Pack Performance and Launch Mode Limits

[Tech_Guy]

I really hope Tesla will address this...

If you don't mind me asking, do you see any yellow doted line indicating that power is cut from the nominal during the launches?

No power limited dotted line under acceleration or warning triangles on the energy graph on the dash... when i charge to 100% i do see the normal dashed lines only on the regen side but nothing on the power side.

 

[Tech_Guy]

Another good thread to look at is @Old Man Mike 's thread on max battery power vs state of charge and temp Max Power vs State of Charge and Temp

heres a graph to look at... Drawing a line following that same arc from 85% at 432 KW my 100% charge with max battery power would yield around 445-450 ish KW which I will confirm with actual testing this weekend...

The conversion for KW to horsepower is 1.341, using this my car made 685HP at 511KW and now makes between 596-603HP at 445-450KW A loss of 82-89HP in 6 months since my purchase. Since Tesla has sandbaged my car down from 511 KW to 450KW, a P100D upgrade would now yield me 160 ish HP over what I'm making now vs the 75 more it would have been before reducing my cars power... I still fear this was an effort to make the 100 upgrade look even better to owners especially since they neglected to let us know before imposing the change...

 

[AnxietyRanger]

I was at the Dallas service center this afternoon and asked an experienced tech about the launch limitations and he confirmed there is indeed a "counter" which tracks the number of perfect launches for Ludicrous model vehicles. He said it affects ALL ludicrous vehicles, be it P85DL, P90DL or P100DL. He couldn't give me a minimum number of launches before the power would be reduced and he couldn't give me an answer as to how much power would be reduced even if he knew. He verified this is to protect the powertrain components. The same powertrain is used for all P versions and has not been beefed up compared the earlier cars like the P90DL. Launch mode must be enabled and the launch must be in perfect conditions to count. I wish I had more time to ask him about additional details. Definitely makes me want to NOT purchase or lease a P model in the future.

Thank you for this update.

One thing to note here (which might also go some way explaining the discrepancy @vgrinshpun pondered about), assuming this report is accurate representation of the tech's words, I do not think there is a claim that all the models use the counter in the same manner.

We have heard P90DL activated the limit of death after the 8.0 software update and I think the concensus has been gravitating towards it doing this based on previously gathered data. Well, what if the counter itself is simply one of many parameters the car follows, but that does not automatically mean anything more? It is just data. It was being collected already on 7.1 (and previous?) and is being collected on any car with launch mode. Maybe it is called a "Launch Mode Counter" or something (though we assume it can get triggered by non-Launch Mode events as well, but that's a separate conversation) and sits next to some other counters like the "Left Seat Heater Turn On Counter" (made that up) that Tesla uses to monitor car conditions for statistical purposes.

Nothing wrong with a counter itself. It is possible Tesla is not using the counter for any performance penalties in P85DL or P100D. A benign use case for counters could be, say, Tesla's own warranty claims to contractors for example (contractor promising a part to last X number of launches). The tech said he didn't know the limitations, so he/she would not necessarily have known where it applies. It seems plausible, he/she may just know the counter because it can be seen in some analytical tool, but not its effects which may not be disclosed to the techs at all (maybe some internal memo was circulated with vague wordings similar to the CYA). He may not know exactly what triggers the counter addition either, beyond perhaps some small descriptive note in the analytical tool that may or may not be the full story.

Now, what is problematic is the precedence being created here. Now that we think we know the counter is global, and Tesla has done this once and added a global disclaimer to boot, who is to say it won't happen again later for other models? Even more so, because there probably are other counters as well and Tesla is known to update software all the time, who is to say they won't disable or diminish other features as well to protect themselves from warranty claims? That seat heater being problematic, why not start limiting its availability pre-emptively by software in the entire fleet... (again, just a made up example)

It certainly is a problematic precedent in that sense, but not necessarily yet indicative of being used beyond P90DL. Pending further data, of course.

As for @vgrinshpun cherry-picking, I personally didn't find that problematic. I am not sure his logic holds regarding the specific criticism levelled aginst the tech (I think my speculation above would be a plausible explanation), but certainly it is not automatically negative cherry-picking to discuss what parts of the tech's comment - or the report of the tech's comment which may be different - could be mistaken or misunderstood. It is certainly possible a tech could know one thing and be mistaken about another, and even more so when considering a later report of the tech's words could have misunderstood and misreported something - the broken telephone or lost in translation effects...

 

[Sunlight]

I have been following this thread out of curiosity but not read every comment so this may have been addressed.

I was wondering if something similar is employed by other manufacturers (especially supercars) in their ICE/hybrid cars? The software must be fiendishly complex and I wonder if anyone has 'analysed' it deeply to see if there isn't some sort of counter/control in a similar manner to that which is suspected in certain Teslas.

Would a Porsche 918 or a LaFerrari (etc) have exactly the same performance and drag times after 2 years/50 launch mode starts/10 000miles...??! If not, would that be put down to 'wear and tear' or something more akin to Tesla's 'trick'? It would be even easier than 'hiding emission control data'..!

Other than TechGuy has anyone else actually compared new and current drag figures on their own cars?

What is the difference? How many are thus affected?

 

[Spidy]

I have been following this thread out of curiosity but not read every comment so this may have been addressed.

I was wondering if something similar is employed by other manufacturers (especially supercars) in their ICE/hybrid cars? The software must be fiendishly complex and I wonder if anyone has 'analysed' it deeply to see if there isn't some sort of counter/control in a similar manner to that which is suspected in certain Teslas.

Would a Porsche 918 or a LaFerrari (etc) have exactly the same performance and drag times after 2 years/50 launch mode starts/10 000miles...??! If not, would that be put down to 'wear and tear' or something more akin to Tesla's 'trick'? It would be even easier than 'hiding emission control data'..!

Other than TechGuy has anyone else actually compared new and current drag figures on their own cars?

What is the difference? How many are thus affected?

Nissan got into some trouble: http://jalopnik.com/5565919/this-is-nissans-gt-r-launch-control-warranty-settlement/

On the other hand there is Porsche...

At the end of the day I don't see any issues with limits or voided warranties AS LONG AS IT IS CLEARLY DISCLOSED WHEN YOU PURCHASE THE CAR.

 

[Old Man Mike]

Just to help everyone understand the significance of exceeding the Launch Limit counter, here is a power chart which shows the drop relative to the various Tesla models:

A few key notes for those that have not kept up with everything in this thread:

1) Only launches made with the special Launch Mode is counted. Just slamming down the accelerator does not count.
2) The only data we have showing the permanent power drop is the one point from Tech_Guy.
3) Tesla has stated that this is real but has not provided much in the way of specific details.

In retrospect, I bet that Tesla wishes they had never added the special Launch Mode. Most said that it did very little to reduce 0-60 times and now look at the turmoil it has caused. I always thought that it had to put tremendous stress on the drive system and never tried it myself. Of course the worse thing is not telling customers about the counter BEFORE they used the feature. I hope they make good for people in the same boat as Tech_Guy.

 

[msnow]

Thank you for sharing this.

Some of the information that this tech provided can't be true. The variants he mentioned, P85DL, P90DL and P100DL have the same drive units, but not the battery packs. The drive units are rated for a combined 568kW (375kW rear and 193kW front) for 30 minutes, so they can't possibly be the limiting factor during launches because the output is less than 568kW, for a lot less time (15 seconds or so vs. 30 minutes).

The maximum output is limited by the battery pack and this maximum output is lower in P85DL than in P90DL. If limit is triggered only for perfect launch, i.e. maximum power in P90DL, it does not make sense that it would be triggered under any circumstances in P85DL as its power pack has lower overall output power.

It is hard to argue that some information provided by this tech is correct, but other is not as it has appearance of cherry picking, but IMO, based on technical considerations he is most likely correct that counter is triggered only when car is in "launch mode" and only for perfect launches (perfect traction, no power cut back due to traction control intervention). The other information can't be correct from the technical point of view.

All the tech said was there was a counter for all variants when perfect launches are achieved. He didn't say the 85 kWh battery would be limited.

 

[lolachampcar]

OMM,
Although the words have come from Tesla that the reductions are tied ONLY to launch mode, I fear greatly that the statement is pure BS. Logically, it is the battery that is the weak link. Logically, it is highly likely to be the individual cell fuse wire interconnects that are fatiguing under the higher currents (or perhaps cell anode or cathode damage) as these were two primary focuses of changes made for the current P100D battery (which Tesla tells us does not CURRENTLY have a limit). If the previous is accurate, cell fuse wires or even cells themselves have no interest in launch mode; they only "see" the current being drawn from them. This simple fact tells me it is way more likely that Tesla is integrating time above a given current threshold to determine when to call no joy.


for the rest
As for the highly technical arguments about spec'ing wiring in structures, I like to keep in mind that it is Tesla's job to design the system and Tesla's job to deliver the 691 hp they originally advertised way back when at the D announce. I could care less if they have squirrels on speed inside that box below my butt provided it delivers the spec for 100K miles. The idea that we cut them some slack because of house wiring is just plain off the reservation.

On another point, adding the MT car to any concept of V1, V2, or V3 batteries seems not to make sense given, again - to me, it is obvious that Tesla provided a "magazine car" to MT so that MT could publish specs Tesla wanted to see in print but not take corporate responsibility for advertising. We simply do not know what battery was in the MT car (100 prototype????); we just think it produced enough power to generate repeatable ten second runs (in the absence of any time slips to back those claims up).

I get quoting written responses from Tesla then parsing the words for hidden meaning or intent. I do not get even considering an utterance from ANY front end Tesla flunky. You may get lucky and garner a grain of truth but that will surely be by shear accident. I simply can not count the number of times I have spoken with anyone lower than the service manager level where I knew more about what is going on than they do (and I do not know very much).

Tesla hires young wide eyed eager kids for the same reason Savimbi put AKs in the hands of children. You can wind them up with any BS you want and point them towards a customer to deliver the party line without regard for truth or reason. They are simply too young and naive to care. The ones that figure it out either buy off on the approach for the return or bail. I've talked to several that have bailed.

I am sorry that the words are coming out so damming and bitter. I'm really not and continue to be blown away by Tesla's overall planning, direction and execution. We are seeing something truly great here and I am appreciative of being able to watch it play out. At the same time, the observations above are beyond obvious to me and I'm not one to ignore reality (as I see it).

 

[P85DEE]

Thank you for this update.

One thing to note here (which might also go some way explaining the discrepancy @vgrinshpun pondered about), assuming this report is accurate representation of the tech's words, I do not think there is a claim that all the models use the counter in the same manner.

We have heard P90DL activated the limit of death after the 8.0 software update and I think the concensus has been gravitating towards it doing this based on previously gathered data. Well, what if the counter itself is simply one of many parameters the car follows, but that does not automatically mean anything more? It is just data. It was being collected already on 7.1 (and previous?) and is being collected on any car with launch mode. Maybe it is called a "Launch Mode Counter" or something (though we assume it can get triggered by non-Launch Mode events as well, but that's a separate conversation) and sits next to some other counters like the "Left Seat Heater Turn On Counter" (made that up) that Tesla uses to monitor car conditions for statistical purposes.

Nothing wrong with a counter itself. It is possible Tesla is not using the counter for any performance penalties in P85DL or P100D. A benign use case for counters could be, say, Tesla's own warranty claims to contractors for example (contractor promising a part to last X number of launches). The tech said he didn't know the limitations, so he/she would not necessarily have known where it applies. It seems plausible, he/she may just know the counter because it can be seen in some analytical tool, but not its effects which may not be disclosed to the techs at all (maybe some internal memo was circulated with vague wordings similar to the CYA). He may not know exactly what triggers the counter addition either, beyond perhaps some small descriptive note in the analytical tool that may or may not be the full story.

Now, what is problematic is the precedence being created here. Now that we think we know the counter is global, and Tesla has done this once and added a global disclaimer to boot, who is to say it won't happen again later for other models? Even more so, because there probably are other counters as well and Tesla is known to update software all the time, who is to say they won't disable or diminish other features as well to protect themselves from warranty claims? That seat heater being problematic, why not start limiting its availability pre-emptively by software in the entire fleet... (again, just a made up example)

It certainly is a problematic precedent in that sense, but not necessarily yet indicative of being used beyond P90DL. Pending further data, of course.

As for @vgrinshpun cherry-picking, I personally didn't find that problematic. I am not sure his logic holds regarding the specific criticism levelled aginst the tech (I think my speculation above would be a plausible explanation), but certainly it is not automatically negative cherry-picking to discuss what parts of the tech's comment - or the report of the tech's comment which may be different - could be mistaken or misunderstood. It is certainly possible a tech could know one thing and be mistaken about another, and even more so when considering a later report of the tech's words could have misunderstood and misreported something - the broken telephone or lost in translation effects...

Now that is my kind of post. Well done sir.

Lengthy, verbose. Using ten words when one will do.

Seven paragraphs. One with nearly 150 words in it. It reads like War and Peace. It's almost like you're "there" when you read it.

Opinionated while remaining feasible.

Token punctuation, run on sentences like the one I'm creating now, and likely a misspelled word or two because it was most likely done on a cell phone where sentence structure can be difficult to follow as it is being written.

It will probably be edited too after reading it and seeing errors in it and an effort to drive its strong points home.

On a serious note though, you mentioned precedent. I appreciate your having done that.

Precedent is what concerns me most about this matter.

On a side note, someone asked if ICE vehicles employ a similar method.

Yes to a degree.

If you've done any ECM tuning, then you've more than likely come across things such as "torque management" and the like.

Methods used to cut power and or torque when certain operating parameters are met so as to save drive train parts and limit warranty exposure.

Many people who do their own tuning, turn these restrictions off. Or increase the power, torque RPM threshold, whichever apply, at which the restriction is triggered.

This is why, or part of why, many manufacturers will not honor a warranty claim in a car which has had its calibration altered or has been tuned.

But yes, manufacturers have historically taken a variety of steps to protect themselves against warranty claims. Even in performance vehicles marketed for aggressive driving.

This latest one by Tesla, where power appears to be cut permanently, and even when certain parameters aren't met, if the triggering event has already occurred, takes it to a new level.

 

[scaesare]

Thank you for sharing this.

Some of the information that this tech provided can't be true. The variants he mentioned, P85DL, P90DL and P100DL have the same drive units, but not the battery packs. The drive units are rated for a combined 568kW (375kW rear and 193kW front) for 30 minutes, so they can't possibly be the limiting factor during launches because the output is less than 568kW, for a lot less time (15 seconds or so vs. 30 minutes).

Long time thread-follower... first time caller...

I wonder about this. Although there is much discussion of this being a protection for the battery, I don't know that's been conclusively proven.

Even if drive components are spec'ed for a given power, it's torque that breaks parts.

Thus where on the torque curve that power is made should determine the max stress on the driveline hard parts. Given electric motors make max torque at 0, and max inertia also must be overcome at 0 RPM, I'd guess that parts that could live with 600HP all day long at 100MPH will suffer a lot more with 600HP launches...

 

[St Charles]

??

This display of rudeness dazzles, but adds little to your point. Totally unnecessary

Merely an observation with a colorful verb. Take it as you will.

No it is not.

To the extent that Teslas behavior is unreasonable. I am glad we agree on this.

I would add, however, that conditions under which this reduction is triggered could vary significantly (for example: 25, 125 or 1025 launches; whether any launches count, or only those done in "launch mode"; done under any conditions, including less than perfect traction, or only when traction is perfect, etc.). This IMO makes the difference in the real impact of this on owners. That is why I very much would like to get more clarity on this, as so far we have a lot of contradictory information on the subject.

I also think that at this stage generalization of hypotheticals on Tesla motives are not helpful.

I would counter that floating standards that may or may not apply is equally unhelpful. Especially since, as you point out, we do not really know what is happening. The core of this argument is ethics, not engineering. It is unethical to sell a feature and remove that feature without warning or refund.

 

[SDRick]

On a lighter note, how accurate are the results generated in Power Tools as it relates to this discussion?

 

[lolachampcar]

"If you've done any ECM tuning, then you've more than likely come across things such as "torque management" and the like."

"Many people who do their own tuning, turn these restrictions off. Or increase the power, torque RPM threshold, whichever apply, at which the restriction is triggered."

Modern ECUs model engine performance with most all functions revolving around "Torque Demand". The gas pedal is the user input for torque demand. The AC compressor kicking on injects an additive request or torque demand.

Torque limiting tables are similar in nature to Tesla's limiting of current ramp during acceleration from zero or current cap once the design current draw level has been reached. Further, component temperature current limiting in Teslas is comparable to torque limiting when inlet charge temperature rises in turbo or supercharged ICE.

These things are indeed normal.

My experience with OEM ecus ends as of about five years ago and, at that time, BMW, Porsche, MB, VAG and others did not have an integration function to amass area under the curve above certain power levels followed by a corresponding reduction in the torque limit table. The table was fixed at calibration and static.

In short, no, other OEMs do not do this as of my last direct experience.

 

[P85DEE]

"If you've done any ECM tuning, then you've more than likely come across things such as "torque management" and the like."

"Many people who do their own tuning, turn these restrictions off. Or increase the power, torque RPM threshold, whichever apply, at which the restriction is triggered."

Modern ECUs model engine performance with most all functions revolving around "Torque Demand". The gas pedal is the user input for torque demand. The AC compressor kicking on injects an additive request or torque demand.

Torque limiting tables are similar in nature to Tesla's limiting of current ramp during acceleration from zero or current cap once the design current draw level has been reached. Further, component temperature current limiting in Teslas is comparable to torque limiting when inlet charge temperature rises in turbo or supercharged ICE.

These things are indeed normal.

My experience with OEM ecus ends as of about five years ago and, at that time, BMW, Porsche, MB, VAG and others did not have an integration function to amass area under the curve above certain power levels followed by a corresponding reduction in the torque limit table. The table was fixed at calibration and static.

In short, no, other OEMs do not do this as of my last direct experience.

My experiences with ECMs ends as of about 1 year to 1.5 yrs ago, and are different from yours, particularly in vehicles equipped with automatic transmissions, having used HP Tuners and tuning my own GM vehicles using it.

Factory A/F ratios, (a car can be made to run leaner or richer than the factory setting and at various points) timing and when in operation to cut or advance it, torque values etc. can be, and are altered from stock using that programming software.

Effectively you can increase to a point, or decrease, area under the curve depending on which parameters you change and where you change them.

Automatic transmission shift points, torque converter lock up, and line pressures affecting how "hard" a 1-2, 2-3, etc. shift is executed, can be altered from factory stock as well.

All of the above parameters are set at a factory range to protect susceptible components from damage and to limit warranty repairs.

If you talk to people who've tuned Duramax diesels, they'll tell you the same thing.

I have been following this thread out of curiosity but not read every comment so this may have been addressed.

I was wondering if something similar is employed by other manufacturers (especially supercars) in their ICE/hybrid cars?..

When he says "something similar", I'm taking this to mean any similar effort to minimize exposure to warranty claims arising out of parts failure from aggressive driving.

 

[St Charles]

My experiences with ECMs ends as of about 1 year to 1.5 yrs ago, are different from yours, particularly in vehicles equipped with automatic transmissions, having used HP Tuners and tuning my own GM vehicles using it.

Factory A/F ratios, (a car can be made to run leaner or richer than the factory setting and at various points) timing and when in operation to cut or advance it, torque values etc. can be, and are altered from stock using that programming software.

Automatic transmission shift points, torque converter lock up, and line pressures can be altered from factory stock as well.

All of the above parameters are set at a factory range to protect susceptible components from damage and to limit warranty repairs.

If you talk to people who've tuned Duramax diesels, they'll tell you the same thing.

Sure, all of these parameters can be messed with. But I think the point that @lolachampcar brings up is that no other car ECU uses driving habits as the sole metric to permanently alter power output. That said, The 2015+ Ford Mustang had a built in Line lock capability that was really a creative use of their ABS system. I had heard rumors that Ford kept track of how many times this was engaged and people at the time were worried that warranty coverage could be denied if it was ever used:

http://jalopnik.com/use-the-2015-mustangs-burnout-control-to-race-lose-you-1567209420

 

[P85DEE]

Sure, all of these parameters can be messed with. But I think the point that @lolachampcar brings up is that no other car ECU uses driving habits as the sole metric to permanently alter power output. That said, The 2015+ Ford Mustang had a built in Line lock capability that was really a creative use of their ABS system. I had heard rumors that Ford kept track of how many times this was engaged and people at the time were worried that warranty coverage could be denied if it was ever used:

http://jalopnik.com/use-the-2015-mustangs-burnout-control-to-race-lose-you-1567209420

See my above post.

When the inquiry was "something similar", I took this to mean any steps taken by an auto manufacturer in an effort to protect drive train parts and limit the manufacturer's exposure to warranty repair.

If we're talking "driving habits", and using those habits as an "active means" in an effort to limit warranty exposure, well then no. No one else has done this. This seems to be a first. It's probably unprecedented, and this is why it stinks to high heaven.

However if we are talking factory parameters, well then yes, manufacturers set the stock calibration in such a manner as to protect drive line parts and thus limit their exposure to covering warranty repairs.

That stock calibration can be altered to unleash more performance capability out of an otherwise stock car.

However if the manufacturer discovers that this has been done, then it often times will affect whether or not they honor a warranty claim.

That was my point.

If by "similar" he meant throttling someone's car based on their driving habits, then no, there is nothing that I know of which is "similar" to that.

However if by "similar" he meant steps taken by a manufacturer to limit their exposure to a warranty repair, well then yes, to an extent. As manufacturers have parameters in place which can be exceeded, but are in place in order to protect driveline parts.

 

[apacheguy]

They are signing all code with a highly secure ECDSA system. The gateway verifies each systems firmware file using this before it will flash the target system. It is still possible to flash a target module with physical access, but it is no longer possible to remotely load unsigned firmware in any system other than the CID and IC itself. The gateway also checks the signatures on any code for itself, so it is no longer possible to roll back to version before the signing started (2.36.31). That was the old-style 7.1 GUI, so at least we have one version of 7.1 that can be run if desired.

Root the gateway and you can remove code signing restrictions.

 

[jbcarioca]

Back in 1989 I had a Nissan 300ZX Turbo. I installed an ECU modification plus raised the boost by a large margin. On the dyno the HP went up by about 30% IIRC. Anyway, Nissan got wind of my mod, the first one made by the vendor, I had SN 901, one fo the first delivered in the US. I received a call from Nissan USA telling me nothing I had done would void the warranty. Probably they'd not do that today.

I could easily accept that Tesla might warn customers in some way about possible damage from too many Launch mode uses. However, it is a standard feature. Not many people use it much at all. I think they should ignore it and eat the increased warranty costs, if any. If they thought it would cause harm they should not publicly issue it, or treat it just as a handful of manufacturers have done when selling aftermarket racing mods, sell it, stating that it's use voids the vehicle warranty and that the mod in question is not intended for use on public roads. tesla wants the PR, they should pay the price, not the users, unless they disclose up front.

I know I am repeating things already said.

 

[P85DEE]

.... I think they should ignore it and eat the increased warranty costs, if any.

I agree and would go even a step further.

Not only should they eat the warranty costs should any occur as a result of use of the launch mode, but they should forget this entire method and concentrate on beefing up susceptible parts if the car is too powerful for the existing ones.

....If they thought it would cause harm they should not publicly issue it, or treat it just as a handful of manufacturers have done when selling aftermarket racing mods, sell it, stating that it's use voids the vehicle warranty and that the mod in question is not intended for use on public roads. tesla wants the PR, they should pay the price, not the users, unless they disclose up front.

That's a reasonable idea. In fact it's an excellent idea.

This way people have a choice as to whether or not they want to jeopardize their current warranty coverage by adding a feature, or if they want to forgo it and keep their warranty intact.

I decide I don't want launch mode, my warranty states in tact and you don't go messing with my power.

I decide I do what launch mode, then I get it at my own risk.

However considering that it doesn't seem to do much, one way around this would be to simply not use it even if they gave it to you.

But it might not just be the use of launch mode, but any hard launch, which triggers and starts the cascade for the loss of power.

 

[stopcrazypp]

OMM,
Although the words have come from Tesla that the reductions are tied ONLY to launch mode, I fear greatly that the statement is pure BS. Logically, it is the battery that is the weak link. Logically, it is highly likely to be the individual cell fuse wire interconnects that are fatiguing under the higher currents (or perhaps cell anode or cathode damage) as these were two primary focuses of changes made for the current P100D battery (which Tesla tells us does not CURRENTLY have a limit). If the previous is accurate, cell fuse wires or even cells themselves have no interest in launch mode; they only "see" the current being drawn from them. This simple fact tells me it is way more likely that Tesla is integrating time above a given current threshold to determine when to call no joy.

I suspect that too, but so far we have two Tesla personnel (one from corporate and one local) which have said this change was related to drivetrain wear from launch control and that it only counts launch control. The main point against that is that if this was about drivetrain wear, then why does the power limit seem to apply universally and not just because of launch control or the short period during launch (at least from the single datapoint that Techguy is providing).

I should point out that I thought of one way for a universal power limit to be consistent would be related to motor winding longevity (I can't find the thread anymore but this was discussed during the Roadster days and the power reduction temperature limits).

A 10C rise in temperature can reduce insulation life by half.
When it comes to motors, how hot is hot? | Plant Engineering

It could be that launch control disproportionately increases motor temperature and the counter is a generic motor life expectancy counter that reduces power so that it can reach the specified warrantied life (except it acts globally instead of temporarily). Like the temporary power reduction, it may be overly strict about the amount of reduction applied.