breser
AutoPilot Nostradamus
As entertaining as these what if's are. I seriously doubt they are mucking with the build to "mess with us."
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Andyw2100
Well-Known Member
MarcG
Active Member
anticitizen13.7
Not posting at TMC after 9/17/2018
Haha that's hilarious, thanks!! Way to end the work day with a good laugh
Funny for us :biggrin: I wouldn't want to be Tesla's software team right now... I am sure that someone is in the doghouse over this
Funny for us :biggrin: I wouldn't want to be Tesla's software team right now... I am sure that someone is in the doghouse over this
Im not so sure. This just doesn't smell like a decision made by one guy at his desk. I'm convinced that there is a very specific reason that they did this, that involved the team leadership to a high level.
Im also not convinced that there isn't a ring of truth in the thought that if they include it in "range mode" that it would look better than it was. And the non-announcement of the addition of the feature also smells funny. Why be so evasive?
Because it still doesn't address the shortfall.
I'm looking forward to seeing the side by side. I don't think this is the fix that solves the problem. I had it installed today and did a short 60 miles - its a run I do fairly regularly and there was no change whatsoever - no charges in temp; weather or traffic; all pretty constant. Something is missing - Id put money on it. No notes because it didn't work as intended.
MarcG
Active Member
Ok after installing 6.1 (2.2.139) last night, I drove the same leg to work this morning and got good results.
Before I shared the detailed data, here are some important points to note:
1. I always pre-heat the cabin to a target 20ºC/68ºF about half an hour before my drive, although it rarely reaches that temp
2. The section of my commute is all highway with little to no traffic, since I start the "clock" after the bay bridge ends
3. There is only 3 feet of net elevation on this stint of the commute
4. I've tried to be as scientific as possible by:
a) Resetting Trip A at the beginning of the traffic-free highway
b) Taking a photo of the range (Rates Miles left) and temperature at point A) above
c) Driving with TACC set to 75 mph, following distance set to 1, and "smoothening" it when it slows down too much for a car ahead
d) Taking a photo of the RM left, temperature, distance traveled, kWh used, and Wh/mi
e) Using the exact time (including the seconds) of each photo to determine time travelled
f) Dividing distance travelled by time travelled to get exact average speed
With that, here's the data!
So it seems that consumption is indeed better with 6.1 (2.2.139), and even though the Wh/mi isn't much lower, the kWh used and RM used definitely are.
BTW this was all done in INSANE mode, as always, but this time RANGE MODE also ON since we're now allowed to use it with INSANE.
Once I return home this afternoon, I will post similar data for the reverse commute. Average distance is about twice as long since there is no traffic on the bay bridge at the time I drive.
Before I shared the detailed data, here are some important points to note:
1. I always pre-heat the cabin to a target 20ºC/68ºF about half an hour before my drive, although it rarely reaches that temp
2. The section of my commute is all highway with little to no traffic, since I start the "clock" after the bay bridge ends
3. There is only 3 feet of net elevation on this stint of the commute
4. I've tried to be as scientific as possible by:
a) Resetting Trip A at the beginning of the traffic-free highway
b) Taking a photo of the range (Rates Miles left) and temperature at point A) above
c) Driving with TACC set to 75 mph, following distance set to 1, and "smoothening" it when it slows down too much for a car ahead
d) Taking a photo of the RM left, temperature, distance traveled, kWh used, and Wh/mi
e) Using the exact time (including the seconds) of each photo to determine time travelled
f) Dividing distance travelled by time travelled to get exact average speed
With that, here's the data!
| Date | Distance | Time | Avg Speed | RM Used | kWh Used | Wh/mi | Avg Temp | Cabin | Roads | Weather | SW Version |
| 1/26/15 | 9.0 mi | 7m18s | 73.97 mph | 12 | 3.4 | 370 | 10ºC/50ºF | Seat Heater (2) | Dry | Cloudy | 6.1 (2.2.113) |
| 1/27/15 | 10.1 mi | 8m34s | 70.74 mph | 12 | 3.3 | 327 | 17ºC/63ºF | None | Dry | Cloudy | 6.1 (2.2.113) |
| 1/28/15 | 9.6 mi | 7m51s | 73.38 mph | 13 | 3.6 | 378 | 17ºC/63ºF | None | Dry | Sunny | 6.1 (2.2.113) |
| 1/30/15 | 9.6 mi | 7m48s | 73.85 mph | 12 | 3.4 | 359 | 12ºC/54ºF | Seat Heater (1) | Dry | Cloudy | 6.1 (2.2.113) |
| 2/2/15 | 9.7 mi | 8m01s | 72.60 mph | 11 | 3.3 | 343 | 12ºC/54ºF | Seat Heater (1) | Dry | Sunny | 6.1 (2.2.113) |
| 2/4/15 | 9.7 mi | 8m14s | 70.69 mph | 9 | 3.0 | 314 | 14ºC/57ºF | None | Dry | Cloudy | 6.1 (2.2.139) |
So it seems that consumption is indeed better with 6.1 (2.2.139), and even though the Wh/mi isn't much lower, the kWh used and RM used definitely are.
BTW this was all done in INSANE mode, as always, but this time RANGE MODE also ON since we're now allowed to use it with INSANE.
Once I return home this afternoon, I will post similar data for the reverse commute. Average distance is about twice as long since there is no traffic on the bay bridge at the time I drive.
Andyw2100
Well-Known Member
MarcG
Active Member
Pleasure. Looking forward to sharing more data on longer trips.
You know what I actually found even more interesting: the "hum" I was hearing at speed (above 60 mph) when motor power was over 15 kW now seems to be gone when RANGE MODE is ON (torque sleep active).
Check out the following thread:
P85D: loud motor humming sound at highway speeds and over 15kW power (or regen)
ArtInCT
Always Learning
MarcG:
First off, thanks so much for your continuing efforts on the P85D Efficiency testing and logging of data.
The 5 previous trips from 1/18 to 2/2 on 2.2.113 averaged about 355 Wh/mile... so the most recent trip of 314 Wh/mile is about a 12% increase in efficiency!
Although only one data point and a short trip as well, it is a very good sign.
314 Wh/mile at a 90% SOC would get about 210 mile range according to my calculations based on 67400 watts in the 90% battery pack usable.
Again thanks for your diligence and commitment.
Art
First off, thanks so much for your continuing efforts on the P85D Efficiency testing and logging of data.
The 5 previous trips from 1/18 to 2/2 on 2.2.113 averaged about 355 Wh/mile... so the most recent trip of 314 Wh/mile is about a 12% increase in efficiency!
Although only one data point and a short trip as well, it is a very good sign.
314 Wh/mile at a 90% SOC would get about 210 mile range according to my calculations based on 67400 watts in the 90% battery pack usable.
Again thanks for your diligence and commitment.
Art
Andyw2100
Well-Known Member
Just checked that thread (which I had seen before, but not in the past few days) and commented.
My comment here, though, is that one interesting aspect of this is the fact that based on your video, range mode is definitely doing something. Before this, we really didn't know, since there were reports of improved range even with range mode left off. Most people reporting improved range have been testing with range mode on, and I don't think anyone yet has reported doing a test with .139 that shows results before .139, with .139 and range mode on, and with .139 and range mode off. So before this, it was possible that there was documentation about the efficiency increase in torque sleep, but that it really existed elsewhere too. Of course that still may be the case, but clearly range mode is now doing something that must be affecting the motors, since your hum is gone.
Although I did just think of something: had you ever tried range mode before? (I know before yesterday I had not.) I mean I guess there could be some possibility that even before whatever Tesla did with respect to efficiency improvements, range mode in and of itself could have eliminated the hum if you had never had range mode on before. Just a thought.
- - - Updated - - -
One important note about Marc's numbers, as compared to Tesla's published range numbers, Art, is that Marc was averaging around 72 MPH as opposed to Tesla talking about range at 65 MPH. You'd be surprised, but that speed difference actually makes a pretty big difference in energy consumption. Just wanted to point that out. If you're planning to drive about the same speed as Marc, it may not be relevant to you, but if someone is reading this casually, and trying to compare numbers to Tesla's numbers, it's important that they be aware of that difference.
EarlyAdopter
Active Member
I did that a couple of days ago, here.
I didn't see any difference. Could either be because torque sleep isn't actually activated yet, or because it's always on (though, I had considerably worse efficiency on .139, which is consistent with the temperature difference), or because the hilly terrain my test route is over doesn't lend itself to torque sleep.
What I would ask everyone making anecdotal observations of .139 and Range Mode On is please do a back-to-back run over the same highway segment, once with Range Mode on and immediately after with it off. That's the only way to get real data. Comparing your usage today with what you saw a few days ago isn't a valid comparison. Too many other variables could have changed (temperature, prevailing windows, traffic, tire inflation, etc).
ArtInCT
Always Learning
Yes I agree. Furthermore, MarcG's logged trips into what appears to be snow country for downhill skiing/boarding often times have an incredible elevation component of + and - along with a brisk MPH rate (and many passengers and gear). None the less, his test data as presented here do characterize the effects of high speed flat cruising with TACC and the so called Torque Sleep in Range mode. Here in the Northeast, on the congested and undersized roads I typically travel, 70MPH rates are not the "norm".
What I do find incredible, is that so far, most Torque Sleep Testers here on TMC have NOT, repeat NOT, mentioned once that they in any way "Felt" or "Were Aware" of the Torque Sleep implementation in the P85D. They point to Wh/m as the tell. This is very important to me personally as I do not want my future P85D to let me know via some performance tell-tales that Torque Sleep is on. The implementation appears to be very good. The so called selection of Torque Sleep functionality is questionable in my mind.
I hope the guys at DragTimes can give us some 0-60 data on Torque Sleep (TS) enabled runs and disabled runs. If they are the same times, then I think that Torque Sleep should probably be ON or enabled all the time as the drive dynamics would auto-magically engage TS.
Last edited:
Andyw2100
Well-Known Member
Ahh, yes, I had seen that.
I guess what I really meant, though I failed to say it, is that we haven't seen someone who has seen an increase in efficiency who did a comparison between pre-.139 and post-.139 with range mode on, then also include a comparison of post-.139 with range mode off.
As for your comment about wanting people making anecdotal observations to do back to back runs, I understand what you're getting at, but I'm not sure you intended for your comment to be taken literally. I mean people like Marc, who have excellent data, should continue to post whatever they have, at least in my opinion, but clearly Marc wasn't going to turn around once he got to work, drive back home, and then drive in to work again with range mode off just to satisfy our desire for a better controlled experiment.
For those that want to do a well-controlled experiment and collect the most meaningful data possible, of course back to back (or side by side) runs would be important. But when that's not possible, the kind of data Marc presented, and any reasonable anecdotal data being presented, as long as it comes with the background information we need to properly evaluate it, is helpful.
That's my opinion, anyway.
Edit: A good argument, by the way, for the value of anecdotal evidence would be how quickly we, as a group, were able to determine that the P85Ds were not delivered with the efficiency numbers we expected them to have. Eventually wk057 did his very well controlled and documented side by side comparison with his fiance's P85, but long before then we had a pretty good idea of how far off the numbers were. That idea came from anecdotal reports. I say keep the anecdotal reports coming!
Last edited:
EarlyAdopter
Active Member
Yes, sorry - I didn't mean to degrade the efforts and data posted thus far. It's just that the engineer in me naturally insists on controlled experiments before being able to draw any definitive conclusions.
While I certainly wouldn't expect anyone to drive to work twice how about this - could a few other people take a drive tonight, say, at 9:00 after the kids have gone to bed or your favorite TV shows are done and when traffic is light? Drive a 10 mile segment of highway with Range Mode off, then immediately go back and drive the same segment again with Range Mode on and post the results. Preheat the cabin on shore power before heading out.
While I certainly wouldn't expect anyone to drive to work twice
how about this - could a few other people take a drive tonight, say, at 9:00 after the kids have gone to bed or your favorite TV shows are done and when traffic is light? Drive a 10 mile segment of highway with Range Mode off, then immediately go back and drive the same segment again with Range Mode on and post the results. Preheat the cabin on shore power before heading out.Now the question becomes, did efficiency improve due to 'torque sleep/distribution' or didn't it and is the difference you are seeing due to other things influenced by range ON/OFF? Ie, what results did you get with RANGE on for that stretch prior to the latest firmware?
How does the engineer in you feel about the whole manner that this is being rolled out? I ask just to get a level set on my inclination to shake my head. It seems so inappropriate that there is such a lack of information on this release and the notes specifically around this suspected fix to the range issue. Case in point, I called Tesla to ask what exactly 2.2.139 was this morning. Called the 800 support number...here is a quick recap:
them: Tesla, <name> speaking.
me: would you like to start with the last 6 of my VIN?
them: yes please
me: I supply - and follow with the background in being a P85D owner seeking to understand that .139 has addressed the range issue.
them: you are the first person that I have heard about having issues about range.
me: really? Are you sure? There are publications on your web pages to explain that a fix is coming.
them: ah yes. We did release a fix.
me: so is that in .139 then?
them: I dont know sir. I mean I think so. You are speaking with sales...
me: (pause/internal sigh/facepalm) who do a speak with to confirm the details?
them: hold please. (1min wait). Let me put you through to tech support.
tech support: (beginning first with the identifying attributes like VIN and address info) How can I help you?
me: as you can see by my account I own a P85D and I am trying to determine what was in the .139 release. Specifically does it address the range issues that were slated to be addressed in January.
tech: ah yes. Well .139 was really sent as a firmware bug fix to address the 6.1 release. There is no specific component that addresses range.
me: I thought that might be the case, but are you sure? I mean 'range mode' now has a disclaimer that it includes 'torque management' as part of the range optimization. Im wondering if this is Tesla's attempt to address the issue. Of course there are no notes hence my call to you.
tech: I agree sir that this is odd. I have no information on that.
me: during a release, is your team usually informed of the updates and material changes?
tech: we certainly are.
me: but in this case you have heard nothing and the new addition to range mode is new information to you as well - anything like an internal briefing or more detailed information that you can reach to?
tech: thats a good question sir. And, you bring up a valid point. Id like to go and do some fact finding. May I call you later with what I find?
me: yes please (confirm contact information etc...)
kinda shaking my head right now....
Vger
Active Member
Received a response from Jerome on a few things. (Jerome seems to be the go to guy for answers these days)
Most notably:
So, no more complaining about it only working in range mode.
This is just as I thought, and surmised yesterday! This is very good indeed.
OK, so due to the long discussion about range mode, and what it does and does not do, and to get a comparison for absolute efficiency to the previous P85 benchmark, I did another couple test runs today, using the same route and methodology as yesterday, but in our "classic" P85. That car is running 6.1(2.2.115). Conditions were similar but a bit more rainy. The P85 has Michelin X-ice3 tires, instead of the Michelin Primacy's on our P85D. They are also fairly new, but probably more broken in. They are a bit quieter than the Primacy's, so perhaps a little less rolling resistance.
In an case, I again did two runs. The first was with the "old" Range Mode ON, and the second with it OFF. Throughout, I had climate control turned completely OFF. The hypothesis was that if Range Mode really ONLY controls the HVAC power on a P85, it should have no effect if HVAC is off.
And the envelope, please:
P85 S02001:
First lap, Range Mode ON = 174 Wh/km (280 Wh/mi)
Second lap, Range Mode OFF = 174 Wh/km (280 Wh/mi)
Compare with, P85D P52512:
First lap, Range Mode OFF = 205 Wh/km (330 Wh/mi)
Second lap, Range Mode ON = 188 Wh/km (303 Wh/mi)
Discussion: So it does seem to be the case that the old-style Range Mode does nothing when HVAC is off. It seems likely that the Range Mode benefit I saw yesterday with the P85D (also with HVAC off) is purely due to the incremental "magic" they are now doing with torque distribution, post .139.
It also seems that the P85D is still not quite as efficient as the P85 in these slower, hillier conditions, but that is perhaps reflective of the difference in EPA 5-cycle range (253 mi vs. 265 mi). However, this might be entirely due to the tire difference between our two cars, so I would not make much of that.
I think the great news is that Tesla does seem to have delivered what the claimed in "D" efficiency, and almost on time! It will be fascinating to see how the 85D's fare when some here get their hands on theirs.
Last edited:
Similar threads
