Are the non P cars geared differently front to back in the diff? If so the gearing of the front motor means it's always more efficient? (at the expense of low end performance)
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Chassis CAN Logging To ASCII Text Plus Graphing
- Thread starter lolachampcar
- Start date
-
- Tags
- Driving Dynamics Model S
Are the non P cars geared differently front to back in the diff? If so the gearing of the front motor means it's always more efficient? (at the expense of low end performance)
No idea, but there's definitely an opportunity for Tesla to do a lot of tuning with two motors to work with.
More detail on drag strip mode please. In particular, is a more aggressive torque ramp available (in addition to the increased power)?
Its just the internal name for max battery mode.
And yeah... power faded instantly after one 0-60 launch. Dropped max to 467 kW. Second launch to 465kW. Did a 0 to 100. Got logs I'll parse later tonight when I get home.... autopilotting to Charlotte right now.
Edit: battery was at 47C
MikeBur
ManualPilot
1. I would imagine so. I believe Bill mentioned that it appeared like there was immediate bias towards front, then shift to rear. I'll let him reiterate the reasoning there.
2. Here you go. Interesting fall off in combined torque ahead of throttle position reduction in Ludicrous.
- - - Updated - - -
Holy moly Batman!
- - - Updated - - -
Battery at 47C *before* or *After* those runs? I'd believe before as I've had this message after doing 5 runs:
!
Battery at 47C *before* or *After* those runs? I'd believe before as I've had this message after doing 5 runs:
View attachment 112717
!
Hmm! I hadn't looked at anything much *after* leaving the house. I did my first 0-60 about 1 mile from my house, and the pack was 47C when I left. I've CAN logged nearly my whole 60 mile trip just now (like 7 million frames...). So, I'll have to pick out some fun data in a bit.
MikeBur
ManualPilot
The data says so, yes. Here's the earlier 0-100 launch zoomed in. I also added the rear RPM as we have that now (thin blue line), showing this is closer to 1.25s
View attachment 112719
- - - Updated - - -
Hmm! I hadn't looked at anything much *after* leaving the house. I did my first 0-60 about 1 mile from my house, and the pack was 47C when I left. I've CAN logged nearly my whole 60 mile trip just now (like 7 million frames...). So, I'll have to pick out some fun data in a bit.
7M frames... yea - you're going to have to break that up before feeding it to anything like Excel - it's now barfing (stuck in CPU busy loop) on a bunch of comparison charts. Why can't it plot 3sets of data with 7,7,and 6 differing sets of data - it just needs to represent ~2.1M data points on a graph - geesh
- - - Updated - - -
When are you starting your reservation list? I'll take one. Do you want any data from a P85 Sig still on 6.2 (or just the P85D that has a Lud upgrade pending)?
Hey Brian, I believe I'm reasonably local if you want to just use the one I have? PM me contact details if you'd like. I won't be using it over the weekend.
- - - Updated - - -
I did manage to capture some data from a P85+, though I had difficulty keeping it's tires stuck to the road on launches. AWD spoils me here
Better on second attempt, though couple of examples of momentarily traction loss
40-75 acceleration, though interesting Torque control on motor; also notice the regen pattern as accelerator pedal is reduced.
btw - let me know if I should keep posting these graphs, or whether it's better just to keep them on the gDrive share?
@MikeB, by all means keep posting charts of interesting runs and events. If you try the run i mentioned maybe 0-30 is enough, safety first...
For the P85+ charts i noticed that the torque is leading the current, which doesn't make sense. Was there a shift in the timebase somehow during data processing? Also one of your charts up above in post 248 didn't show up, just listed as an attachment..?
For the P85+ charts i noticed that the torque is leading the current, which doesn't make sense. Was there a shift in the timebase somehow during data processing? Also one of your charts up above in post 248 didn't show up, just listed as an attachment..?
MikeBur
ManualPilot
-------------------------------------------------------------------------------------------------------------------------
Ok, back to P85D Insane / Ludicrous / etc
This next graph is going require a bit of explaining...
First, the context: In order to answer Bill's question re: are we really seeing significantly different torque response curves when accelerating from 0-70+, vs. 50-70+. So, any issues with brain exploding on information overload, blame him!
Second, the insane graph (pun intended)
4 sets of data
- Insane 50-75 == Dotted lines
- Ludicrous 50-75 == Thin Translucent lines
- Insane 0-90 == Thick Dashed lines
- Ludicrous 0-100 == Thick Translucent lines
Battery SoC/Temp data for the various runs in this order along top.
For simplicity, I summed the front and rear motor torque. Rear motor RPM is useful, though not available on Insane data (wasn't available).
I'm still studying this, though some things pop out at me, eg.
a) Immediately apparent that Battery SoC difference of 8.8% on Ludicrous vs. insane launch is very apparent on the combined torque, and speed curves, though interestingly not power
b) torque/power/rpm relationship vs. mph. Appears to be some smart limits being applied here, though I was told there would be no math
and here it is:
- - - Updated - - -
I'll check the P85+ charts, maybe I fluffed it. Will be a couple of hours though - life things to deal with here first
hmmm... don't know what went wrong with attachment. Here it is again, thanks for noticing:
Ok, back to P85D Insane / Ludicrous / etc
This next graph is going require a bit of explaining...
First, the context: In order to answer Bill's question re: are we really seeing significantly different torque response curves when accelerating from 0-70+, vs. 50-70+. So, any issues with brain exploding on information overload, blame him!
Second, the insane graph (pun intended
)4 sets of data
- Insane 50-75 == Dotted lines
- Ludicrous 50-75 == Thin Translucent lines
- Insane 0-90 == Thick Dashed lines
- Ludicrous 0-100 == Thick Translucent lines
Battery SoC/Temp data for the various runs in this order along top.
For simplicity, I summed the front and rear motor torque. Rear motor RPM is useful, though not available on Insane data (wasn't available).
I'm still studying this, though some things pop out at me, eg.
a) Immediately apparent that Battery SoC difference of 8.8% on Ludicrous vs. insane launch is very apparent on the combined torque, and speed curves, though interestingly not power
b) torque/power/rpm relationship vs. mph. Appears to be some smart limits being applied here, though I was told there would be no math
and here it is:
- - - Updated - - -
I'll check the P85+ charts, maybe I fluffed it. Will be a couple of hours though - life things to deal with here first
hmmm... don't know what went wrong with attachment. Here it is again, thanks for noticing:
My best graphing attempt.
(I couldn't make the left and right axis line up exactly, but they're close enough)This was at 97% SoC. Pack avg temp of 47.3C. 0 MPH to 60 MPH in 3.01s (no rollout) seems pretty dang fast. 0-100 MPH in 7.57s also seems pretty fast. Max combined torque of 750.7 ft/lbs (should have graphed that...).
For this run I estimate the 1-ft point to be 0.232655s in, so "0"-60 with rollout would be 2.78s. Using that I estimate drag strip 60-ft time at 1.62s and 1/8th mile time at 7.07s with a 98.3 MPH 1/8th mile trap speed.
This was the 3rd full throttle event after leaving the garage with 100% charge and a max battery power heated pack. The other graphs are similar enough that I wont really bother.
MikeBur
ManualPilot
Just redid all from the source data, though looking at the one that looked most egregious, this is the best I could redo - still looks the same to me, though added more granularity. Apologies if res of graph is low, working from laptop at the mo.
If this really isn't better then I'm afraid it looks like something that needs addressing in the logger.
- - - Updated - - -
My best graphing attempt.
This was at 97% SoC. Pack avg temp of 47.3C. 0 MPH to 60 MPH in 3.01s (no rollout) seems pretty dang fast. 0-100 MPH in 7.57s also seems pretty fast. Max combined torque of 750.7 ft/lbs (should have graphed that...).
For this run I estimate the 1-ft point to be 0.232655s in, so "0"-60 with rollout would be 2.78s. Using that I estimate drag strip 60-ft time at 1.62s and 1/8th mile time at 7.07s with a 98.3 MPH 1/8th mile trap speed.
This was the 3rd full throttle event after leaving the garage with 100% charge and a max battery power heated pack. The other graphs are similar enough that I wont really bother.
Nice! Tempted to do 100% SoC to see if I can repro. Think 90-100 SoC could make this much difference? (Looks like it does). Only other alternative is later FW version you're on?
Hmm.... sweet!
Edit: Showing max rear torque estimate at 688.5 Nm (507.81 ft/lb) and max front torque estimate at 332.5 Nm (245.24 ft/lb) for that data.
Wk, what Packconfig and FW ver are you running?
Definitely seems to make a difference. It's not something I'd plan on doing often though, as charging to 100% and then heating the pack is a surefire way to cause rapid degradation of the cells. Doing so is probably as hurtful to them as 20 cycles worth of normal use.
I'm on 2.12.45 right now. I have 2.12.126 pending, but I haven't installed it yet.
Since I logged nearly my whole trip to Charlotte, I'm going to try and graph the BMS_maxDischargePower variable to show the drop off of max power vs SoC. At 90% SoC it was already down in the 440s kW.
Last edited:
