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

Performance of P85D with Ludicrous upgrade review

Discussion in 'Model S: Driving Dynamics' started by thimel, Nov 30, 2015.

  1. thimel

    thimel Member

    Joined:
    Feb 27, 2015
    Messages:
    268
    First a quick summary for those who don’t want all the details

    The P85D with ludicrous upgrade is significantly faster than without. There is 19% more power above 30 mph, 0-60 time drops from 3.2 to 2.9 seconds and the quarter mile time drops from 11.8 to 11.5. So it was fast before and is faster now. You get about two-thirds of the improvement if you don’t heat the battery with max battery power.

    Now, all the nitty gritty for the rest of you.


    Before Lud run 1 Before Lud run 2 After Lud run 1 After Lud run 2 After Lud run 1 After Lud run 2

    Max Batt off Max Batt ? Max Batt on Max Batt on Max Batt off Max Batt off
    0-10 1 ft rollout 0.20 0.19 0.20 0.20 0.20 0.21
    0-20 1 ft rollout 0.62 0.63 0.63 0.62 0.63 0.65
    0-30 1 ft rollout 1.08 1.08 1.07 1.06 1.06 1.09
    0-40 1 ft rollout 1.64 1.64 1.55 1.56 1.57 1.60
    0-50 1 ft rollout 2.35 2.34 2.15 2.19 2.23 2.23
    0-60 1 ft rollout 3.213 3.200 2.893 2.970 3.059 3.044
    0-70 1 ft rollout 4.30 4.26 3.74 3.93 4.06 4.02
    0-80 1 ft rollout 5.58 5.53 4.92 5.08 5.29 5.22
    0-90 1 ft rollout 7.14 7.05 6.24 6.47 6.79 6.70
    0-100 1 ft rollout 9.020 8.873 7.843 8.162 8.649 8.494
    0-60 no rollout 3.543 3.485 3.168 3.271 3.302 3.376
    0-100 no rollout 9.349 9.159 8.118 8.462 8.892 8.825
    30-60 starting 0 2.133 2.120 1.823 1.910 1.997 1.955
    60-90 starting 0 3.92 3.85 3.347 3.50 3.731 3.656
    1/8 mile time 7.483 7.468 7.211 7.281 7.363 7.350
    1/8 mile trap speed 90.561 90.959 94.806 93.541 91.888 92.448
    ¼ mile time ------ 11.820 11.406 11.532 11.694 11.654
    ¼ mile trap speed ------ 112.176 116.013 114.541 112.309 113.150
    30-60 starting 29 2.175 2.166 1.902 1.967 ------ ------
    60-90 starting 59 ------ 3.81 3.36 3.55 ------ ------
    Note that my before ludicrous times are about 0.1 second slower than some people get. I probably didn’t stomp the pedal hard enough. I did it by bending my ankle. I tried to do it the same way on the after ludicrous upgrade runs. A few measurements are missing because I let up on the accelerator too soon.

    Below is a plot showing the before ludicrous upgrade quarter mile performance. Note the track is quite level. Only 5 feet of elevation change.

    quarter_mile1.PNG

    Plot below is same as above, after ludicrous upgrade and with max battery power on.

    quarter_mile2.PNG

    Below is a before ludicrous plot showing a test intended to measure acceleration onto a freeway. I started with the cruise control set on 30 mph (like you have just come around a curve on an on-ramp) (actual speed from vbox was 29.5) and then floored it.

    The 60-90 test was done in a fashion similar to the 30-60 and is intended to simulate passing someone on a highway. The reason for doing these tests separately from accelerating from a stop is to see if the acceleration might be quicker because nothing has started to overheat during the initial acceleration from 0. It also could be slower if Tesla chose to gradually ramp the power.

    30-60.PNG

    Plot below is same as above, after ludicrous with max battery power on. Notice the maximum acceleration increased from about 0.85 to 1.0 g.

    30-60v2.PNG

    A few more details about the testing procedure.

    I started the before Ludicrous tests early in the morning and had not driven the car for many hours, but had charged it that night. The ambient temperature in my garage that morning was 59 degrees and it was 50 degrees outside during the tests. I started with a 90% charge and by the time I was done the charge was 70%. Creep mode was off to help avoid a very slow start. Insane mode was on of course. Climate control was off. I drove a few miles before starting the first test but did nothing else to warm the battery. I drove 5 miles at moderate acceleration and speeds between acceleration passes. This was both to give a chance for things to cool down and to return me to the same starting point for each acceleration pass.

    Between the 3 passes of run 1 (1/4 mile, 30-60 and 60-90) and those of the second run, I set max battery power going to heat the battery. I waited half an hour and it still showed “preparing” so I gave up and did run 2. Hence, I’m not sure if the batteries had warmed up.

    The Powertools app indicated that my maximum power was 380 kW during the before Ludicrous tests and 451 kW after the ludicrous upgrade and max battery on and 426 with max battery off (after considerable driving). A graph of the speed and power vs time is plotted below.

    power_vs_speed.PNG

    Does anyone understand why both before and after the upgrade the power gradually drops as the speed increases? With the instrumentation available, I can’t tell if it is battery current or voltage that drops. VisibleTesla shows the battery current, but only about once a minute which is not useful for this purpose.

    The tests were done with a PerformanceBox made by RaceLogic kindly loaned to me by Pete90D. It had an external antenna mounted to the top outside of the windshield. All the included plots and most of the numbers in the table were obtained by uploading the data to vboxverify.com. I used the performance tools software that came with the PerformanceBox to calculate the 60-90 times and to check the calculations done by vboxverify.com.

    My first attempt to measure the after Ludicrous upgrade performance was half done when I realized I’d forgotten to put the memory card back in the PerformanceBox. Oops ;(. Got up early for nothing.

    I restarted the after Ludicrous upgrade with max battery power on tests early in the morning and had not driven the car for many hours, but had charged it that night, timing the charge to end just before I turned on max battery power. I waited about 40 minutes with the car still plugged in while the battery heated. It consumed about 6 kW from the outlet while warming the battery and running the climate control. The ambient temperature in my garage that morning was 52 degrees and it was 35 degrees outside during the tests. I started with a 90% charge and by the time I was done the charge was 69%. Creep mode was off to help avoid a very slow start. Ludicrous mode was on of course. Climate control was off.

    The next morning I tested the Ludicrous upgrade with max battery power off, timing the charge to end just before I left and remotely turning on the climate control for 15 minutes before leaving. The ambient temperature in my garage that morning was 50 degrees and it was 35 degrees outside during the tests. I started with a 90% charge and by the time I was done the charge was 77%. Creep mode was off to help avoid a very slow start. Ludicrous mode was on of course. Climate control was off. This time I only did 3 0.25 mile runs, skipping the runs starting at 30 and 60 mph. The maximum power seen in the three runs were 409, 425 and 426 kW. I figure the battery wasn’t warm enough for the first run and hence did not analyze that data. (I also forgot to turn creep mode off for that run.)

    I used VisibleTesla to measure how much power it takes to heat up the battery using max battery power. I had driven about 6 miles and it was about 50 degrees out. I plugged into my 240 V EVSE and then turned on the max battery power. The plot below shows the charger current drawn as a function of time as measured by VisibleTesla. The initial warming took about 45 minutes and consumed 19 Amps on average. That is 4,600 Watts for 0.75 hour or 3.4 kW-hr for initial warming. To keep it warm with the car parked it is heating about 1/6 of the time for an average draw of 0.75 kW. Presumably driving would heat the battery some, so the draw would be somewhat less for that case but I have no way to measure it. A colder outside temperature would presumably increase the average draw. If one is driving at freeway speeds this will cause about a 4% range reduction for my average of 340 W-hr/mi. Note that max battery power turned itself off after 3 hours of the car sitting idle.

    warming.PNG

    It has always been much more impressive to demonstrate the P85D acceleration by launching from a stop. The instant 1g acceleration that pushes one back in the seat is what scares/impresses/shocks/creates-adrenaline-rush-in people. One has to search for opportunities to do this by getting lucky and being the first in line at a red light (where one can only accelerate to 30 or 40 depending on the speed limit.) Freeway onramps usually have too much traffic to allow one to stop and then accelerate. However, it is quite natural on a freeway onramp to need to accelerate rapidly from around 30 mph to 65. So, my hope was that ludicrous would be as shocking from 30 mph as from 0. To quantify this, I have massaged the performance box data in Excel to smooth the acceleration and plot acceleration vs speed. The resulting plot is below.

    fig_accel_vs_speed.PNG

    Ludicrous does give significantly more g force. It is still over 1 g up to about 33 mph instead of 25. (This is pretty much the same information as it putting out more power, just in a more physically interesting way.)

    Comments about the results:

    1. Each run was repeated twice. The times are quite consistent (typically within 0.1 second).
    2. As expected, the performance below 30 mph is the same before and after ludicrous upgrade as the acceleration is torque limited there.
    3. Above 30 mph, ludicrous clearly has more power. This is seen directly with the PowerTools readout using the REST API which shows the maximum power increased from 380 to 451 kW, a 19% increase and by the shorter times to achieve speeds above 30 mph.
    4. If you don’t turn on max battery power, the maximum power was reduced to 426. This is not a very solid number as it clearly depends on the battery temperature which we have no way to measure. Anyway, using this number one gets about 2/3’s of the total Ludicrous improvement if you don’t turn on max battery power.
    5. The times to accelerate from 30 to 60 starting at 0 and starting at 29 are nearly identical. So there does not seem to be a limited slug or surge of power. (sorry lolachampcar.) The same is true for 60-90 starting from 0 and starting at 59. Another way to say this is that while we know if one takes the car on a track and constantly accelerates and brakes that the software limits the power after a while, presumably to protect something that is overheating. During a 0-100 run, with normal driving before it, this does not happen. This is good news.
    6. The max power measured from the battery was 451 kW. This compares to 458 kW that Pete90D measured on his P90DL. So the battery doesn’t make much of a difference. http://www.teslamotorsclub.com/showthread.php/52082-Happy-Birthday-to-me-it-s-a-P90D/page18
    7. The 0-60 time I got of 2.89 is also nearly identical to that Pete90D got of 2.901 s. I’m not sure of his state of charge or whether max batt power was on. http://www.teslamotorsclub.com/showthread.php/52082-Happy-Birthday-to-me-it-s-a-P90D/page20
    8. DragTimes got 0-30, 0-60 and 0-100 times of 1.1, 2.86, 7.56 respectively for their P90DL. http://insideevs.com/dragtimes-takes-delivery-ludicrous-tesla-model-s-p90d-video/
    9. This compares to my results of 1.06, 2.97, 7.84 for the P85DL. So they are a little bit faster, but the ludicrous upgrade clearly is more important than the battery upgrade as far as performance is concerned. This is exactly what Elon said would be the case.
    10. Of course, if there is ever a software update that lets people’s P90DL’s perform like the one Motor Trend tested then the P90DL’s will get a lot faster.
    11. All my measurements were taken between 70 and 90% charge. We need to see how the performance changes at lower charge levels, particularly comparing max battery power on and off. I plan to do this on my next long trip (not presently planned), just accelerating from 50 to 80 and checking the maximum power with the PowerTools app.
    Summary: The P85D with ludicrous upgrade is significantly faster than without. There is 19% more power above 30 mph, 0-60 time drops from 3.2 to 2.9 seconds and the quarter mile time drops from 11.8 to 11.5. So it was fast before and is faster now. You get about two-thirds of the improvement if you don’t heat the battery with max battery power.
     
    • Helpful x 2
    • Informative x 2
    • Like x 1
  2. tezzla

    tezzla Member

    Joined:
    Jul 14, 2013
    Messages:
    545
    Location:
    SoCal
    Nice job thimel.
     
  3. lolachampcar

    lolachampcar Active Member

    Joined:
    Nov 26, 2012
    Messages:
    4,449
    Location:
    WPB Florida
    Thank you!!!
     
  4. andrewket

    andrewket 2014 S P85DL, 2016 X P90DL (soon 100)

    Joined:
    Dec 20, 2012
    Messages:
    5,043
    Performance of P85D with Ludicrous upgrade

    Thank you for the detailed report!

    Now we need tesla to expose an API that allows us to set max battery remotely. I'd use it to heat the battery using shore power before a long trip. I suspect it would work better than just pre-heating the cabin.
     
  5. benjiejr

    benjiejr Technogeekextraordinaire

    Joined:
    Nov 20, 2014
    Messages:
    502
    Location:
    San Antonio, Texas
    Excellent write up @thimel. I really do appreciate all the time and effort you put into this. I'm sure you enjoyed it too!

    Regards, Ben
     
  6. tomas

    tomas Traded in 9 rep bars for M3, used to be somebody!

    Joined:
    Oct 22, 2012
    Messages:
    2,106
    Location:
    Chicago/Montecito
    Honest officer, it was in the name of science!
     
  7. mkspeedr

    mkspeedr Member

    Joined:
    Jun 14, 2015
    Messages:
    245
    Location:
    Santa Clara, CA
    Excellent information - thanks.
     
  8. yo mama

    yo mama Member

    Joined:
    Jul 22, 2015
    Messages:
    391
    Location:
    san jose, CA
    Well done!
     
  9. sorka

    sorka Active Member

    Joined:
    Feb 28, 2015
    Messages:
    3,263
    Location:
    Merced, CA
    Do you know what the SOC was on the 60-90 passes with Ludicrous max battery? My P85D 60-90 runs with max battery at 90% are:

    3.49
    3.52
    3.56

    I'm hoping that your best run of 3.347 was at lower than 90% and that the 3.7 run was way below 90%...hopefully. It certainly appear that max battery has much larger effect post L than before. if I charge to 90% I'll get something like 409KW without max battery vs 415KW with max battery.

    On your run where you got 451KW, do you have the pbox file for that run and does it include one of the 60-90 runs?
     
  10. P85DEE

    P85DEE Active Member

    Joined:
    Oct 10, 2015
    Messages:
    1,214
    Location:
    TMC
    #10 P85DEE, Nov 30, 2015
    Last edited: Nov 30, 2015
    Excellent write up. Very well done and nicely detailed.

    Down to 11.4 in the quarter puts it about a tenth behind th P90D with Ludicrous.
     
  11. shokunin

    shokunin P85 & S40

    Joined:
    Feb 28, 2012
    Messages:
    960
    Location:
    Irvine, CA
    Excellent write up! Seems nearly on-par with the current P90DL software version, that may change if Tesla ever releases the software they gave to Motor Trend.
     
  12. garwood

    garwood Member

    Joined:
    Jun 11, 2015
    Messages:
    11
    Location:
    Fredericksburg, VA
    Thanks for the very informative write up, I hope you realize that this may be very costly for all us fence sitters!
     
  13. LetsGoFast

    LetsGoFast Active Member

    Joined:
    Oct 13, 2014
    Messages:
    1,342
    Location:
    Virginia
    Just to be clear, this is speculation. We don't actually know how Motor Trend got the results -- we hope that it was software related, but it may be for a different reason.
     
  14. JQ11

    JQ11 Member

    Joined:
    Aug 19, 2015
    Messages:
    75
    Location:
    Jackson, WY
    Thanks as well for the detailed info - very helpful!
     
  15. hly

    hly Member

    Joined:
    Jan 31, 2015
    Messages:
    44
    Location:
    NJ
    Thank you very much for the detailed results.
    Regarding future software upgrades no one knows for sure but it is also possible that the "MT" version could bring similar improvements to P85DL
     
  16. Evee

    Evee Member

    Joined:
    Nov 24, 2015
    Messages:
    153
    Location:
    Las Vegas
    Awesome!!
     
    • Like x 1
  17. NSX1992

    NSX1992 Member

    Joined:
    Sep 2, 2014
    Messages:
    253
    Location:
    Thousand Oaks,CA
    Thanks for the hard work. I assume the Terminal Velocity for the drag runs are the actual speed obtained at the end of the 1/4 or 1/8 mile runs and the trap speed is over the last 60? feet you would get at the drag strip.
     
  18. tstafford

    tstafford Member

    Joined:
    Jul 4, 2015
    Messages:
    790
    Location:
    Nashville, TN
    Thank you so much. The 60-90 starting at 59 is so sweet to see. That's why I want the upgrade.
     
  19. thimel

    thimel Member

    Joined:
    Feb 27, 2015
    Messages:
    268
    In my table, I gave the trap speed which I believe is over the last 60 feet as you suggest. (It is an output from the vboxverify.com website.) Both trap and terminal velocity. Show beneath the first two figures.

    - - - Updated - - -

    Sorry about that.
     
  20. LetsGoFast

    LetsGoFast Active Member

    Joined:
    Oct 13, 2014
    Messages:
    1,342
    Location:
    Virginia
    There is a lot of good, related material in the other thread that started out discussing the scheduling of the upgrades and wandered into performance characteristics of the first couple of upgrades. It might be nice to merge the two together, probably by extracting the performance metric posts from the original thread.
     

Share This Page