Pack Performance and Launch Mode Limits

[wk057]

So, I picked apart some code from the car and did find a few related things. Unforunately my dissection is based on slightly older firmware (I've been documenting as much as possible in my reverse engineering of a 7.1 variant), so it's possible this information has changed or is out of date by now. However, I think it is still relevant.

First, a bit of a smoking gun: There is a hidden flag on the BMS labeled: BMS_w159_SW_WOT_Cycles_Exceeded (WOT being Wide Open Throttle, of course). This appears to get set when some predetermined number of WOT events is exceeded. I've not been able to determine that exact number.

Additionally, there appear to be two counters. One is labeled WOTCycles and the other is dsWOTCycles. I can not find or come up with a description for the "ds" abbreviation.

These counters to not appear to be broadcast over CAN anywhere. They are 16-bit counters, which suggests the trigger number is > 256.

In any case, when this is tripped the BMS broadcasts it as a hidden alert (only visible when the car is in diagnostic mode). However, this should be visible on CAN as a 1 in bit30 of 0x3C2 on the power train bus. This ID probably won't exist on the bus otherwise... almost all of the other error bits in it are catastrophic failures that aren't likely to happen, and alertMatrix IDs are only sent when one or more bits are set. I'm also not sure if this alert is cleared at some point after the event.

Suffice it to say, I'm reasonably certain that there is no way for service to actually remotely check these counters. I'll try to make time to dive into this in a newer firmware version to see if anything has been updated.

 

[qwk]

Just vow!
So, they've realized that most of the warranty motor replacement cases are caused by instant acceleration. So instead of testing their cars properly and fixing their design, they've decided to limit power of the owner's cars who are accelerating quickly.(which is what Tesla's marketing based on).

I've upgraded to P90DL, knowing that the tech difference doesn't worth the £30.000 difference between 90D, just the rear motor is slightly more powerful, but Tesla is also charging this money for possible warranty issues, since they are pushing the limits of the hardware. But, removing the power of the car to get away with it and expect drivers to not notice that reduction is not acceptable.

This is not exactly a new phenomenon. The "firmware upgrade", which is actually a firmware downgrade started in early 2013. The UMC could not handle 40A charging in the heat of summer, so instead of redesigning it so that it could, tesla simply cut the charging power via a "firmware upgrade". After much complaining, they issued another "firmware upgrade", so that the power charging remains cut, but the screen still shows 40A.

My advice to you all, is give Tesla a middle finger when they want to "upgrade" your firmware. After all, it's your car.

If they keep pulling these shinanigans, they will soon be sued out of business.

 

[smac]

@wk057 stab in the dark but could ds be (duration in seconds). To count time at WOT as well as the number of occurrences, you would have thought both would be important.

 

[davidc18]

So thanks to wk57 we know that even staying on fw 7.1 won't prevent counter death.

Tesla needs new advertising:.

Our performance models are a limited time deal or perhaps, model P - use it and lose it!

 

[wk057]

@wk057 stab in the dark but could ds be (duration in seconds). To count time at WOT as well as the number of occurrences, you would have thought both would be important.

Definitely plausible. I won't know more until I dig more into the BMS firmware.

 

[Tech_Guy]

To add insult to injury, your situation is causing you to inadvertently do a smidge bit of false advertising (just like Tesla)? Your moniker still shows "Always in Ludicrous Mode"!

Fixed LOL

 

[P85DEE]

So thanks to wk57 we know that even staying on fw 7.1 won't prevent counter death.

Tesla needs new advertising:.

Our performance models are a limited time deal or perhaps, model P - use it and lose it!

It only makes sense that they would not have just until very recently developed a method of "counting" the number of launch events and 100% pedal positions or WOT events.

I'm not surprised at all that they have a way of having the car mine this information, and then a way of later retrieving it from the car.

Now the question is, "What have they been doing with that information?"

The logical answer, is that they have been mining and comparing it after retrieving it from cars that have come back to them for warranty repairs of various types.

This is likely how they came up with their nice, round, 625 number.

 

[SDRick]

Fixed LOL

Too funny! Too bad Tesla does not react as fast as you did. Don't worry, rather than a refund I'm sure they would be happy to replace the letter badges that fell off your car.

 

[Tech_Guy]

So, I picked apart some code from the car and did find a few related things. Unforunately my dissection is based on slightly older firmware (I've been documenting as much as possible in my reverse engineering of a 7.1 variant), so it's possible this information has changed or is out of date by now. However, I think it is still relevant.

First, a bit of a smoking gun: There is a hidden flag on the BMS labeled: BMS_w159_SW_WOT_Cycles_Exceeded (WOT being Wide Open Throttle, of course). This appears to get set when some predetermined number of WOT events is exceeded. I've not been able to determine that exact number.

Additionally, there appear to be two counters. One is labeled WOTCycles and the other is dsWOTCycles. I can not find or come up with a description for the "ds" abbreviation.

These counters to not appear to be broadcast over CAN anywhere. They are 16-bit counters, which suggests the trigger number is > 256.

In any case, when this is tripped the BMS broadcasts it as a hidden alert (only visible when the car is in diagnostic mode). However, this should be visible on CAN as a 1 in bit30 of 0x3C2 on the power train bus. This ID probably won't exist on the bus otherwise... almost all of the other error bits in it are catastrophic failures that aren't likely to happen, and alertMatrix IDs are only sent when one or more bits are set. I'm also not sure if this alert is cleared at some point after the event.

Suffice it to say, I'm reasonably certain that there is no way for service to actually remotely check these counters. I'll try to make time to dive into this in a newer firmware version to see if anything has been updated.

My theory so far has been that the car had the ability to track this information in 7.1 but 8.0 added the ability to act on it. For example the day I upgraded my car to 8.0 I noticed the reduction in power at least to me that seemed to be to much of a coincidence, I'm quite curious about what you find!!

 

[P85DEE]

My theory so far has been that the car had the ability to track this information in 7.1 but 8.0 added the ability to act on it. For example the day I upgraded my car to 8.0 I noticed the reduction in power at least to me that seemed to be to much of a coincidence, I'm quite curious about what you find!!

Makes sense. The timing is just too coincidental.

 

[lolachampcar]

I second that the counters have been there since the beginning of time. Acting on the counters is new in 8.0 which is why the power cut showed up in that release.

 

[kirkbauer]

How do you guys track your power output? My 1.5yo P85DL seems less powerful than it used to be, but perhaps I'm just used to it.

 

[winfield100]

perhaps while you folks are discussing Tesla batteries and available power, charge, etc, etc, you might care to read a recent patent filing about Tesla batteries
".......BRIEF SUMMARY OF THE INVENTION

Disclosed is a system and method for identifying and responding to exceptional charge events of series-connected energy storage elements. The following summary of the invention is provided to facilitate an understanding of some of technical features related to detecting to exceptional charge events of series-connected elements and responses thereto, and is not intended to be a full description of the present invention. A full appreciation of the various aspects of the invention can be gained by taking the entire specification, claims, drawings, and abstract as a whole. The present invention is applicable to other implementations in addition to electric vehicles such as stored energy cases providing energy time shifting of renewable energy generation (e.g., solar and wind generators), to other arrangements of series-connected energy storage elements, and may be applied to other cell chemistries........."

United States Patent: 9529048

 

[Impulse20110]

Add your battery/power output on this thread to help us track and compare current and future power losses.

For those that do not know, To get Battery output download "PowerTools" on an apple device and log into your vehicle.
Then run app during drive and wala. Your exact power output specs.

Maximum Battery Kwatts Output By Owner

Would be sweet if someone made it onto a spread sheet that owners can update like the google drive sheet used for owners waiting on production.

 

[Tech_Guy]

Add your battery/power output on this thread to help us track and compare current and future power losses.

For those that do not know, To get Battery output download "PowerTools" on an apple device and log into your vehicle.
Then run app during drive and wala. Your exact power output specs.

Maximum Battery Kwatts Output By Owner

Would be sweet if someone made it onto a spread sheet that owners can update like the google drive sheet used for owners waiting on production.

you mean like this spreadsheet, you can click in the corner to add your car...

Tesla Model S PxxxD(L) Comparisons (Responses)

 

[Impulse20110]

you mean like this spreadsheet, you can click in the corner to add your car...

Tesla Model S PxxxD(L) Comparisons (Responses)

Wow. And it's already done.
Thanks!

Hopefully more people continue to upload so we can monitor what tesla is doing!

 

[Andyw2100]

Hopefully more people continue to upload so we can monitor what tesla is doing!

Looks like it may be spreading to other features of the car:

Headrest adjustment

 

[P85DEE]

Wow. And it's already done.
Thanks!

Hopefully more people continue to upload so we can monitor what tesla is doing!

Well I can certainly appreciate your enthusiasm. But in order to get any kind of a grip on this, one would have to do baseline measurements, which is what we have now and then compare them against periodic measurements in the same car to determine if any power had been robbed from that car.

The thing that stinks about it that, is the mere act of taking a measurement, might put you closer to the point where permanent power reduction will occur.

Do it enough times, and checking to see if you've been robbed, can result in you getting robbed.

And robbed blind.

Each time you run the test, could get you closer to the tipping point just for having run it.

 

[brianman]

Somewhat on-topic...

Does anybody have captured data to report max power output for Sport mode on a P85D, P90D, or P100D? I'm considering leaving my vehicle in Sport mode while we wait for this counter stuff to shake out.

 

[AWDtsla]

Somewhat on-topic...

Does anybody have captured data to report max power output for Sport mode on a P85D, P90D, or P100D? I'm considering leaving my vehicle in Sport mode while we wait for this counter stuff to shake out.

Didn't save the logs, but it's clearly a lot less power. Not that we can tell if that counts against the counters or not. If it's just looking for 100% throttle then you might get power limited sooner, not later.

Regardless... why? there's no situation in which you win here, unless it's just to buy enough time to give ludicrous test drives to the next owner.