13, 2013
    Messages:
    610
    Location:
    Annapolis, MD
    I've been deliberately running my battery down into the "quarter tank" range to gain confidence with that. The most obvious thing I've noticed is an indicated 185 km range at about 50% indicated, which is about 43% of full rated range, 426 km.

    At first, i worried this might reflect loss of battery capacity, even though I've only had the car a couple of months and less than 5000 km. Then I made my first supercharger stop and range charge, and it charged up to a full indicated 425 km range at 100% on the gauge. So evidently, the charge gauge graduations are not linear in rated range.

    Has anyone else noticed this? Has anyone else mistakenly attributed this to range loss as I did? Thanks for any comparison notes.
     
  2. Zextraterrestrial

    Joined:
    Mar 11, 2010
    Messages:
    3,636
    Location:
    Humboldt/Los Altos
    I think the curve is a tiny bit bow shaped but there is also some % of charge below 0 mi (/km) rated and this may explain what you are 'seeing'
     
  3. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    I think what is going on is that at zero miles (km), the gauge still shows several percent (the below zero reserve). Below is a picture of my battery from several versions ago at 34 miles while Supercharging an "A" battery. It is reading about 20%. Note that 20% of 260 miles is 52 miles. However is we assume the reserve below zero is about 22 miles, then 20% of 282 (260+22) miles is about 56 miles. 56 miles just happens to be the 22 reserve miles plus the 34 miles indicated.

    I don't know what the reserve is, and have never tested more than a couple of miles :eek:, but 22 miles makes this example plausible. Note that 22/282 is about 8%, so in this example, zero miles would read as 8% of the battery left.

    250-Amps.PNG
     
  4. ThosEM

    ThosEM Space Weatherman

    Joined:
    Dec 13, 2013
    Messages:
    610
    Location:
    Annapolis, MD
    That certainly seems plausible, and I guess I'd have to really run it down low to find out for sure. I thought that the rated mileage was simply proportional to the battery charge state, but there is a claimed reserve tank of about 5 kWh (25 km), as well as a further 4 kWh of "bricking" protection that can't be accessed in normal use.

    Has anyone seen zero indicated range when the "green bar" still indicates charge remaining? Where is "E" on this scale, really? And what ever happened to that thread where this stuff was studied and documented?
     
  5. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    I thought that I had a nice "zero" picture, but here is a 19 picture that I got just as a Supercharger session was starting. Here, (17+19)/277 = 13%, about what you see above, and 17/277 = 6%, so here "zero" miles equals about 6% on the battery. Note that at "zero" miles and about 6% on the battery picture (or just slightly higher), the battery turns red. Should have gotten that picture...

    As for documentation, it's all reverse engineering done on this forum. There is a nice battery picture around with "above zero." reserve, and "anti-brick" nicely charted for an 85, but I could not find it.

    Battery-19.PNG
     
  6. brianman

    brianman Burrito Founder

    Joined:
    Nov 10, 2011
    Messages:
    15,487
    Here's a couple shots of zero from February 2013:

    20130210Zero17.jpg

    20130210ZeroIC.jpg
     
  7. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    Based on other threads on balancing and capacity, I decided to do a few days of range charging while away from my MS, followed by a discharge to zero and another range charge. I started capturing the discharge data with percent battery, and plotted the data.

    This was all done with v5.8.8, and the the discharge was over three days. I have had my Sig S#37 for about 16 months and it has almost 14,000 total miles. Battery percent was estimated by doing iPhone screen shots from the App and enlarging the battery to get an estimated percent full, easily +/- 1%. Below are the data with a plot.

    As I collected the data, I started a plot with a linear regression fit. Everything was a great straight line, but the percent full for zero miles was being estimated at a little over 12%, which was different than my other calculations based on one low rated miles point. As the rated miles went below 40, the data were not fitting as well, and then I noticed that it looked like two lines that met at my 42 mile point. This plot shows two linear regression fits, one for 42 rated miles and greater, and one for 42 rated miles and less. Both fit well, with the zero mile percentage at about 7%, extrapolated to a 14 mile "below zero" reserve.

    Please take this as one set of data. I want to do at least another charge/discharge cycle or two to see if I can repeat these results. If this is repeatable, it appears that Tesla changes how percent full is calculated relative to rated miles above and below 40 or so rated miles. :confused:


    Rated-percent-1.png
     

Share This Page