Performance Model 3 Drag Race - 2021 vs 2018 - Is it Faster?

[Marius A]

my battery has lost 13% and it is still within 0.01 seconds on a 0-60 and 1/8 mile time. So I’m not sure if a higher % loss makes a difference, but mine has not, at all.

Normal battery degradation doesn't affect performance. Performance is a combination of peak voltage, nominal voltage and output current. Normal battery degradation reduces the charge capacity for a battery and after a while peak voltage, but not the nominal voltage and current output.
When enough degradation has happened maximum current output will be affected, and finally nominal voltage.
According to the subsystem specification the nominal voltage is 320V and max currents are:
980 motor (RWD, Performance, Stealth) - 800A, max power 202kW
990 motor ((most) LR) - 600A, max power 180kW
Front motor - 600A, max power 137kW

Of course, they don't output all of that at the same time. The max effect is at a certain RPM, which is different for all 3 motor types, but just to give an example we could say max simultaneous power of 320kW on the Performance (it isn't) makes 320V nominal voltage and a 1000A current (which isn't correct).

However, that's only theoretical numbers from the subsystem specification, the battery specification (according to unconfirmed sources I've read, so grain of salt, et.al) says 350V nominal voltage and peak output current at 1200A, a theoretical nominal power that would amount to 420kW. So given these nominal values you'd be able to max out the motor until the battery has degraded to ~24% of it's maximum nominal output.

Of course, this is only a simplification I made, in reality you the peak voltage (400V) to achieve the max power of the motors, which requires a high state of charge and healthy battery modules. However, unless you're using the car for drag-racing all the time then your TM3P/stealth will likely maintain it's performance for several years before you start seeing a occasional drops in 0-60 times. Especially considering that the battery consists of several modules which can be replaced for a reasonably-ish amount.

 

[Knightshade]

According to the subsystem specification the nominal voltage is 320V and max currents are:
980 motor (RWD, Performance, Stealth) - 800A, max power 202kW

Also all LR AWD made in 2018, and at least some made first half-ish of 2019 (these are the ones eligible for the ghost upgrade from a 3rd party that gives it P performance)

Likewise it's all LR RWD, but only some SR RWD as there's been reports those now come with 990s instead (no idea when a cutover date would've been, or if it's specific to one factory or the other- just seen owners report having a 990)

 

[MasterC17]

i hate to tell you, but you did not prove it. There are always differences between cars, even though they are in theory the same. .1 is such a small difference.

And as said above, Dragy leaderboard might say something when the temperature goes up, not now.

So interesting data, but there is more to explore.

The Dragy leaderboard is not going to change in any notable way when the temperatures increase. I already explained the reason why - that's not how the drivetrain is designed/works for peak power. I'm confident in my data. As of now, it's up to someone else to prove me wrong.

And an EV is not like an ICE, there is really no difference between cars because the power output is all electrically controlled. If Tesla wanted too, they could probably roll out an update tomorrow that increased power 20% at the expense of more warranty failures.

 

[Marius A]

Also all LR AWD made in 2018, and at least some made first half-ish of 2019 (these are the ones eligible for the ghost upgrade from a 3rd party that gives it P performance)

Yeah, that's why I put :

990 motor ((most) LR) - 600A, max power 180kW

Likewise it's all LR RWD, but only some SR RWD as there's been reports those now come with 990s instead (no idea when a cutover date would've been, or if it's specific to one factory or the other- just seen owners report having a 990)

I didn't know that. My 2 cents is that this means older LR and SR models will never be fully unbinned. It would make owners of new cars very upset.

Sidenote: If you have a binned LR with a 980 motor, is that a Tesla Model 3 Sleeper Stealth? Or maybe a Model 3 Aurora?

 

[TwoK4drSi]

Yeah, that's why I put :



I didn't know that. My 2 cents is that this means older LR and SR models will never be fully unbinned. It would make owners of new cars very upset.

Sidenote: If you have a binned LR with a 980 motor, is that a Tesla Model 3 Sleeper Stealth? Or maybe a Model 3 Aurora?

Please don't say "binned" at all anymore. That's like saying unicorns and leprechauns....it doesn't exist. The 980 motor has 6 more mosfets than the 990. PERIOD. That's why it more powerful.

Elon has and always will lie to us to make a sale. That's business.

 

[Marius A]

Please don't say "binned" at all anymore. That's like saying unicorns and leprechauns....it doesn't exist. The 980 motor has 6 more mosfets than the 990. PERIOD. That's why it more powerful.

Elon has and always will lie to us to make a sale. That's business.

Binning is a term that has been used for many years, for example by PC hardware makers.
One example of binning is when a component cannot meet the quality or performance it was meant for, it will be limited and sold as a lower end component. A real life example would be an Intel CPU sold as an i5 because it can't keep the clock values of an i7
Another example of binning is when the demand for a higher-end component is low but the demand for a lower-end component is high, so the producer limits the high-end component and sells it like a lower-end. A real life example would be a Mercedes 280 engine which was the same as a 320 engine, but it was software limited.

I don't think Elon is a shiny examples of truth and openness, but saying he's lying because of this is silly. It's a common practice in many industries. AFAIK he has never promised that you'll be able to turn your LR AWD into a Performance.

Also, based on his interaction with the world it looks like he's much more interested in rockets than which motor is fitted to which VIN-number. If you wanna blame someone then blame the TMC members who buys TSLA stocks.

 

[Sam1]

Normal battery degradation doesn't affect performance. Performance is a combination of peak voltage, nominal voltage and output current. Normal battery degradation reduces the charge capacity for a battery and after a while peak voltage, but not the nominal voltage and current output.
When enough degradation has happened maximum current output will be affected, and finally nominal voltage.
According to the subsystem specification the nominal voltage is 320V and max currents are:
980 motor (RWD, Performance, Stealth) - 800A, max power 202kW
990 motor ((most) LR) - 600A, max power 180kW
Front motor - 600A, max power 137kW

Of course, they don't output all of that at the same time. The max effect is at a certain RPM, which is different for all 3 motor types, but just to give an example we could say max simultaneous power of 320kW on the Performance (it isn't) makes 320V nominal voltage and a 1000A current (which isn't correct).

However, that's only theoretical numbers from the subsystem specification, the battery specification (according to unconfirmed sources I've read, so grain of salt, et.al) says 350V nominal voltage and peak output current at 1200A, a theoretical nominal power that would amount to 420kW. So given these nominal values you'd be able to max out the motor until the battery has degraded to ~24% of it's maximum nominal output.

Of course, this is only a simplification I made, in reality you the peak voltage (400V) to achieve the max power of the motors, which requires a high state of charge and healthy battery modules. However, unless you're using the car for drag-racing all the time then your TM3P/stealth will likely maintain it's performance for several years before you start seeing a occasional drops in 0-60 times. Especially considering that the battery consists of several modules which can be replaced for a reasonably-ish amount.

Would think it would affect it somewhat though during high draw events. Doing some of my top end runs at 95% SOC, I've seen my voltage dip below 300v once you start getting over 140mph or so. More available kwh would not stabilize that a bit more?

 

[SomeJoe7777]

Please don't say "binned" at all anymore. That's like saying unicorns and leprechauns....it doesn't exist. The 980 motor has 6 more mosfets than the 990. PERIOD. That's why it more powerful.

I agree with Marius above ... separating identically-built parts into higher and lower performance categories is a real thing, and has been done for years by manufacturers.

This is the primary way that Intel makes CPUs. They take 100 CPUs out of a wafer, and test them at higher and higher clock speeds. Each chip is then sold based on the highest clock speed it can sustain without any errors. Out of the 100 CPUs, many can run at 2.8 GHz, some can't, but can still be sold if they are run at 2.2 GHz, and others for whatever reason are perfectly fine running at 3.4 GHz, so you sell them at a premium.

I see no reason why Tesla wouldn't do this as well.

Take 100 Model 980 motors and put them on a test apparatus. Run maximum power through them (800A / 202kW) several times in a row with identical timings for the pulses and rests. Now measure the inverter temperature at the hottest junction. Over the 100 motors, you get a range of temperatures. All of them cluster around a median, but there are some that are lower and some that are higher.

You take the lower temperature ones and put them aside to go in Performance vehicles, and take all the others and put them in non-performance vehicles. That gives the performance vehicles the greatest margin to failure. And you don't have to worry about the others because the software on the non-performance vehicles limits the current anyway, thus limiting the temperature.

This is a great way to minimize warranty issues without having to engineer a beefier motor.

 

[Knightshade]

Binning is a term that has been used for many years, for example by PC hardware makers.
One example of binning is when a component cannot meet the quality or performance it was meant for, it will be limited and sold as a lower end component. A real life example would be an Intel CPU sold as an i5 because it can't keep the clock values of an i7
Another example of binning is when the demand for a higher-end component is low but the demand for a lower-end component is high, so the producer limits the high-end component and sells it like a lower-end. A real life example would be a Mercedes 280 engine which was the same as a 320 engine, but it was software limited.

Yes- but when chip makers do that, they give them different part numbers

That it literally the reason they bin

So they can stamp a higher end PN on the part that tests better, and a lower end PN on the one that does not.


The fact Tesla slapped the SAME PN on every 980 (and the 990 is a physically different part) is yet more evidence there was never any "binning" going on.

Go to Tesla, say you need a replacement 980.

They don't even need your VIN to give you one.

That'd be impossible if Ps got special magic binned ones and other cars did not.

 

[TwoK4drSi]

Binning is a term that has been used for many years, for example by PC hardware makers.
One example of binning is when a component cannot meet the quality or performance it was meant for, it will be limited and sold as a lower end component. A real life example would be an Intel CPU sold as an i5 because it can't keep the clock values of an i7
Another example of binning is when the demand for a higher-end component is low but the demand for a lower-end component is high, so the producer limits the high-end component and sells it like a lower-end. A real life example would be a Mercedes 280 engine which was the same as a 320 engine, but it was software limited.

I don't think Elon is a shiny examples of truth and openness, but saying he's lying because of this is silly. It's a common practice in many industries. AFAIK he has never promised that you'll be able to turn your LR AWD into a Performance.

Also, based on his interaction with the world it looks like he's much more interested in rockets than which motor is fitted to which VIN-number. If you wanna blame someone then blame the TMC members who buys TSLA stocks.

You’re skirting the topic and truth on this subject matter. It’s is 100% a lie. There are no binned motors. Stop talking about CPUs and stick to actual part numbers and physical parts.

https://www.reddit.com/r/teslamotor..._source=amp&utm_medium=&utm_content=post_body

 

[TwoK4drSi]

Also start learning from someone that has physically cracked open a 990 motor, swapped inverters, and put it back together and MADE a P car

AGAIN...there are no binned motors. The 990 has 6 less mofsets hence lower power. Ingenext has PROVEN this, cracked the code and made ghost. How do I know? Bc I’m a 980 LRAWD and I have ghost and it’s EXACTLY identical to P. So keep dreaming of a world where Elon has spoken to us truthfully.

 

[Knightshade]

Not to defend Elon.... but if you must, here's the defense that'd actually fit the facts.

Check the date on his tweet.

It's months before the first P (or LR AWD) was ever actually delivered to a customer.

It's 100% possible it was intended as true when he wrote it- but turned out in their testing 100% of motors they tested were good for P specs, so they all got stamped 980 and they stopped bothering to test by the time they were in full production of the P/LR AWD.

Hell it might be that's WHY they began looking into creating the 990...realizing the 980 was overbuilt and they could save a few bucks on the non-P cars making a cheaper and less capable unit.

 

[SomeJoe7777]

AGAIN...there are no binned motors. The 990 has 6 less mofsets hence lower power. Ingenext has PROVEN this, cracked the code and made ghost. How do I know? Bc I’m a 980 LRAWD and I have ghost and it’s EXACTLY identical to P. So keep dreaming of a world where Elon has spoken to us truthfully.

I disagree. Yes, the 980 motors in the LR RWD and the 980 motors in the AWD Performance are indeed physically identical, with the exact same parts, same inverter, same number of switching elements. That does not mean that any two 980 motors that you pick out of the field will perform identically. There are manufacturing tolerances in every component.

Just because all of the 980 motors are physically identical is not evidence or proof that there was no "binning". In fact, the very definition of "binning" is a process that is applied to identically-built parts. It would be expected that all 980 motors are physically identical.

 

[TwoK4drSi]

Not to defend Elon.... but if you must, here's the defense that'd actually fit the facts.

Check the date on his tweet.

It's months before the first P (or LR AWD) was ever actually delivered to a customer.

It's 100% possible it was intended as true when he wrote it- but turned out in their testing 100% of motors they tested were good for P specs, so they all got stamped 980 and they stopped bothering to test by the time they were in full production of the P/LR AWD.

Hell it might be that's WHY they began looking into creating the 990...realizing the 980 was overbuilt and they could save a few bucks on the non-P cars making a cheaper and less capable unit.

I believe the latter. 6 mosfets saved can go into the next motor BUT performance motors had to have been in production at the time of tweet. Switching on or off lights, heated seats, power, etc would make production super streamlined and faster BUT not cost effective.

Trying to stay on topic....no such thing as binned Tesla motors. PERIOD. 100000000% proven. Don’t talk about CPUs

 

[Sam1]

At least you got to 1.5 pages before the “have to be right online” kiddos popped in @MasterC17

 

[Devils son]

The Dragy leaderboard is not going to change in any notable way when the temperatures increase. I already explained the reason why - that's not how the drivetrain is designed/works for peak power. I'm confident in my data. As of now, it's up to someone else to prove me wrong.

And an EV is not like an ICE, there is really no difference between cars because the power output is all electrically controlled. If Tesla wanted too, they could probably roll out an update tomorrow that increased power 20% at the expense of more warranty failures.

First of all: in the video at the end, you are saying that there are small differences between cars. And now you say that they are all the same. Uh-huh.

As far as Dragy leaderboard goes: there is a sweetspot in temperature to get the best 0-60 time. Not too hot, not too cold.

So, interesting data, but nothing is proven as of yet.

Basically, what you are saying is that Tesla is lying about the 2021 versions being quicker than the 2020 versions. I would assume that a lawyer would actually prevent them from lying about that.

 

[MasterC17]

First of all: in the video at the end, you are saying that there are small differences between cars. And now you say that they are all the same. Uh-huh.

As far as Dragy leaderboard goes: there is a sweetspot in temperature to get the best 0-60 time. Not too hot, not too cold.

So, interesting data, but nothing is proven as of yet.

Basically, what you are saying is that Tesla is lying about the 2021 versions being quicker than the 2020 versions. I would assume that a lawyer would actually prevent them from lying about that.

I'm really not looking to argue with you. I'll leave the book open and say that there is a possibility the 2021 may be quicker due to less drivetrain loss or something that I can't account for. That being said, the power draw between the cars is effectively the same and I have provided hard data to support that statement. If someone finds different, I'm all ears.

Regardless, Tesla has changed the 0-60 numbers on the configurator more times than I can count. They also use the 1-ft rollout ONLY on the Performance models and not on the non-Performance (is that "lying"). When the 5% power boost was released in late 2019 the configurator was not changed. After that update, people were getting a 3.1 second 0-60 (with 1ft rollout). The update for 2021 from 3.2 to 3.1 seconds was likely to reflect that 5% power boost. The 2018 was originally advertised with a 3.5 second 0-60. But now a 2018, 2019, and 2020 all have similar 0-60 times and the same power draw.

They also changed the 0-60 time for the LR AWD models. It was originally 4.5 seconds, then 4.4, and now 4.2.

They also changed the $2,000 "Power Boost" upgrade on the LR AWD from 3.9 seconds to 3.7 seconds for all model years, but again tests have shown no power increase. Tesla Improves Model 3 LR, $2K Acceleration Boost Now Hits 60 in 3.7s

2021 Model 3 Acceleration Boost Videos?

Tesla isn't really "lying", they are picking and choosing when to update the numbers because ultimately no one cares if the car is faster than advertised only if it is slower. I am guessing they have found it beneficial to change these numbers to incentivize more sales, despite the car not being faster in reality. The flip-side of that coin is that so far all owners have benefitted from more power via Software.

I believe @Knightshade has the data to back this all up.

 

[Devils son]

I'm really not looking to argue with you. I'll leave the book open and say that there is a possibility the 2021 may be quicker due to less drivetrain loss or something that I can't account for. That being said, the power draw between the cars is effectively the same and I have provided hard data to support that statement. If someone finds different, I'm all ears.

Regardless, Tesla has changed the 0-60 numbers on the configurator more times than I can count. They also use the 1-ft rollout ONLY on the Performance models and not on the non-Performance (is that "lying"). When the 5% power boost was released in late 2019 the configurator was not changed. After that update, people were getting a 3.1 second 0-60 (with 1ft rollout). The update for 2021 from 3.2 to 3.1 seconds was likely to reflect that 5% power boost. The 2018 was originally advertised with a 3.5 second 0-60. But now a 2018, 2019, and 2020 all have similar 0-60 times and the same power draw.

They also changed the 0-60 time for the LR AWD models. It was originally 4.5 seconds, then 4.4, and now 4.2.

They also changed the $2,000 "Power Boost" upgrade on the LR AWD from 3.9 seconds to 3.7 seconds for all model years, but again tests have shown no power increase. Tesla Improves Model 3 LR, $2K Acceleration Boost Now Hits 60 in 3.7s

2021 Model 3 Acceleration Boost Videos?

Tesla isn't really "lying", they are picking and choosing when to update the numbers because ultimately no one cares if the car is faster than advertised only if it is slower. I am guessing they have found it beneficial to change these numbers to incentivize more sales, despite the car not being faster in reality. The flip-side of that coin is that so far all owners have benefitted from more power via Software.

I believe @Knightshade has the data to back this all up.

Let me just add one thing: the 2021 has a bigger battery. If it didn’t get quicker, it would be the first Tesla not to gain performance after that.

 

[Knightshade]

I'm really not looking to argue with you. I'll leave the book open and say that there is a possibility the 2021 may be quicker due to less drivetrain loss or something that I can't account for. That being said, the power draw between the cars is effectively the same and I have provided hard data to support that statement. If someone finds different, I'm all ears.

Regardless, Tesla has changed the 0-60 numbers on the configurator more times than I can count. They also use the 1-ft rollout ONLY on the Performance models and not on the non-Performance (is that "lying"). When the 5% power boost was released in late 2019 the configurator was not changed. After that update, people were getting a 3.1 second 0-60 (with 1ft rollout). The update for 2021 from 3.2 to 3.1 seconds was likely to reflect that 5% power boost. The 2018 was originally advertised with a 3.5 second 0-60. But now a 2018, 2019, and 2020 all have similar 0-60 times and the same power draw.

They also changed the 0-60 time for the LR AWD models. It was originally 4.5 seconds, then 4.4, and now 4.2.

They also changed the $2,000 "Power Boost" upgrade on the LR AWD from 3.9 seconds to 3.7 seconds for all model years, but again tests have shown no power increase. Tesla Improves Model 3 LR, $2K Acceleration Boost Now Hits 60 in 3.7s

2021 Model 3 Acceleration Boost Videos?

Tesla isn't really "lying", they are picking and choosing when to update the numbers because ultimately no one cares if the car is faster than advertised only if it is slower. I am guessing they have found it beneficial to change these numbers to incentivize more sales, despite the car not being faster in reality. The flip-side of that coin is that so far all owners have benefitted from more power via Software.

I believe @Knightshade has the data to back this all up.

Yup.

Basically as long as it's at least a fast as Tesla claims, no lawyers are going to be involved in anything.

see attached for the various LR AWD and P times from the various SW updates.

Basically, what you are saying is that Tesla is lying about the 2021 versions being quicker than the 2020 versions. I would assume that a lawyer would actually prevent them from lying about that.

Nope.

They're just really bad about posting accurate times- but in a way that's totally legal.



See above.

The LR AWD was listed at 4.4 when it was doing 4.0 flat if you measured it the same way they measured the P on the website listed time.

Today it does 3.9 measured the same way but they list it at 4.2.

Nothing about the car changed, they just put a less sandbagged time on the website.


Same with the P. Originally they were, oddly for tesla, actually honest about the P in that they listed it measured the same way as the non-P (this was back mid 2018)... the original LR AWD was listed at 4.5, and the P at 3.5... both were pretty close to the actual real world no rollout times.

Then someone at Tesla noticed they weren't cheating on the 3P the way they always have on the S/X and "fixed" it.

The P suddenly dropped to 3.3.

Nothing about the car changed, they simply posted the rollout time now (only for the P).

Then bumped it to 3.2 after the first 5% update.

After the 2nd 5% update they didn't change the website (for any 3).

NOW they have.

Again the car isn't any faster this week than last, or this year vs last.

They just finally updated the public #s.

Which are least equal to or slower than the real numbers, so no lawsuits.

Let me just add one thing: the 2021 has a bigger battery. If it didn’t get quicker, it would be the first Tesla not to gain performance after that.

It does not have a bigger battery. That would require a physical design change.

It has a very slightly denser one though. And those really are legitimately different things for a bunch of reasons.

 

[TwoK4drSi]

Let me just add one thing: the 2021 has a bigger battery. If it didn’t get quicker, it would be the first Tesla not to gain performance after that.

Also one thing to note....don’t get confused on “bigger battery” equal more power bc it doesn’t. It’s the “current” to the DUs the software allows that equates to more power. Bigger / denser / Efficient batteries is only used power storage.