Taycan is the least reliable of all EV's

[McgR]

So your dealer claims the failure rate of Taycan gen 2 is lower than gen 1?

Yes he does. He has a very nice gen 2 CT 4. It doesn’t have RWS and HUD but all other options I have now.

I am tempted but I think I am going for a PHEV and may go back to EV in 2-3 years. Want to know how they perform long time first.

Sponsored

 

[whitex]

So your dealer claims the failure rate of Taycan gen 2 is lower than gen 1?

They are learning from Elon. They can honestly say without fear of contradiction that the US Gen 2 Taycan failure rate is 0%, which is much lower than Gen 1. Come on, I dare you to find a single Gen 2 Taycan Gen2 owner in the US who had any technical issues with their Gen 2 Taycan so far.

 

[McgR]

They are learning from Elon. They can honestly say without contradiction that the US Gen 2 Taycan failure rate is 0%, which is much lower than Gen 1. Come on, I dare you to find a single Gen 2 Taycan Gen2 owner in the US who had any technical issues with their Gen 2 Taycan so far.

Very true.

They say the OBC and heater at redesigned and more reliable. But we only know after a year (a winter) at least if that is correct.

Sure they have done the testing but I presume they have done the same on gen 1 and didn’t find the problems we are all having.

I think it will only be confirmed if the changes they made will make a difference in a year.

 

[WasserGKuehlt]

The stereotypical IT response "Have you turn it off and on again" still applies, as cars are becoming large, complex computers nowadays. With increasing levels of complexity, it is becoming impossible to test all possible states the software can get itself into, therefore starting fresh is not a bad idea. Heck, I saw a computer science whitepaper a while back suggesting architecting systems which regularly reboot themselves in order to return to a known state. The paper was even suggesting resetting hardware/software in the middle of normal operation (while idling for example, or even between unrelated operations).

I’m of a less understanding position on this. Software engineering is severely behind in tackling correctness - and that’s the only way one can hope to tame complexity. Not one single person can understand the full stack of a contemporary cloud-based service, for instance, to the degree that they can assert its correctness. Decomposition and sticking with clear interfaces achieves verifiable local correctness, but that does fuckall for side effects or other bad assumptions that one can’t verify from within. There’s obviously a need for starting from a known state (updates, resets - in the cleanup for next tenant sense) but these should be explicit actions taken by the customer rather than “it’s that time of the month”.

Of course the drawback of restarting cars, it takes an ever increasingly longer amount of time. Most cars today already don't boot when you get in. There are different approaches, Porsche seems to shut down the car after some inactivity locked. It might even reboot and suspend to RAM instead, so that when you get back into the car, it appears to startup quickly. Rebooting performs a bunch of integrity/safety checks, as well as brings the ECUs back to a known state. Suspend to RAM just halts everything until you are ready to use it, but provides a very quick startup (compared to actual boot). If car is parked for longer, it might drop down to suspend to storage or actually total power off to save power, which is why some cars take longer to startup after a prolonged time parked.

Imo the Taycan is very quick to start/achieve operational status while not using power in its idling state. I know your findings are different, but based on my observations it doesn’t use one electron if, say, it’s parked in the snow for a full day of skiing, or several days in a garage.

I’ll definitely try the reset procedure detailed by @McgR next time i encounter the slightest issue. (Might be years, though )

 

[Scandinavian]

I’m of a less understanding position on this. Software engineering is severely behind in tackling correctness - and that’s the only way one can hope to tame complexity.

I agree with this. It is difficult with software, and then the fact that German companies are not very good at software in general. The VW group certainly is at the back of the field by a clear margin. US companies are much more advanced in SW in general even if there of course still can be errors in it.

I learnt that from my Service advisor at an early stage. I had a lot of small issues that did not register in the fault logs. He advised me to come in next time I had an issue, but to not turn off or lock the car. By locking the car all error records were deleted ! Brilliant engineering from Porsche.
By following his advice they could capture a hell of a lot of errors that were sent to the mothership. The response from Porsche was: “”We have never seen these error codes before””! NO WONDER!

 

[Tooney]

I learnt that from my Service advisor at an early stage. I had a lot of small issues that did not register in the fault logs. He advised me to come in next time I had an issue, but to not turn off or lock the car. By locking the car all error records were deleted ! Brilliant engineering from Porsche.

That's an interesting tip.
But won't the car turn itself off after um 30 minutes of inactivity?

 

[Scandinavian]

That's an interesting tip.
But won't the car turn itself off after um 30 minutes of inactivity?

I think it does, but I drove straight to the workshop and parked outside the door and walked in. He was in the car within a minute and straight to a service guy with the Porsche computer and connected up.

 

[Vim Schrotnock]

FWIWConsumer Reports in it’s annual Best /worse auto issue rates the Taycan as above average for reliability!

NeIn. The 2020 Porsche Taycan is listed in Consumer Reports as having 12 recalls. There is no data on this car for reliability.

 

[Old man]

NeIn. The 2020 Porsche Taycan is listed in Consumer Reports as having 12 recalls. There is no data on this car for reliability.

Well on page 73 the Taycan ( years not specified) is listed as having above average reliability as demarcated by an upward green arrow

 

[DJP]

I've had a number of 911's over the years so I'm familiar with the bottleneck on dealer resources
But the Taycan takes that to a whole new level - serious issues taking 4-5months to get a slot at a main dealer (thats as of March 24 - 3 dealers London & -Northwest) or just leave it here and register it as a breakdown with Porsche Assist and we'll get around to it when we can - could be weeks we don't know & no commitment.
Dealer capacity to address issues for me is the principal frustration - accepting new tech is likely to go wrong far more frequently than ICE equivalents.

 

[TJH13]

I'm going to guess that Porsche owners are just a little more particular about what constitutes a problem,

Meanwhile, on the electric SUV list the Mach-E has a 100% reliability score.

100% with the Porsche statement. As for the Mach-E they must not have any responses. ;-)

 

[whitex]

I’m of a less understanding position on this. Software engineering is severely behind in tackling correctness - and that’s the only way one can hope to tame complexity. Not one single person can understand the full stack of a contemporary cloud-based service, for instance, to the degree that they can assert its correctness. Decomposition and sticking with clear interfaces achieves verifiable local correctness, but that does fuckall for side effects or other bad assumptions that one can’t verify from within. There’s obviously a need for starting from a known state (updates, resets - in the cleanup for next tenant sense) but these should be explicit actions taken by the customer rather than “it’s that time of the month”.

You are not wrong on this, I could tell you some software stories that would make your hairs stand up (different forum scope, also not all suitable for the internet) - never underestimate how badly something can be coded.

However, besides the fact that a lot of software is often badly written, even if you do everything right, there is still an issue of complexity outpacing the ability to validate. A restart does offer a lot of benefits, as it brings everything back to a known state. A restart doesn't have to be system wide. A processor could reset itself between instructions (needs to be design to reset quickly) and that actually simplifies verification tremendously. To oversimplify, imagine having 100 instructions a CPU can handle, you can easily verify that each instruction starting after reset will complete what it needs to. However, add almost infinite possible sequences, and now validation becomes exponentially more complex. Now scale this to a distributed system like a car with up to 100 ECU's, each in their own state, etc. It's still easier to validate that the car car survive 24hrs ON and simply reset it every time it's parked, than try to validate that it can run flawlessly for 15 years straight. It's all a matter of probability math - there is much lesser chance of the car getting into an undefined state over 24hrs than over 15 years. Frequent return to a known state, such as reset, is just a way to simplify validation.

Just for completeness, yes there is a branch of computer science where mathematical proofs are written to prove that the code is error free. Having seen this long ago in school, I can tell you the math involved is way beyond my abilities or even interest. I am glad this is used for things like code controlling nuclear reactors (IIRC the one I was shown in school was from Pickering Nuclear Generation Station), but doing the same for a car is simply not feasible with current computing power. Who knows, maybe some day in the future AI will be able to create such mathematical proofs for large scale software, though perhaps we'll need a Dyson sphere to power one of those AI's though.

Imo the Taycan is very quick to start/achieve operational status while not using power in its idling state. I know your findings are different, but based on my observations it doesn’t use one electron if, say, it’s parked in the snow for a full day of skiing, or several days in a garage.

We'll have to agree to disagree on this one. You could put a power meter on the 12V battery while parked if you want some proof.

 

[Vrm001]

I have had numerous problems with my Gts in addition to the common ones of failure of heating system and glitchy entertainment system. Perhaps the most worrying is repeated failure of braking system .. on 3 separate occasions. Also there have been two inexplicable wheel buckling. I have other complaints about the quality of the car but won’t bore you with them.

This is my first Porsche and have been totally underwhelmed by the response from Porsche UK. Having owned several BMWs, Mercedes and a Jaguar F type… this car has had numerous more issues than the others combined.

 

[Hirschaj]

wheel buckling

What does "wheel buckling" mean?

 

[whitex]

What does "wheel buckling" mean?

Example from google (picture is worth a thousand words):

Sponsored