4 ms·

Yes, so the claim is that the complexity electric motors + batteries + planetary gears + all the ICE/electric interfaces are more reliable than a traditional tr

by bumby 1y ago

Yes, so the claim is that the complexity electric motors + batteries + planetary gears + all the ICE/electric interfaces are more reliable than a traditional transmission. I’m open to that argument but nobody has really elaborated in detail. Admittedly, I’m slightly skeptical (at least in the case of a manual gearbox) but would like to hear more detail.

vel0city 1y ago
> at least in the case of a manual gearbox Ok, let's say you go stupid simple for the transmission and have a super basic classic four-speed manual. Right off the bat, you've got clutch wear. In that e-CVT, there are no sliding clutches. Everything is connected all the time. Immediately, we see a wear component that will eventually need replacing. Not might need replacing, will need replacing. It is a consumable part, designed to wear. You now have cable assemblies which will eventually stretch over the life of the car. Those will eventually need adjustment. Once again, these just don't exist at all with the e-CVT system. Now you have a shifter and gear selector. This will need to slip in and out of other gears. Often this is not a perfect shift, imparting wear on the transmission components. Once again, you don't have a gear selector in an e-CVT, this wear never happens. This wear can cause premature failure of the transmission. This is largely controlled by the skill of the operator, sure. Just a few quick examples. But this is then also an incredibly basic manual transmission, you won't find such a thing on pretty much any recent mass market car. These days you'll see complicated automatics with servos and what not to control the gear ratios, massively more gear ratios, rely on fluid channels to push things around inside, rely on computers to operate them effectively, etc. And as Retric mentioned, there are analoges for most of the hybrid components in a pure-ICE car. You already have a DC motor set to drive the car in the starter motor. You already have an inverter, the alternator. You're not really adding a ton of new things, you're just massively upsizing some of the things you already have and massively simplifying a lot of the other components.
- bumby 1y ago
  >there are analoges for most of the hybrid components in a pure-ICE car. The error in this is in assuming the analogs have equivalent reliability or cost in each system.* (The original point is about cost of ownership). They don’t. So, while both have batteries, the reliability and cost of each is very different. Same with the ICE component etc. My issue is broad generalizations about reliability without speaking to the nuance. I’m not set against the idea of a hybrid being more reliable or cheaper, but more against the superficial generalization. * also the analogs miss some of the complexity. Yes, both have a battery, but a hybrid requires high voltage and auxiliary batteries, meaning the battery system is by definition more complex.
  - vel0city 1y ago
    > The error in this is in assuming the analogs have equivalent reliability The majority of my comment points out how the equivalent analoge on the pure ICE is massively less reliable than the hybrid. Who is ignoring the reliability of the different components here again? Sure, there is a HV and a LV battery. They're both solid state devices and thus generally pretty reliable when it comes to cars. The LV battery faces far less wear. The overall system is considerably more reliable in the hybrid than the same in the ICE. It's near impossible to generally talk about the prices of something like the HV battery, it varies greatly based on what models you're talking about. One car might have a replacement used battery that's good for many years be $700, another might be extremely bespoke and rare and be $10k. If I were to judge prices of pure ICE transmissions based on extremely rare hyper cars I might say an ICE transmission replacement costs $20k or more. The details matter greatly when judging TCO on potentially low market cars.
    - bumby 1y ago
      If your point is you can’t talk specifics about price of ownership, you’ve wasted both of our time during this entire discussion. That’s what it’s about. A better answer would have just been like the commenter below, who said the information is hard to come by. A hardliner stance where there’s a lot of uncertainty isn’t a great hill to die on.
      - Retric 1y ago
        That’s not quite what they said, in 98% of hybrids sold these components are significantly more reliable than in ICE cars. He can’t speak to the last 2%, but I doubt your in the market for a hyper car and really this concerned about TCO.
        bumby 1y ago
        The hyper car argument is really just a digression and it cuts both ways (which I think was the posters point). If you want to compare hyper car prices, you need to compare hyper gas-only to hyper hybrid. But that’s not really germane to the generalizations being used here. Where is your 98% claim coming from? The other poster made no such claim. The irony is there is published info on TCO from companies like Edmunds and Kelly Blue Book. We can take umbrage with some of the factors or weighting, but it’s a better starting point in real data than anything bright up so far. A claim could easily be crafted using such information that addresses everything I’ve raised (maintenance, depreciation, different designs, etc). Yet the discussion seems almost wholly based on vibes and near-ideological bias. It reads to me as conclusions searching for data instead of the other way around. The best I can tell from that information, TCO is largely a wash. For comparable manufacturers/designs, TCO over the first 5-10 years seems to be about the same. You can pick some designs that favor one side or the other, but in general they are roughly within 10% (usually considerably less) of one another. Given the uncertainty due to the variables in question, it seems reasonable to conclude there aren’t strong conclusions on TCO differences between them.
Retric 1y ago
> all the ICE/electric interfaces are more reliable than a traditional transmission The argument goes like this, ICE cars have an alternator (electric motor 1), starter (electric motor 2), battery, and transmission. A beefier alternator (generator) + starter (electric motor driving the car) + battery adds less complexity than a transmission. That’s the simplest EV design where the engine only ever charges the battery. It’s perfectly viable for a long range plug in hybrid that only ever uses the engine on long trips. The downside is batteries have conversion losses, so most hybrids have various ways of directly using engine power which then adds complexity. But ultimately hybrids are more complicated than EV’s but very much can be simpler than modern ICE cars. PS: Technically some old ICE designs like dynastart used to do the same as hybrids where the same electric motor acted as a starter and alternator but in modern ICE vehicles the tradeoff around now little time the starter is needed and how little power the alternator needs to generate means it’s more efficient to separate it out. http://www.isettadoc.com/files/dynastart.pdf http://www.isettadoc.com/files/dynastart.pdf
- bumby 1y ago
  Thanks for taking the time to detail it. I think that’s a reasonable take. A couple caveats: 1) most modern hybrids use the ICE at higher speeds for efficiency, right? What does that mean in terms of added complexity and reliability? 2) somewhat of a tangent, but in the original post was regarding cost of ownership, and reliability was brought into the discussion because cost is a function of reliability. But the alternator vs electric motor aspect misses the original point about cost, considering the motor may cost 8x-10x to replace. All that to say, reliability and cost of ownership is complicated. I was pushing back on the overly simplistic takes and appreciate you adding some nuance.
  - Retric 1y ago
    > most modern hybrids use the ICE at higher speeds for efficiency, right? What does that mean in terms of added complexity and reliability? There’s a lot of tradeoffs involved which I’m not an expert on. However, ICE cars need an engine capable of low end torque and a good efficiency across a huge RPM range, hybrids can use a much simpler engine design optimized for where the engine operates best because the EV side handles the low end just fine. A hybrid engine is also used for fewer hours of operation over its lifespan so in general (because exceptions exist) the gas engine in an hybrid is more reliable than the gas engine in an equivalent ICE. That said, car manufacturers can use up that margin to save weight etc so it’s not a guarantee. In the end it’s a huge design space, saying something is a hybrid doesn’t actually tell you much about what’s under the hood.

by bumby 1y ago

vel0city 1y ago

> at least in the case of a manual gearbox Ok, let's say you go stupid simple for the transmission and have a super basic classic four-speed manual. Right off the bat, you've got clutch wear. In that e-CVT, there are no sliding clutches. Everything is connected all the time. Immediately, we see a wear component that will eventually need replacing. Not might need replacing, will need replacing. It is a consumable part, designed to wear. You now have cable assemblies which will eventually stretch over the life of the car. Those will eventually need adjustment. Once again, these just don't exist at all with the e-CVT system. Now you have a shifter and gear selector. This will need to slip in and out of other gears. Often this is not a perfect shift, imparting wear on the transmission components. Once again, you don't have a gear selector in an e-CVT, this wear never happens. This wear can cause premature failure of the transmission. This is largely controlled by the skill of the operator, sure. Just a few quick examples. But this is then also an incredibly basic manual transmission, you won't find such a thing on pretty much any recent mass market car. These days you'll see complicated automatics with servos and what not to control the gear ratios, massively more gear ratios, rely on fluid channels to push things around inside, rely on computers to operate them effectively, etc. And as Retric mentioned, there are analoges for most of the hybrid components in a pure-ICE car. You already have a DC motor set to drive the car in the starter motor. You already have an inverter, the alternator. You're not really adding a ton of new things, you're just massively upsizing some of the things you already have and massively simplifying a lot of the other components.

bumby 1y ago
>there are analoges for most of the hybrid components in a pure-ICE car. The error in this is in assuming the analogs have equivalent reliability or cost in each system.* (The original point is about cost of ownership). They don’t. So, while both have batteries, the reliability and cost of each is very different. Same with the ICE component etc. My issue is broad generalizations about reliability without speaking to the nuance. I’m not set against the idea of a hybrid being more reliable or cheaper, but more against the superficial generalization. * also the analogs miss some of the complexity. Yes, both have a battery, but a hybrid requires high voltage and auxiliary batteries, meaning the battery system is by definition more complex.
- vel0city 1y ago
  > The error in this is in assuming the analogs have equivalent reliability The majority of my comment points out how the equivalent analoge on the pure ICE is massively less reliable than the hybrid. Who is ignoring the reliability of the different components here again? Sure, there is a HV and a LV battery. They're both solid state devices and thus generally pretty reliable when it comes to cars. The LV battery faces far less wear. The overall system is considerably more reliable in the hybrid than the same in the ICE. It's near impossible to generally talk about the prices of something like the HV battery, it varies greatly based on what models you're talking about. One car might have a replacement used battery that's good for many years be $700, another might be extremely bespoke and rare and be $10k. If I were to judge prices of pure ICE transmissions based on extremely rare hyper cars I might say an ICE transmission replacement costs $20k or more. The details matter greatly when judging TCO on potentially low market cars.
  - bumby 1y ago
    If your point is you can’t talk specifics about price of ownership, you’ve wasted both of our time during this entire discussion. That’s what it’s about. A better answer would have just been like the commenter below, who said the information is hard to come by. A hardliner stance where there’s a lot of uncertainty isn’t a great hill to die on.
    - Retric 1y ago
      That’s not quite what they said, in 98% of hybrids sold these components are significantly more reliable than in ICE cars. He can’t speak to the last 2%, but I doubt your in the market for a hyper car and really this concerned about TCO.
      - bumby 1y ago
        The hyper car argument is really just a digression and it cuts both ways (which I think was the posters point). If you want to compare hyper car prices, you need to compare hyper gas-only to hyper hybrid. But that’s not really germane to the generalizations being used here. Where is your 98% claim coming from? The other poster made no such claim. The irony is there is published info on TCO from companies like Edmunds and Kelly Blue Book. We can take umbrage with some of the factors or weighting, but it’s a better starting point in real data than anything bright up so far. A claim could easily be crafted using such information that addresses everything I’ve raised (maintenance, depreciation, different designs, etc). Yet the discussion seems almost wholly based on vibes and near-ideological bias. It reads to me as conclusions searching for data instead of the other way around. The best I can tell from that information, TCO is largely a wash. For comparable manufacturers/designs, TCO over the first 5-10 years seems to be about the same. You can pick some designs that favor one side or the other, but in general they are roughly within 10% (usually considerably less) of one another. Given the uncertainty due to the variables in question, it seems reasonable to conclude there aren’t strong conclusions on TCO differences between them.

Retric 1y ago

> all the ICE/electric interfaces are more reliable than a traditional transmission The argument goes like this, ICE cars have an alternator (electric motor 1), starter (electric motor 2), battery, and transmission. A beefier alternator (generator) + starter (electric motor driving the car) + battery adds less complexity than a transmission. That’s the simplest EV design where the engine only ever charges the battery. It’s perfectly viable for a long range plug in hybrid that only ever uses the engine on long trips. The downside is batteries have conversion losses, so most hybrids have various ways of directly using engine power which then adds complexity. But ultimately hybrids are more complicated than EV’s but very much can be simpler than modern ICE cars. PS: Technically some old ICE designs like dynastart used to do the same as hybrids where the same electric motor acted as a starter and alternator but in modern ICE vehicles the tradeoff around now little time the starter is needed and how little power the alternator needs to generate means it’s more efficient to separate it out. http://www.isettadoc.com/files/dynastart.pdf http://www.isettadoc.com/files/dynastart.pdf

bumby 1y ago
Thanks for taking the time to detail it. I think that’s a reasonable take. A couple caveats: 1) most modern hybrids use the ICE at higher speeds for efficiency, right? What does that mean in terms of added complexity and reliability? 2) somewhat of a tangent, but in the original post was regarding cost of ownership, and reliability was brought into the discussion because cost is a function of reliability. But the alternator vs electric motor aspect misses the original point about cost, considering the motor may cost 8x-10x to replace. All that to say, reliability and cost of ownership is complicated. I was pushing back on the overly simplistic takes and appreciate you adding some nuance.
- Retric 1y ago
  > most modern hybrids use the ICE at higher speeds for efficiency, right? What does that mean in terms of added complexity and reliability? There’s a lot of tradeoffs involved which I’m not an expert on. However, ICE cars need an engine capable of low end torque and a good efficiency across a huge RPM range, hybrids can use a much simpler engine design optimized for where the engine operates best because the EV side handles the low end just fine. A hybrid engine is also used for fewer hours of operation over its lifespan so in general (because exceptions exist) the gas engine in an hybrid is more reliable than the gas engine in an equivalent ICE. That said, car manufacturers can use up that margin to save weight etc so it’s not a guarantee. In the end it’s a huge design space, saying something is a hybrid doesn’t actually tell you much about what’s under the hood.