Why fuel cell cars don't work • Hacker News Board

Re: Why fuel cell cars don't work

#81

Post by maxxxxx » Sun, Feb 22, 2015, 10:31 PM UTC

threeseed wrote: post #76

sandworm wrote: post #57

The threat to the car companies is that all-electric cars could dramatically decrease demand. They hope the complexity of fuel cells will keep demand going a few more decades. An all-electric car could have as few as a hundred moving parts. A few motors and connecting rods, some suspension componants, fans and a couple manual controls. You just don't need the thousands of tiny metal bits necessary to keep a gas/diese…

> THAT's what the car companies really fear. No it's not. Cars today are all basically the same and have been for decades. People aren't buying new cars because the components are failing. Generally it's for a new interior, new technologies e.g. support their new iPhone and lifestyle changes.

This may be true for some people but a lot of people I know buy a car when the old one falls apart.

Re: Why fuel cell cars don't work

#82

Post by allemagne » Sun, Feb 22, 2015, 10:38 PM UTC

astrocat wrote: post #63

sandworm wrote: post #57

The threat to the car companies is that all-electric cars could dramatically decrease demand. They hope the complexity of fuel cells will keep demand going a few more decades. An all-electric car could have as few as a hundred moving parts. A few motors and connecting rods, some suspension componants, fans and a couple manual controls. You just don't need the thousands of tiny metal bits necessary to keep a gas/diese…

I doubt this is the case. If anything, the reasons you give boil down to: "the core of the car - the moving vehicle bit - will become commodity" and when this happens, cars will become "cars" like phones are now "phones," as in, they're not so much phones anymore as tiny computers that can also make calls. And this will definitely NOT increase their lifetime. If anything, it will decrease it... just like we replace o…

This is an interesting idea, but I'd be curious to know what kinds of innovations in car functionality are possible. Also, assuming that you're referring to a kind of planned obsolescence, how can a company convince you to replace the entire vehicle if the moving parts are working fine?

If the computer is broken, replace the computer. Update the self-driving software over the internet. Why would you need to replace the whole car?

Unless, of course, a company tries to integrate points of failure all throughout the body, so that "replacing the computer" of the car or what have you becomes impossible due to the delicate electronic parts. Call me an optimist but at the price of cars today I don't think anyone but the super rich are gullible or decadent enough to fall for something like that instead of a cheaper, more reliable modular alternative. Unless the price of a brand new car becomes similar to that of a new phone, I don't think that future is very likely.

Re: Why fuel cell cars don't work

#83

Post by CalRobert » Sun, Feb 22, 2015, 10:39 PM UTC

parsnips wrote: post #45

krschultz wrote: post #40

Earlier quoted context omitted.

No, subways are very appropriate in certain situations. It wouldn't make sense to build an 8 lane highway to connect two towns in the backwoods of Alaska, but that doesn't mean 8 lane highways are a solution in search of a problem.

The point was I'm addressing is this idea that "sprawl" is somehow the problem that needs curing. It reminds me of a young child asking: "If people are so poor in ___________ why don't they move to _______?"

Sprawl is problematic for a number of reasons.

Environmental - a great deal of land is taking out of either agricultural usage or natural usage for sprawl. Large amounts of water are used on lawns.

Economic - Some people can't afford cars or motorcycles, but cars and motorcycles are the only practical way to get around most sprawling cities and towns

Safety - Cars are a dangerous way to get around. They kill half a million people a year. They also make roads even more dangerous for other users, like walkers or people on bicycles.

Social - Social isolation from a lack of interaction with neighbours is a large problem, and being spread out makes it harder to see other people. This is particularly true if you're elderly, can't drive, and don't have other ways of making social connections.

That's just a few. Not every place needs to be the City of London or Manhattan, but there are smarter ways to make suburbs.

Re: Why fuel cell cars don't work

#84

Post by astrocat » Sun, Feb 22, 2015, 10:39 PM UTC

maxxxxx wrote: post #81

threeseed wrote: post #76

Earlier quoted context omitted.

> THAT's what the car companies really fear. No it's not. Cars today are all basically the same and have been for decades. People aren't buying new cars because the components are failing. Generally it's for a new interior, new technologies e.g. support their new iPhone and lifestyle changes.

This may be true for some people but a lot of people I know buy a car when the old one falls apart.

the average car ownership time is somewhere around 6 years[1]. It feels like 3-year leases are becoming more and more common so I wouldn't be surprised to see this number drop significantly in the future.

[1] http://www.kbb.com/car-news/all-the-latest/average-length-of...)

Re: Why fuel cell cars don't work

#85

Post by bobbles » Sun, Feb 22, 2015, 10:40 PM UTC

thrownaway2424 wrote: post #58

Shivetya wrote: post #37

Get back to me when a practical, affordable, and go over 300 mile on a charge, battery powered car is here. Currently batteries are not a practical replacement. Currently if you want to hit near 300, you don't quite get there, your in the eighty thousand plus range if that low after common options for your only real choice. Home charging is still a four hour plus affair with the best chargers. Odd he claims that a hu…

Nearly all car trips are way under 300 miles. I believe the P95 trip distance (in the USA) is 30 miles each way. Most people can use an affordable pure-electric car such as the Leaf, e-Golf, or Fiat 500e for most trips. From a system optimization perspective, it's OK if some people (like you) always need the range of an internal combustion engine and everyone else switches to electric.

yeah 90% of my trips might be short, but i'm not going to go and rent a separate car when I do need a long trip

Re: Why fuel cell cars don't work

#86

Post by cma » Sun, Feb 22, 2015, 10:41 PM UTC

toomuchtodo wrote: post #68

astrocat wrote: post #63

Earlier quoted context omitted.

I doubt this is the case. If anything, the reasons you give boil down to: "the core of the car - the moving vehicle bit - will become commodity" and when this happens, cars will become "cars" like phones are now "phones," as in, they're not so much phones anymore as tiny computers that can also make calls. And this will definitely NOT increase their lifetime. If anything, it will decrease it... just like we replace o…

You know how you swap the powertrain on a Model S? You pull the motor (~70 lbs) and you drop the battery pack. 10 minutes tops. Sure, you could argue the interior will wear. Big deal. You'll have interiors you can swap on the power sled (battery + suspension + electric motor). Those sleds should last forever.

Why did you list the lbs for the motor and not the battery pack? =P

Re: Why fuel cell cars don't work

#87

Post by greglindahl » Sun, Feb 22, 2015, 10:43 PM UTC

Shivetya wrote: post #37

Get back to me when a practical, affordable, and go over 300 mile on a charge, battery powered car is here. Currently batteries are not a practical replacement. Currently if you want to hit near 300, you don't quite get there, your in the eighty thousand plus range if that low after common options for your only real choice. Home charging is still a four hour plus affair with the best chargers. Odd he claims that a hu…

Lots of people who have yet to own an electric car think like you do. People who've tried it and have home chargers discover:

1) It's really nice that your car is fully charged every morning... sure beats having to visit gas stations on a regular basis. 4 hours is not a big deal if you're asleep.

2) If you want to go long distance, a network like Tesla's supercharger does the trick. (Tesla owners drive SF-LA by stopping twice for 20 minutes, or pay for a 5-minute battery swap once.)

3) While the only long-range electric car today costs 80k+, Chevy and Tesla are both planning on releasing a cheaper long-range car in 2017.

Is it the case that anyone other than cars is betting on hydrogen fuel cells? The only big fuel cells that I've ever heard of are used in the datacenter industry, where Bloom Energy is the main fuel cell company, and they power theirs on hydrocarbons, not hydrogen.

Re: Why fuel cell cars don't work

#88

Post by nabla9 » Sun, Feb 22, 2015, 10:43 PM UTC

derekp7 wrote: post #36

gambiting wrote: post #32

Earlier quoted context omitted.

I am really really curious how anyone is trying to solve the problem of Hydrogen leaking out of literally any container you could put it into. Smallest molecule+huge pressure = leaks out of pure lead canisters. When Mercedes made their Hydrogen-powered concept car they said the whole tank would evaporate naturally within 3 weeks if I remember correctly? As hydrogen forms an explosive mixture with air, I cannot see th…

I thought the solution was to have an insulated tank and a refrigeration solution to keep the hydrogen liquid. Of course you will still have "leakage", as some of the hydrogen needs to be spent periodically to run the refrigeration, but this can be minimized with enough insulation.

30% of the heating value of hydrogen is wasted for liquefaction.

Re: Why fuel cell cars don't work

#89

Post by TheLoneWolfling » Sun, Feb 22, 2015, 10:47 PM UTC

allemagne wrote: post #82

astrocat wrote: post #63

Earlier quoted context omitted.

I doubt this is the case. If anything, the reasons you give boil down to: "the core of the car - the moving vehicle bit - will become commodity" and when this happens, cars will become "cars" like phones are now "phones," as in, they're not so much phones anymore as tiny computers that can also make calls. And this will definitely NOT increase their lifetime. If anything, it will decrease it... just like we replace o…

This is an interesting idea, but I'd be curious to know what kinds of innovations in car functionality are possible. Also, assuming that you're referring to a kind of planned obsolescence, how can a company convince you to replace the entire vehicle if the moving parts are working fine? If the computer is broken, replace the computer. Update the self-driving software over the internet. Why would you need to replace t…

If only it were so simple. You're assuming discrete components. And you're assuming consumer sanity.

"Every protocol the car uses is proprietary and protected under DCMA / etc. There are chips in every component that don't allow the car to start if any component isn't a first-party part."

We're already seeing this sort of thing. Look at the Tesla. Or the Lightning connector from Apple.

Re: Why fuel cell cars don't work

#90

Post by astrocat » Sun, Feb 22, 2015, 10:55 PM UTC

allemagne wrote: post #82

astrocat wrote: post #63

Earlier quoted context omitted.

I doubt this is the case. If anything, the reasons you give boil down to: "the core of the car - the moving vehicle bit - will become commodity" and when this happens, cars will become "cars" like phones are now "phones," as in, they're not so much phones anymore as tiny computers that can also make calls. And this will definitely NOT increase their lifetime. If anything, it will decrease it... just like we replace o…

This is an interesting idea, but I'd be curious to know what kinds of innovations in car functionality are possible. Also, assuming that you're referring to a kind of planned obsolescence, how can a company convince you to replace the entire vehicle if the moving parts are working fine? If the computer is broken, replace the computer. Update the self-driving software over the internet. Why would you need to replace t…

as automation increases, cars will likely become more computers than cars. And at that point I imagine the advances in computation and sensing will drive product refreshes. Even battery tech. Which all sound familiar...

Sure some of it can be over-the-air as Tesla has already shown, but the same happens on our current tech yet we still update the hardware pretty frequently. A new, better autopilot may not be able to run on a 3 year old car processor. Could you swap out the cars brains? Sure. Would that be lucrative for the car manufacturer? maybe not.

I think the biggest argument against what I'm proposing is that computers/phones have huge form-factor constraints that favor tightly integrated components - it's why RAM is no longer user replaceable in new, thinner macs. Cars, given their physical size, will likely never face that constraint making the modular approach you suggest much more feasible.