I got a chance to try out the revised version of the Power Tools for Tesla app this weekend and pulled a 2.7 second 0-60 time in my P85D. Max KW was 375, SOC = 82%. First record (at 250 ms) was 15 mph, so maybe I was just lucky in how the data was acquired from Tesla's telemetry servers? Nevertheless, an interesting result <Sorry the image is so big>
Here's the raw data (a few columns omitted): TimeStamp_msSpeedSOCElevationEst_HeadingPower_kWShift_StateRangeEst_Range1.431617030371E+1215829241165D2041341.431617030619E+1221829241231D2041341.431617030867E+1225829241274D2041341.431617031117E+12318292221337D2041341.431617031367E+12358292221364D2041341.431617031617E+12408292221376D2041341.431617031867E+12438292221376D2041341.431617032117E+12488291221375D2041341.431617032367E+12508291221375D2041341.431617032617E+12548291221373D2041341.431617032867E+12578291221375D2041341.431617033118E+12608290221374D204134
These are my KW draws for various SOCs: 90% = 413 KW 89% = 411 KW 81% = 396 KW 70% = 383 KW 66% = 376 KW 60% = 366 KW So 375 looks low for 82%. I've done each of these 3 or 4 times on different days and get remarkably the same results. In fact, I pulled more that at 66%.
@Volt Can you add the 0 speed timestamp row as well? Looking at your data points, I'm seeing 2.747sec from 15mph to 60mph -- rather than from 0 to 60mph. 2747=33118-30371
@Brianman - I've included all rows. Do you think it's possible that I need to add 250 ms to account for the 0-15 time. that would put the overall time at just under 3.0 for 0-60, right?
Your first line notes the speed as 15mph. Without at least the preceding row (and maybe the row before that), it's hard to know how much adjustment you need.
Yea, the data file is incomplete and the app erroneously assumed the first stamp from the run was from 0. So it's not real.
Yeah, and to actually measure a 0-60 that anyone's going to believe you need a Vbox. Tesla's GPS is not high res enough.
What does Tesla's GPS have to do with anything. That doesn't factor into this at all. Did you think Tesla's speedometer input was GPS driven? it's driven by optical rotary disc from one of the wheel axels like all cars these days are.
That's fine and dandy, but if it were that easy then folks should just mount an iphone and measure 0-60 times by video. No need to complicate things further by grabbing REST API data as suggested by the OP. But it's not that easy. You really need a Vbox.
Well, I'll readily concede it may be a limitation of the app. That said, I did include all rows of data from the log file and I did start the run from a complete stop. If I understand it right (and I'll concede I'm no expert) our cars sample every 250 milliseconds to send data to Tesla's telemetry servers (from which this data is drawn). To be sure, i'm not asserting that my car actually did a 2.7 sec run...just that the app measured it as such.
A vbox will get you that extra degree of accuracy but both will be within a tenth of a second of each other. My g-tech pro always less than tenth off of what I'm getting the through the API. Granted, I have an older g-tech without GPS, but it was always within a tenth ET at the track back in the late 90s. You can't just use the speedometer with a video as the speedo is programmed to be a few MPH over the actual speed.
Ah, I see. So you are suggesting the speed provided thru the API is more accurate than that displayed on the dash? Interesting never knew that. Setting aside other sources of uncertainty, the base error in this measurement is +- .25 sec, as you correctly point out the 250 ms sampling frequency.