Tesla BEV Competition Developments

[ammulder]

It's called "Highway Teammate", so I'll go out on a limb and predict it will be geofenced for highways....

Well, Autopilot is highways-only, officially... yet you can use it virtually anywhere (anywhere with road markings, perhaps?). I haven't driven a Cadillac, but my impression is that you can't even activate SuperCruise outside the premapped (highway) areas. Part of the difference being that I assume SuperCruise relies on that pre-generated high-resolution mapping data. It makes a big difference whether this new implementation relies on pre-mapped data like that, or even if not whether they put location restrictions in place for safety/reliability.

OTA is more of a marketing issue, e.g. do their dealer agreements allow it, do their customers want it, etc.

I would have said OTA was more of a technology issue than a marketing issue. Can they get it working any better than VW and their parking lots full of unsold inventory and USB drives for software updates? Will the scope be "anything in the car" more like Tesla, or only self-driving logic (perhaps also infotainment)? No matter what the capabilities are, can they effectively switch to a model of e.g. monthly updates instead of annual model-year updates? I think that would require a huge change in software development practices, so it's by no means a given even if the OTA capabilities are there. Are they willing to ditch the model of "want a new feature, buy a new car!" to drive new car sales?

By way of comparison, note how iPhones get regular updates, but often Android handsets get very infrequent updates from manufacturers/carriers even though Google publishes regular updates to the Android core. Just because it's possible doesn't mean the OEMs will do it or do it well.

It would be interesting to measure the number of take-it-to-a-dealer recalls before and after OTA updates are introduced for each OEM, as one measurement of the scope of their OTA implementation.

 

[colettimj]

Well, Autopilot is highways-only, officially... yet you can use it virtually anywhere (anywhere with road markings, perhaps?). I haven't driven a Cadillac, but my impression is that you can't even activate SuperCruise outside the premapped (highway) areas. Part of the difference being that I assume SuperCruise relies on that pre-generated high-resolution mapping data. It makes a big difference whether this new implementation relies on pre-mapped data like that, or even if not whether they put location restrictions in place for safety/reliability.
[snip]

It will be interesting to see if this section of I95 in Philly (my highlighting in yellow) remains off the Supercruise map while construction on the roadway is underway. The construction won't be completed until 2023 or 2025.

As good as SC is / can be, it would stink if it was unavailable for years on roadways that are under construction. I have been using AP through that area since v9 was released.

 

[RobStark]

Wish we in the States could get the Skoda Enyaq instead of the VW ID.4

IMO it would sell much better.

 

[Fact Checking]

Interesting way to avoid Osborning. Don’t announce/unlock improvements until several months after they have been implemented.

Yeah, I've been grinning about this all morning already: beyond avoiding the Osborning factor Tesla also found the right way to avoid hurting recent buyers like @Cherry Wine, by phasing in the new hardware in a stealth fashion, software-locking the range to the old one as if nothing was going on, and then increasing the range in one big step when the opportunity was given to smooth over seasonal demand weakness.

Feedback on Reddit and other social media is overwhelmingly positive, with none of the usual "I just bought my Model S and then Tesla does this!! " posts whatsoever.

Elon also announced that the Plaid will have the same range, i.e. apparently there's no big reason to wait for Plaid unless you want to spend $120k+ on a racing car ...

Whoever figured out this particular trick to use software locking and OTA updates in such a creative fashion deserves a big bonus.

BTW., I'll just list a couple of range comparisons:

• E-Tron EPA range (2019): 204 miles
• Model X EPA range: 351 miles, +72% higher

Or:

• Taycan 4S EPA range (2020): 201 miles
• Model S EPA range: 390 miles, +94% higher

Not only are legacy OEMs not catching up with Tesla, they are falling further behind.

 

[egn1]

• Taycan 4S EPA range (2020): 201 miles
• Model S EPA range: 390 miles, +94% higher

On the other side the Taycan has at 150 km/h a similar range as the Model 3 LR RWD.

There is still a large difference between EPA and reality.

Charging Speed of Taycan is also much better.

Here in there is a considerable drop of charging speed during winter, not only because of charge gate.

I can only shake my head when someone says that Tesla is years ahead of the others. Tesla has been a bit lazy the last few years. With NMC the others have the better battery technology with higher cycle numbers and better charging speed.

So I hope Tesla is now pushing the limits to go ahead again.

 

[Compton]

On the other side the Taycan has at 150 km/h a similar range as the Model 3 LR RWD.

View attachment 511425

There is still a large difference between EPA and reality.

Charging Speed of Taycan is also much better.

View attachment 511426

Here in there is a considerable drop of charging speed during winter, not only because of charge gate.

I can only shake my head when someone says that Tesla is years ahead of the others. Tesla has been a bit lazy the last few years. With NMC the others have the better battery technology with higher cycle numbers and better charging speed.

So I hope Tesla is now pushing the limits to go ahead again.

So about three times more expensive car with a 25% larger battery has a worse range? But Tesla has been lazy? That's amazing.

 

[Fact Checking]

On the other side the Taycan has at 150 km/h a similar range as the Model 3 LR RWD.

The fact that the Taycan cannot exceed even the Model 3's range in the Taycan's carefully maintained 150 km/h sweet spot (let alone the Model S) is a very poor showing for the Taycan.

The big mistake of the EPA test cycle, which tanked the Taycan's range to 193-201 miles, is that it implements the horrible city driving pattern of cars frequently accelerating and braking, which driving pattern I've never seen from a single Porsche driver: they are all maintaining a smooth, constant, disciplined speed, right in accordance with the benchmarketing recommendations of Porsche PR.

There is still a large difference between EPA Porsche claims and reality.

FTFY.

Come on Porsche, ask your Volkswagen pals how to do this, they are the experts on how to cheat on government tests. I'm sure if you work hard enough you'll be able to claim twice the range of Tesla. Install sails perhaps?

I can only shake my head when someone says that Tesla is years ahead of the others.

LOL, today I learned that "egn1" is a new bird species.

 

[electracity]

So about three times more expensive car with a 25% larger battery has a worse range? But Tesla has been lazy? That's amazing.

After all the disappointing "Tesla Killer" car releases I'm just pleased that Porsche made a car that some high end buyers might prefer over an X or S.

Tesla completely dominating the EV space is not a good thing, perhaps even for TSLA investors.

 

[KarenRei]

Charging Speed of Taycan is also much better.

View attachment 511426

Here in there is a considerable drop of charging speed during winter, not only because of charge gate.

LOL. I don't know whether they're deliberately trying to be misleading or whether it was accidental.

First off, you don't need to charge to "X %". You need to charge to "X km". Thus all comparisons should be "How long does it take to charge from X km to Ykm". Anything else is a nonsensical rigged comparison that has no applicability to the real world.

In the below graphs, Teslas are red (Model 3 variants = bright red), Taycan (4x variants) bright gold.

Gee, nope, I can't see any sort of difference, no sirree....

(The above graphs include an initial connection and rampup delay. The shape of the charging curves on the "other" EVs which haven't been released yet is assumed, based on stated charge powers, charge rates, and compared to pack sizes. Existing vehicles are based on filmed or published charging curves. For example, Model 3 on V3 is based on this, and Taycan is based on this.)

Your video is rigging the comparison - deliberately or not, you decide - by forcing the Model 3 to go to high SOCs where it has to taper significantly.... except the Model 3 at high SOCs is more range than the Taycan has total. And then they try to distort it even more by relying on an unrealistically high speed for the vast majority of drivers, to reduce the Taycan's appalling efficiency at lower speeds. It's a face-palmingly bad graph.

And BTW, this is assuming that you can actually get the Taycan to charge at high rates. I've seen report after report of people having trouble with the Taycan at fast chargers. There was even a recent review-road trip where repeated troubles at every charger tried left the driver stranded at a Porsche dealership. And Porsche advises owners to not fast charge often regardless.

And lastly - EPA ranges have long been considered to be the most reliable range standards out there. When you score a bad score on a standardized test that everyone else takes, and your response is to shout, "well, the test is rigged!", you are not leading.

Don't get me wrong, the EPA test could be better (I'd prefer they up the average speed but reduce the 30% penalty, for example). But as things stand, it's the gold standard. Much better than WLTP, and way better than NEDC.

 

[StealthP3D]

The only cases in which the range of the Taycan becomes a critical issue is when driven in particularly cold climates, or particularly fast in areas with an insufficient distribution of rapid chargers, but we know that the number of rapid chargers will grow rapidly (heh!) in the next few years.

I think you are under-estimating the value of range for many people that goes well beyond cold weather and fast driving in rural areas. I've found that, mostly for convenience reasons, about 240-260 miles is the minimum I would accept even though I could make 200 miles work. At 200 miles trip planning becomes MUCH more important. 40-60 miles might not sound like much difference but it would impact me greatly even though I tend to not be a high miles driver on a daily basis.

Here's the interesting thing. You imply Taycan's target market is more interested in performance than range. But think about how a sports car is used, especially a four-door sports car like the Taycan. If you are going to use its capabilities you will either be on a track (ideally) or on scenic rural roads that are not heavily trafficked. That's why sports car clubs always have their rallies out in beautiful rural areas. And if you're on a track, range matters even more because there is precious little of it when driving that hard. Given these factors, I think the Taycan's 200 mile rated range is a VERY significant limitation to their target audience.

But recent reviews of the Taycan seem to indicate that the official 201 mile range is seriously sandbagged.

We will have to re-visit this after the Taycan has been out in the real world long enough for some real results but I think the EPA range is very likely accurate and realistic for legal driving on a mix of public roads, similar to the Model S. We will see if it has an advantage over the rated range of the Model S when driven at higher speeds on a racetrack but it would take quite a divergence from the EPA numbers relative to that of the Model S for it to even approach the track range of the Model S. I don't see that happening. Even if it approached the useable range of the Model S while burning up the Autobahn at triple-digit speeds (due to the higher efficiency allowed by the two-speed gearbox) that is a small consolation considering that is a VERY limited use type of situation. It may very well turn out that the Plaid Model S is better at everything!

An interesting contest for performance EV's would be a measure of which one could do the most laps (of a road racing course) in a fixed amount of time, say an hour or so. The exact amount of time would be short enough to keep speeds high enough to be interesting but long enough that neither car could drive the allotted amount of time if driven at 10/10ths without depleting the battery before the time ran out.

 

[bhtooefr]

Another bit of FUD I see, perpetuated by Porsche in the Taycan's EPA test documents, is a whole idea of (paraphrased) "the Taycan stays in first gear during the EPA testing, but in real world driving it's usually in second". I think they vastly overestimate the time spent in second outside of freeway driving, TBH.

...and I think I see the problem they have, planetary gearsets can have plenty of friction (it's easy for them to absolutely slaughter efficiency if they're not done extremely carefully - I can feel the reduction in friction in my e-bike shifting from 4th (where every stage of planetary gearing is engaged) to 5th (which is direct drive)), and they're using one as a reduction gear in first, and then bypassing it in second. I almost wonder if they'd have been better off using the planetary set as an overdrive and shifting it in above, oh, 130 km/h - it'd hurt efficiency up at higher speeds, but it'd maintain efficiency at lower speeds.

That said, I think we are to the point where mandating the US06, SC03, and Cold FTP tests for range/efficiency testing is needed.

It is worth noting that the 30% penalty is from the EPA test as of 1978 (when the raw results from FTP-75 were reported for city MPG, and the raw results from HWFET were reported for highway MPG), and nowadays only for EVs - the EPA hasn't considered it appropriate for ICEs for a few years now (it used to be that you could use either the 30% penalty or a slope/intercept method to get "derived 5-cycle" numbers, and when the EPA changed the slope/intercept, they changed the penalty as well).

If you actually do the full 5-cycle test procedure, true 5-cycle results AFAIK don't get the penalty, but rather modify the city and highway numbers using data collected from the three additional cycles. (Or, you get to adjust your penalty based on the other cycles, as Tesla does.)

 

[wdolson]

On the other side the Taycan has at 150 km/h a similar range as the Model 3 LR RWD.

View attachment 511425

There is still a large difference between EPA and reality.

Charging Speed of Taycan is also much better.

View attachment 511426

Here in there is a considerable drop of charging speed during winter, not only because of charge gate.

I can only shake my head when someone says that Tesla is years ahead of the others. Tesla has been a bit lazy the last few years. With NMC the others have the better battery technology with higher cycle numbers and better charging speed.

So I hope Tesla is now pushing the limits to go ahead again.

Another factor that is unknown right now, but is a major consideration when choosing a battery chemistry is the life of the battery. Tesla has been working hard to come up with battery chemistries that last a long time. Elon's goal is the million mile battery pack and claims to be close to that goal now.

Most things in engineering are a trade off. Rolls Royce makes ultra compfy ride cars, but they aren't exactly very good on the track because of the engineering decisions made. Li-ion batteries are a class of batteries as opposed to just one chemistry and there are many factors to weigh in choosing a chemistry: charge density, safety, lifetime and some other factors. There is a chart part way down in this article:
Types of Lithium-ion Batteries – Battery University

We will find out what decisions Porsche made for their batteries in a few years when Taycans have been out on the road a while. Porsche may have gone with a short lifespan chemistry to get other factors they wanted. This could result in Taycans needing a new battery at 50K miles. We don't know yet.

The battle for the perfect battery chemistry is a heated one and Tesla has been laser focused on it for longer than any other car maker. Though there have been many other non-car makers researching new chemistries.

It's possible that VW managed to get a hold of a vastly superior battery chemistry than Tesla and their EVs will be fantastic, but the more likely probability is that Porsche made some different decisions with battery chemistry compromises possibly sacrificing battery life for short term performance gains. I doubt this, but it's also possible they pushed the safety margins of the batteries, and that will result in battery fires. But considering the bad press Tesla's small number of battery fires got, I suspect they avoided doing anything that would increase the risk of fires in the name of performance.

At least some other car makers are beginning to act like they are serious about EVs. In some details other makers have thought of things Tesla didn't. Other makers also have different design approaches. For example Tesla's interior design concept is minimalist and the European luxury car makers are known for lots of gadgets and features. Porsche has always been a performance car brand and while Tesla makes cars that can outperform Porsches in some areas, Tesla has been focused on making all around cars that do more things than Porsches do. You probably won't see many Taycans at the loading bay at Ikea for example.

At the moment Tesla is way ahead of everybody in the areas that count. Tesla has been producing EVs from the ground up since 2012 and they are the only EV maker outside of China to mass produce one and they have had that car on the market for a couple of years already. Nobody has come close to Tesla's range yet. All of Tesla's car offerings have at least one model with a range over 300 miles. The competition is still struggling to break 260 with only a few cars available with ranges above 220. Tesla had a 265 rated range car in 2012.

Tesla has advertised that they are working on drive trains and batteries that will last 1 million miles. They have already tested Model 3 motors to that many miles (under laboratory conditions) and they made it with minimal wear. Tesla's battery management is possibly the best in any industry. They go to extremes to insure batteries have long useful lives.

According to Sandy Munro Tesla's electronics are up there with cutting edge mil spec and is, at minimum, a generation ahead of the rest of the car industry. He has been very critical of Tesla's metal bending ability and has pointed out they are behind the rest of the industry in fit and finish, but he has also said they are way ahead in a lot of other areas. His conclusion from taking apart the Model 3 is that the rest of the car industry should be very scared.

 

[egn1]

NMC of Audi/Porsche has a much better lifetime and fast charging ability than NCA. This is also the reason why Tesla uses NMC for energy storage and not NCA, which is dead after about 500 full cyles or 1200-1400 typical charge cycles, that EM mentioned in the investor call after indroduction of the power packs.

Charge- and Battery gates are the symptoms of the weakness of NCA.

Please read the following studies:
NCA
NMC

 

[Compton]

NMC of Audi/Porsche has a much better lifetime and fast charging ability than NCA. This is also the reason why Tesla uses NMC for energy storage and not NCA, which is dead after about 500 full cyles or 1200-1400 typical charge cycles, that EM mentioned in the investor call after indroduction of the power packs.

Charge- and Battery gates are the symptoms of the weakness of NCA.

Please read the following studies:
NCA
NMC

A) You are referencing old data from 2014 (while comparing it to Audi's/Porche's current performance) and b) Teslas have on average >90% capacity left after 250 000 km corresponding to at least 500 full cycles or ~1500 typical cycles. More crap you want to make up?

Why not compare old Roadster data from 2008 to most recent competitor offerings? Do you use the same logic with computers and mobile phones or only cars?

 

[SmartElectric]

2013 Smart Electric Drive. NMC cells. Charged to 100% indicated for 1000 days at 1/3 duty, so 400 full cycles. 99% original capacity. Full 20kW regen braking available in cold temperatures (-5C).

2013 Tesla Model S85. NCA cells. Charged to 90% indicated daily at 140000 km. 97% original capacity. Supercharged 200+ times. No regen below 4C until warmed up.

so: looks like these are different battery cells and managed differently but similar result

 

[wdolson]

2013 Smart Electric Drive. NMC cells. Charged to 100% indicated for 1000 days at 1/3 duty, so 400 full cycles. 99% original capacity. Full 20kW regen braking available in cold temperatures (-5C).

2013 Tesla Model S85. NCA cells. Charged to 90% indicated daily at 140000 km. 97% original capacity. Supercharged 200+ times. No regen below 4C until warmed up.

so: looks like these are different battery cells and managed differently but similar result

Tesla has also been through multiple generations of batteries, all of which fall under the NCA umbrella. Li-ion battery chemistry tech is very, very complex. There are so many variables that the only way to effectively find a new tweak to Li-ion chemistry is to make up an experimental mix and test it.

The second gen chemistry used for the 90 KWh packs proved to degrade a bit too fast when fast charged too much, so they put firmware limits on it. Older battery chemistries got extra checks in firmware updates last year.

Unless you're both an expert deep down into the weeds of Li-ion battery chemistry combined with an expert in the electronics of cutting edge battery management, it's pretty much impossible to say that x NMC chemistry is superior or inferior to y NCA chemistry or even compare a and b NCA chemistries.

What Tesla has demonstrated is that a Li-ion battery pack can have a long life with many charge cycles if managed correctly. And they have also demonstrated that they are the world leader in battery management control systems.

As an engineer and someone who has written firmware control systems, I'm deeply impressed at the quality of Tesla's battery management software (combined with state of the art hardware too). We don't know yet how good or how bad the battery management systems are on the new EVs coming to market. We don't even know how good the systems are in EVs that have come out in the last 3-4 years because few of those cars have enough miles on them to show long term battery wear yet.

I think the best the competition can do is parity with Tesla. In the area of battery management I doubt anyone can do more than very slight improvements over Tesla and if Tesla becomes aware of a better way, they will make those changes very quickly. If it's a software change, the whole fleet will get them soon, if it's a hardware change only newer cars will get it, but they are focused on the longest life possible for their batteries.

 

[egn1]

@Compton

A) You are referencing old data from 2014 (while comparing it to Audi's/Porche's current performance) and

b) Teslas have on average >90% capacity left after 250 000 km corresponding to at least 500 full cycles or ~1500 typical cycles. More crap you want to make up?

The shown graph is missleading, because it doesn't contain data up to now. If you add data of vehicles experiencing the Battery Gate you see a considerable jump down. Additionally they can only charge at SuCs at half the speed of new cars.

And if you analyse the above graph you see a very high variance in the values. This os also a quality issue of battery cell production.

I also have added values for my S85 to the above spreadsheet, and compared the result to data read with scan my tesla from the CAN bus. The degradation calculated with the spreedsheet was about 5 % at about 70.000 miles, but the degradation read from ScanMyTesla was 8 %.

I bought a brand new MX100D with starting usable capacity of 98.4 kWh according to Jason Hughes. After about 13.000 miles usable full pack is now 87.6kWh. This translates into a degradation of about 12 % in about 11 months. Part is real chemical degradation, but part is also Tesla intentionally turning down the capacity, because of long-term risk of lithium plating.

So please be quiet with your unqualified accusations and look at real data instead of quoting useless crap.

@wdolson
Both papers are from experts in lithium battery technology.

At current data Tesla is far away from having shown that their batteries have a long lifetime, at least not the generation used in S85 and SX90.

NMC battery technology is far ahead NCA. The only advantage of NCA is the high energy specific weight. But even here Tesla lost ground. The M3 battery has similar specific energy like the previous generation. It was expected that they would have 300+Wh/kg, but they still have only about 250Wh/kg.

There is a reason why Tesla is using NMC in there storage products - much more cycles under worse conditions.

 

[Yuri_G]

@Compton



The shown graph is missleading, because it doesn't contain data up to now. If you add data of vehicles experiencing the Battery Gate you see a considerable jump down. Additionally they can only charge at SuCs at half the speed of new cars.

And if you analyse the above graph you see a very high variance in the values. This os also a quality issue of battery cell production.

I also have added values for my S85 to the above spreadsheet, and compared the result to data read with scan my tesla from the CAN bus. The degradation calculated with the spreedsheet was about 5 % at about 70.000 miles, but the degradation read from ScanMyTesla was 8 %.

I bought a brand new MX100D with starting usable capacity of 98.4 kWh according to Jason Hughes. After about 13.000 miles usable full pack is now 87.6kWh. This translates into a degradation of about 12 % in about 11 months. Part is real chemical degradation, but part is also Tesla intentionally turning down the capacity, because of long-term risk of lithium plating.

View attachment 512038
So please be quiet with your unqualified accusations and look at real data instead of quoting useless crap.

@wdolson
Both papers are from experts. in lithium battery technology.

At current data Tesla is far away from having shown that their batteries have a long lifetime, at least not the generation used in S85 and SX90.

NMC battery technology is far ahead NCA. The only advantage of NCA is the high energy specific weight. But even here Tesla lost ground. The M3 battery has similar specific energy like the previous generation. It was expected that they would have 300+Wh/kg, but they still have only about 250Wh/kg.

There is a reason why Tesla is using NMC in there storage products - much more cycles under worse conditions.

Nice anecdotal evidence on battery degradation. Try actual statistical trends across Tesla's fleet next time.

 

[JRP3]

NMC battery technology is far ahead NCA. The only advantage of NCA is the high energy specific weight. But even here Tesla lost ground. The M3 battery has similar specific energy like the previous generation. It was expected that they would have 300+Wh/kg, but they still have only about 250Wh/kg.

Which suggests they gave up increased specific energy for improved longevity and higher charging rates, a reasonable trade off. Chemistry is not the only determiner of cell characteristics, other parameters can be adjusted.

 

[JRP3]

NMC of Audi/Porsche has a much better lifetime and fast charging ability than NCA.

options trader on Twitter
options trader
@VWCHARTS
@Porsche
Taycan exploded in Florida garage.

10:42 AM · Feb 17, 2020·Twitter for iPhone