I'm a little confused on your point that regulation could work "just by charging"... Wouldn't that only work if all the EVs were already part of the baseline load? How would charging at 5kW be regulating load down when I was charging at 0kW (not 10kW) before with my trusty non-grid-connected gasoline engine?
I'm not sure how else to explain it. A car capable of 20kW (but charging at 5kW) can provide up to 15kW of regulation down, or 5kW of regulation up. Your gas car is always zero load, and has zero regulation capacity. These are basically the definitions of regulation up and regulation down.
One of the arguments that many EV enthusiasts make is that you can add a huge number of EVs to the road without adding any new baseline load because most charging would occur at off-peak times.
No; the benefit is that you don't need to add additional peaking generation and infrastructure. Look at this graph of the real-time load in California: www.caiso.com/outlook/outlook.html
On a hot summer afternoon, the load (red line) will approach the available generation (green line), at which point all available generation is online. Adding more generation would require building more generators, and eventually more infrastructure (transmission lines, etc.). Fortunately, most EV charging fits into the valley on the left of the graph (in the middle of the night) -- we can literally add millions of EVs over there, without any infrastructure upgrades.
My question is which benefit will we see - low up-front infrastructure costs because we won't have to build any new power plants? Or huge savings because we get to use V2G after we build the costly new power plants to bring up baseline load?
Yes on the first; the second isn't really correct. EV charging profiles (i.e. mostly scheduled at night) mean we won't need new infrastructure for a long time, but that has nothing to do with V2G.
For V2G and regulation, let's go back to my car -- it can vary its charge between 0 and 20kW. That means a fleet of 1000 cars can superimpose a 0-20megawatt signal on top of the CalISO load graph above. By doing so, we can avoid having to do the same thing with generators as they balance generation with load -- highly desirable because generators are inefficient and dirty when they're ramping up and down. (We still need to add generation as load increases, of course -- but we can do it in larger, more-optimal steps, leaving the fine-grained balancing act to the vehicle fleet.) This is the basic idea behind "ancillary services" and it's where the money is on the table right now.
Disclaimer: I worked on some of the software in the vehicle mentioned in the article. The article was a little light on technical details. Dr. Kempton is much more qualified to comment with respect to V2G technology, but I'll try to preemptively clear a few things up, here.
Why would I let the big bad utility company wear out my expensive battery?
Because they'd pay you more than enough to make it worthwhile. The details of the business model are undefined, but as TFA explains, there is a lot of money on the table (at least $4K/year), so there is considerable financial incentive to put a fleet of vehicles to use. The basic idea is that a vehicle owner would sign on with an aggregator, who would control a fleet (thousands or hundreds of thousands of vehicles) and sell regulation services to the utilities at the megawatt level. It could be that you'd lease your battery from the aggregator.
The most-valuable proposition is called ancillary services. Very simplistically, in this model you're not really moving much energy; you're really just selling the availability to provide fast-reacting regulation. Grid operation is a giant, complicated balancing act -- balancing generation with load.
Right now the balancing is done by ramping generator output up and down. As greater amounts of solar and wind make their way into the power mix, generators will end up doing even more regulation. Unfortunately, generators are generally least efficient and most polluting when ramping, so a fleet of vehicles that can provide small amounts of regulation within milliseconds is extremely attractive to grid operators.
But what if the utility company drains my battery when I need it for that long trip?
Obviously the system would have to be designed to take your individual driving needs into account. The good thing is that it doesn't really matter what you do as an individual -- the statistical behavior of the fleet as a whole remains predictable.
Furthermore, with a sufficiently large fleet of vehicles, it's possible to provide all the necessary regulation just by charging. If a vehicle is charging at 10kW, but is capable of charging at 20kW, then it can adjust its power up or down by 10kW, subject only to the constraint that it needs to be full by morning (or whenever). I've seen estimates by people more knowledgeable than I that we could regulate all of California with a fleet on the order of hundreds of thousands of EVs.
If you're doing all your regulation via charging, then you can't claim you're wearing out your battery prematurely (unless you were never planning to charge it again, of course).
However we all know that improperly charged NiCd and Lithium ion batteries can explode by themselves
No we don't. This is a misrepresentation of the truth, and you're also confusing two issues. Let's clear some of that up. I don't claim to know everything, but I am an electric vehicle engineer, and I have considerable experience with enormous Nickel- and Lithium-chemistry batteries.
First of all, the fires caused by Li-Ion cells have nothing to do with charging. The failures are the result of tiny manufacturing defects that cause a microscopic short circuit to develop inside the cell over time. The short could appear while charging, discharging, or while sitting on the shelf. The short causes extreme local heating inside the cell, and since Lithium is quite flammable it often burns. It is absolutely a dangerous condition, but it is not an explosion in any sense of the word. The root cause (the manufacturing defects) are thankfully limited to high-capacity cutting edge cells.
The other issue: Any chemical battery (or cell, technically) will overheat and suffer ill effects if overcharged, or if subjected to excess charge or discharge current. Depending on the type of cell, this usually means fluid leaks or venting of gas or nothing at all. Calling this an explosion is a gross exaggeration in almost every case. Yes, some types (e.g. large format lead-acid batteries) could develop enough heat and pressure to erupt and spray some electrolyte (yes, it's acid, and no, it won't melt your eyeballs) but it's just fear mongering to associate this with explosives.
First, efficiency of generating electricity (work done/energy produced) is 60% tops.
This is roughly correct for state of the art gas-fired plants. Efficiently numbers like the above don't even make sense, though, for the large (and growing) fraction of our power mix that is nuclear, and the growing portion that is wind-powered.
Then there is attenuation loss while its delivered to the consumer.
The power grid is over 90% efficient overall. Locally generated energy (say, from PV panels on your roof) is even better in this regard.
Then it has to be stored in batteries which lose energy over time.
It's starting to sound like you're outside your area of expertise. IAAEVE (I am an electric vehicle engineer.) Are you?
Electric engine has THEORETICAL top efficiency of around 45%.
This is total BS. Are you spreading misinformation deliberately or do you actually believe this? AC Propulsion's AC-150 drive system is about 90% efficient over a typical driving cycle. Follow the link to a spec sheet with the detailed efficiency map. Tesla Motors' propulsion system is based heavily on ACP's, and will be roughly the same in terms of efficiency.
The theoretical efficiency of gasoline engine (which I don't remember at the moment) is 2-3 times that.
The BS is flying thick, now. I don't know what you mean by "theoretical efficiency", but it's clear that you don't, either. Gasoline engines in the real-world cars I drive are around 15-18% efficient. (Did you really think they were 3 * 45 or 135% efficient?)
So for every calory [sic] of heat we burn (and release into atmoshere) with a gasoline engine we'd get 2-3 as much work.
Somehow you managed to get your conclusion in the right ballpark, but you have it backwards. Most modern EV propulsion systems are at least 3x as efficient as gasoline cars in a real-world, fair, wells-to-wheels energy comparison, making them about equivalent to 120-140 miles per gallon. You can do your own homework on this -- it's well documented. Tesla Motors' website has some interesting whitepapers and other material on the subject that's pretty easy to understand.
These cars will just end up burning more coal and release massive amount of greenhouse gases. But hey, it's cool to be green.
Spreading FUD when you don't know what you're talking about isn't cool at all. Even from coal, EVs are substantially more efficient and clean. This. Is. Well. Documented. And coal is just part of the power mix. Electricity is the ultimate flex fuel. And EV charging is biased towards off-peak times, when baseline (e.g. nuclear) energy is a larger part of the grid mix.
I love the way the article refers to the eleven cameras as "nearly a dozen". Using that system, there are nearly 61 minutes in a hour, 2 + 2 is nearly 5, and I nearly ate one gnome for breakfast this morning.
I wouldn't anticipate the price of platinum to vary much unless you need more platinum for a fuel cell as you do for the already mandatory emissions control devices.
You do. Fuel cells require on the order of 20x the platinum.
If what you say is true, then who do the makers of NiCd batteries say instruct you to fully deplete batteries on the instructions that ship with the products?
Some do, some don't.
They don't know the properties of their own products?
Correct -- generally they don't know the properties of their own products. Some literature advises against complete discharge, but some does not; the lack of consistent advice is more evidence of their general cluelessness. The literature included with cordless tools, etc., is likely written by marketing types who don't know any better; the "run it all the way down" myth is extremely pervasive in society. It's also not a problem for them at all if you need a new battery in a year or two.
Nickel-Cadmium cells (the only type that EVER had a documented "memory" problem) are effectively obsolete now, but the myth lives on...
One other anecdote:
Back in the day, RC racers would often do a complete discharge on a battery pack before charging it again. This can be good for the NiCd cells in some circumstances, and it's a low-tech but effective way of balancing a series of cells -- if they all start out completely discharged, then they'll all be fully charged at about the same time if charged in series. Anyway, they'd discharge the batteries by placing them in special trays with resistors that discharged each cell individually, rather than discharging the series pack as a whole.
Memory is a very specific occurrence in very specific conditions with a very specific type of cell (sintered plane nickel-cadmium). It exists. You've never seen it.
The above is spot on.
Another common cause of what is incorrectly thought to be "memory" is the corollary myth that you MUST deplete NiCd batteries completely before charging. While a full discharge can, in fact, sometimes be useful for certain types of cells, this is generally untrue for real-world batteries (comprised of multiple cells). A battery with several cells in series will always have slightly unbalanced cells, and the weaker cells will lose charge first. As the weakest cell begins to collapse, its neighbors in the string will crush it to zero volts, and then to a negative (reverse) voltage. To permanently damage a cell more effectively, you'd really have to apply yourself.
ALWAYS stop using the battery at the first sign of depletion -- continuing to use it will just kill one or more of its cells.
... which isn't very helpful, because at least on Earth, 100% of the hydrogen (in round numbers) is already bonded with something else -- like oxygen, as in H20 -- because that's what hydrogen likes to do. Separating it from other atoms requires energy. Lots of energy.
but you can use nuclear, wind, or solar power to perform that extraction.
And you could put that same (electrical) energy into a lithium-ion battery and go at least 4 times as far. Whether your goal is reducing CO2 emissions, reducing pollution, or simply saving money, it's hard to see why you'd opt for an expensive 20%-efficient solution.
Disclaimer: IAAEVE. (I am an electric vehicle engineer.)
Fuel cells for the electricity will be even better.
Fuel cells will make sense the day we have so much renewable or other "clean" energy that we can afford to throw 80% of it away on the hopelessly inefficient Water electrolysis->Hydrogen->Fuel Cell cycle. Right now qualified renewables in the USA are some fraction of 1%. When do you anticipate we'll hit the 500% mark so that 4/5ths of it can be discarded to make hydrogen?
On the other hand, battery energy density is doubling every 8-10 years, meaning that battery EVs will have a 600-1000 km range within 2 decades -- enough driving for one day, even for the most car-addicted nation on the planet. Battery costs will benefit tremendously from economies of scale, but platinum (for your fuel cells) is not exactly going to get cheaper in quantity.
Have you considered getting a job as a futurist? At this point I can guarantee that your track record will be better than many of the ones actually out there.
Okay, I'll give you one more freebie: 512 petabyte solid-state media is on the way.
To run a toaster we will need 40 square meters of solar panel
That's why you tie your system to the grid -- that way the solar only needs to cover your average energy needs instead of your peak power requirements.
No one would ever deploy an off-grid (i.e. battery backed) system and expect it to make financial or other sense. The only reason to deploy such a system is if you have no way to connect your site to the grid. Electricity is unique in that it must be consumed at the same instant it is created. Sharing with the grid allows you to draw your transient peaks as needed, and share your surplus the rest of the time, all at high efficiency.
I think solar is a great idea but a low tech solution makes more sense to me than a high tech solution and [...] our wasteful energy consumption is a very very serious problem
No disagreements there. Both approaches are part of the solution. Energy efficiency is extremely important (and valuable) and energy costs rise. Solar at 1$/watt, if ever realized, will propel mass adoption of a much-needed *clean* resource to handle daytime peak loads. Also a good thing.
I looked around for your 6.5 billion per week figure and could not find it.
Actually I wrote 6.5B/month, not per week, and I obtained that number here.
It looks like 1kw is 22,000 (for a non battery backed up system) so that's $33,000 for a non battery backed up 1.5kw system.
No, that's outrageous. (You pulled that number out of some turkey's comments on that blog page instead of researching the actual cost.) Installed (grid-tied, i.e. not battery-backed) PV is well documented to be under $10/watt. The panels themselves are only ~$5/W. Here is some data to start with. Our 3300W system cost $24K, and at least 25% of that went to the installers, and that price is before any state subsidies or tax credits.
Battery-backed PV technology is stagnant, and totally impractical and uneconomical unless you have no hope of obtaining a grid connection. Battery wear-out costs drown any remote chance of saving money. Tying PV to the grid is better for everyone -- the grid is a fantastic virtual storage device.
Solar just converts current energy that was shining on the house anyway.
I did a napkin calculation a year or so ago and at that time, you could give 100k houses free 1.5mw solar power (with inverters, trackers, and batteries) each year for the cost of the Iraq war.
Sorry, but your napkin math isn't even close.
First, I'm going to assume you meant 1.5 kilowatts instead of 1.5 megawatts, because the latter is just crazy in the context of private homes.
100,000 houses X 1.5kW = 150,000 kW or 150 MW.
Installed solar (unsubsidized) is between $8 and $9/watt. Let's call it $10 to make the math napkin-friendly. That brings us to $1.5 billion for your hypothetical install. The US is currently spending over $6.8 billion each MONTH on the war, and that's just the pentagon's estimate, which ignores plenty of hidden costs.
In other words, we could install 150MW per WEEK for the same cost of the Iraq war.
Nonetheless, my feeling is that there's no time like the present -- to put off a solar installation.
Yeah, that's probably why Wal*Mart is installing so much solar. I mean, it's not like Wal*Mart is famous for being pathologically frugal or anything. Obviously they haven't done their homework -- they're throwing money away on solar and they need you to set them straight.
We've just invented a new-and-improved [solar cell|battery|ultracapacitor] and it's really really great but we're not going to quote any actual absolute metrics.
We have units for expressing solar cell performance, but I didn't see any in TFA.
I was looking at the Tesla car this morning, I noticed that it has 450lbs. of batteries
Actually, it's closer to 900 pounds, or about 400kg.
then I was think at that time "Imaging a truck that hauls 10 tons of cargo, do I need 5 tons of batteries?
Imagine also an electrified rail system hauling that cargo most of the required distance, using grid power directly. Now you have a lot of liquid fuel left to power vehicles like your truck, for short-haul applications. Realize that this is how much of the world (excepting the US, of course) moves freight. This is precisely why the US is going to suffer more greatly than the rest of the world as fuel costs continue to rise.
Then I checked the energy density of hydrogen and lithium battery, I found that you need 1500kg of lithium ion batteries to contain same amount of energy in 1kg of hydrogen
Your figures are incorrect. A 40kWh lithium pack (roughly equivalent to 1 US gallon of gasoline or 1kg of hydrogen) weighs less than 300kg based on cars like AC Propulsion's eBox. In addition, that energy will power a motor at ~90% efficiency, instead of a ~15% efficient gasoline engine, or a 50% efficient fuel cell. The latter, of course, doesn't account for the inefficiency of producting the hydrogen in the first place. Don't forget to include the weight of a 10,000PSI hydrogen tank in your estimates.
Based on these data, I am not convinced that the battery is a pure better solution for transportation than hydrogen fuel cell hybrid configuration.
Except that, even if you ignore the poor efficiency, hydrogen fuel cells don't exist today in any practical form, and have serious problems that make even their staunch supporters estimate them to be decades away from practical use.
Here is the fundamental flaw with Hydrogen, as clearly as I can explain it.
Hydrogen, as we all know, is not an energy resource -- we have to make it somehow. (There isn't a vast pool of it somewhere just waiting to be tapped.) In this sense hydrogen is a battery. But we already have much better batteries.
To make hydrogen, you can split water, using electricity, in a process called electrolysis. The round-trip efficiency for this process (energy -> hydrogen -> energy) is quite poor -- around 25%. This means you get one unit of energy out for every 4 you put in. If you put the same electrical energy into a lithium-type battery pack, you could drive ~4 times farther using the same energy.
The other practical way to make hydrogen is to reform it from other hydrocarbons -- typically natural gas. The problem in this case is that, if you have natural gas, you're far better burning it directly in a reciprocating engine. Converting the gas to hydrogen is inefficient, regardless of whether you burn the hydrogen (as in this BMW) or convert it to electricity (as in a fuel cell).
In addition to being inefficient, hydrogen fuel cells (which convert hydrogen into electrical energy) have a long list of problems that are presently not talked about much, because they're obscured by more fundamental problems. One amazing dealbreaker is the fact that hydrogen fuel cells only have a useful life of a few years.
This is hardly anything I'd consider "details". As TFA clearly states, Microsoft still hasn't offered any specifics, or evidence, and they probably never will.
Slashdot Mirror
I'm a little confused on your point that regulation could work "just by charging"... Wouldn't that only work if all the EVs were already part of the baseline load? How would charging at 5kW be regulating load down when I was charging at 0kW (not 10kW) before with my trusty non-grid-connected gasoline engine?
I'm not sure how else to explain it. A car capable of 20kW (but charging at 5kW) can provide up to 15kW of regulation down, or 5kW of regulation up. Your gas car is always zero load, and has zero regulation capacity. These are basically the definitions of regulation up and regulation down.
One of the arguments that many EV enthusiasts make is that you can add a huge number of EVs to the road without adding any new baseline load because most charging would occur at off-peak times.
No; the benefit is that you don't need to add additional peaking generation and infrastructure. Look at this graph of the real-time load in California:
www.caiso.com/outlook/outlook.html
On a hot summer afternoon, the load (red line) will approach the available generation (green line), at which point all available generation is online. Adding more generation would require building more generators, and eventually more infrastructure (transmission lines, etc.). Fortunately, most EV charging fits into the valley on the left of the graph (in the middle of the night) -- we can literally add millions of EVs over there, without any infrastructure upgrades.
My question is which benefit will we see - low up-front infrastructure costs because we won't have to build any new power plants? Or huge savings because we get to use V2G after we build the costly new power plants to bring up baseline load?
Yes on the first; the second isn't really correct. EV charging profiles (i.e. mostly scheduled at night) mean we won't need new infrastructure for a long time, but that has nothing to do with V2G.
For V2G and regulation, let's go back to my car -- it can vary its charge between 0 and 20kW. That means a fleet of 1000 cars can superimpose a 0-20megawatt signal on top of the CalISO load graph above. By doing so, we can avoid having to do the same thing with generators as they balance generation with load -- highly desirable because generators are inefficient and dirty when they're ramping up and down. (We still need to add generation as load increases, of course -- but we can do it in larger, more-optimal steps, leaving the fine-grained balancing act to the vehicle fleet.) This is the basic idea behind "ancillary services" and it's where the money is on the table right now.
Hope this helps at least a little.
Disclaimer: I worked on some of the software in the vehicle mentioned in the article. The article was a little light on technical details. Dr. Kempton is much more qualified to comment with respect to V2G technology, but I'll try to preemptively clear a few things up, here.
Why would I let the big bad utility company wear out my expensive battery?
Because they'd pay you more than enough to make it worthwhile. The details of the business model are undefined, but as TFA explains, there is a lot of money on the table (at least $4K/year), so there is considerable financial incentive to put a fleet of vehicles to use. The basic idea is that a vehicle owner would sign on with an aggregator, who would control a fleet (thousands or hundreds of thousands of vehicles) and sell regulation services to the utilities at the megawatt level. It could be that you'd lease your battery from the aggregator.
The most-valuable proposition is called ancillary services. Very simplistically, in this model you're not really moving much energy; you're really just selling the availability to provide fast-reacting regulation. Grid operation is a giant, complicated balancing act -- balancing generation with load.
Right now the balancing is done by ramping generator output up and down. As greater amounts of solar and wind make their way into the power mix, generators will end up doing even more regulation. Unfortunately, generators are generally least efficient and most polluting when ramping, so a fleet of vehicles that can provide small amounts of regulation within milliseconds is extremely attractive to grid operators.
But what if the utility company drains my battery when I need it for that long trip?
Obviously the system would have to be designed to take your individual driving needs into account. The good thing is that it doesn't really matter what you do as an individual -- the statistical behavior of the fleet as a whole remains predictable.
Furthermore, with a sufficiently large fleet of vehicles, it's possible to provide all the necessary regulation just by charging. If a vehicle is charging at 10kW, but is capable of charging at 20kW, then it can adjust its power up or down by 10kW, subject only to the constraint that it needs to be full by morning (or whenever). I've seen estimates by people more knowledgeable than I that we could regulate all of California with a fleet on the order of hundreds of thousands of EVs.
If you're doing all your regulation via charging, then you can't claim you're wearing out your battery prematurely (unless you were never planning to charge it again, of course).
However we all know that improperly charged NiCd and Lithium ion batteries can explode by themselves
No we don't. This is a misrepresentation of the truth, and you're also confusing two issues. Let's clear some of that up. I don't claim to know everything, but I am an electric vehicle engineer, and I have considerable experience with enormous Nickel- and Lithium-chemistry batteries.
First of all, the fires caused by Li-Ion cells have nothing to do with charging. The failures are the result of tiny manufacturing defects that cause a microscopic short circuit to develop inside the cell over time. The short could appear while charging, discharging, or while sitting on the shelf. The short causes extreme local heating inside the cell, and since Lithium is quite flammable it often burns. It is absolutely a dangerous condition, but it is not an explosion in any sense of the word. The root cause (the manufacturing defects) are thankfully limited to high-capacity cutting edge cells.
The other issue: Any chemical battery (or cell, technically) will overheat and suffer ill effects if overcharged, or if subjected to excess charge or discharge current. Depending on the type of cell, this usually means fluid leaks or venting of gas or nothing at all. Calling this an explosion is a gross exaggeration in almost every case. Yes, some types (e.g. large format lead-acid batteries) could develop enough heat and pressure to erupt and spray some electrolyte (yes, it's acid, and no, it won't melt your eyeballs) but it's just fear mongering to associate this with explosives.
First, efficiency of generating electricity (work done/energy produced) is 60% tops.
This is roughly correct for state of the art gas-fired plants. Efficiently numbers like the above don't even make sense, though, for the large (and growing) fraction of our power mix that is nuclear, and the growing portion that is wind-powered.
Then there is attenuation loss while its delivered to the consumer.
The power grid is over 90% efficient overall. Locally generated energy (say, from PV panels on your roof) is even better in this regard.
Then it has to be stored in batteries which lose energy over time.
It's starting to sound like you're outside your area of expertise. IAAEVE (I am an electric vehicle engineer.) Are you?
Electric engine has THEORETICAL top efficiency of around 45%.
This is total BS. Are you spreading misinformation deliberately or do you actually believe this? AC Propulsion's AC-150 drive system is about 90% efficient over a typical driving cycle. Follow the link to a spec sheet with the detailed efficiency map. Tesla Motors' propulsion system is based heavily on ACP's, and will be roughly the same in terms of efficiency.
The theoretical efficiency of gasoline engine (which I don't remember at the moment) is 2-3 times that.
The BS is flying thick, now. I don't know what you mean by "theoretical efficiency", but it's clear that you don't, either. Gasoline engines in the real-world cars I drive are around 15-18% efficient. (Did you really think they were 3 * 45 or 135% efficient?)
So for every calory [sic] of heat we burn (and release into atmoshere) with a gasoline engine we'd get 2-3 as much work.
Somehow you managed to get your conclusion in the right ballpark, but you have it backwards. Most modern EV propulsion systems are at least 3x as efficient as gasoline cars in a real-world, fair, wells-to-wheels energy comparison, making them about equivalent to 120-140 miles per gallon. You can do your own homework on this -- it's well documented. Tesla Motors' website has some interesting whitepapers and other material on the subject that's pretty easy to understand.
These cars will just end up burning more coal and release massive amount of greenhouse gases. But hey, it's cool to be green.
Spreading FUD when you don't know what you're talking about isn't cool at all. Even from coal, EVs are substantially more efficient and clean. This. Is. Well. Documented. And coal is just part of the power mix. Electricity is the ultimate flex fuel. And EV charging is biased towards off-peak times, when baseline (e.g. nuclear) energy is a larger part of the grid mix.
I love the way the article refers to the eleven cameras as "nearly a dozen". Using that system, there are nearly 61 minutes in a hour, 2 + 2 is nearly 5, and I nearly ate one gnome for breakfast this morning.
I wouldn't anticipate the price of platinum to vary much unless you need more platinum for a fuel cell as you do for the already mandatory emissions control devices.
You do. Fuel cells require on the order of 20x the platinum.
If what you say is true, then who do the makers of NiCd batteries say instruct you to fully deplete batteries on the instructions that ship with the products?
Some do, some don't.
They don't know the properties of their own products?
Correct -- generally they don't know the properties of their own products. Some literature advises against complete discharge, but some does not; the lack of consistent advice is more evidence of their general cluelessness. The literature included with cordless tools, etc., is likely written by marketing types who don't know any better; the "run it all the way down" myth is extremely pervasive in society. It's also not a problem for them at all if you need a new battery in a year or two.
Nickel-Cadmium cells (the only type that EVER had a documented "memory" problem) are effectively obsolete now, but the myth lives on...
One other anecdote:
Back in the day, RC racers would often do a complete discharge on a battery pack before charging it again. This can be good for the NiCd cells in some circumstances, and it's a low-tech but effective way of balancing a series of cells -- if they all start out completely discharged, then they'll all be fully charged at about the same time if charged in series. Anyway, they'd discharge the batteries by placing them in special trays with resistors that discharged each cell individually, rather than discharging the series pack as a whole.
Memory is a very specific occurrence in very specific conditions with a very specific type of cell (sintered plane nickel-cadmium). It exists. You've never seen it.
The above is spot on.
Another common cause of what is incorrectly thought to be "memory" is the corollary myth that you MUST deplete NiCd batteries completely before charging. While a full discharge can, in fact, sometimes be useful for certain types of cells, this is generally untrue for real-world batteries (comprised of multiple cells). A battery with several cells in series will always have slightly unbalanced cells, and the weaker cells will lose charge first. As the weakest cell begins to collapse, its neighbors in the string will crush it to zero volts, and then to a negative (reverse) voltage. To permanently damage a cell more effectively, you'd really have to apply yourself.
ALWAYS stop using the battery at the first sign of depletion -- continuing to use it will just kill one or more of its cells.
... which isn't very helpful, because at least on Earth, 100% of the hydrogen (in round numbers) is already bonded with something else -- like oxygen, as in H20 -- because that's what hydrogen likes to do. Separating it from other atoms requires energy. Lots of energy.
but you can use nuclear, wind, or solar power to perform that extraction.
And you could put that same (electrical) energy into a lithium-ion battery and go at least 4 times as far. Whether your goal is reducing CO2 emissions, reducing pollution, or simply saving money, it's hard to see why you'd opt for an expensive 20%-efficient solution.
Disclaimer: IAAEVE. (I am an electric vehicle engineer.)
Fuel cells for the electricity will be even better.
Fuel cells will make sense the day we have so much renewable or other "clean" energy that we can afford to throw 80% of it away on the hopelessly inefficient Water electrolysis->Hydrogen->Fuel Cell cycle. Right now qualified renewables in the USA are some fraction of 1%. When do you anticipate we'll hit the 500% mark so that 4/5ths of it can be discarded to make hydrogen?
On the other hand, battery energy density is doubling every 8-10 years, meaning that battery EVs will have a 600-1000 km range within 2 decades -- enough driving for one day, even for the most car-addicted nation on the planet. Battery costs will benefit tremendously from economies of scale, but platinum (for your fuel cells) is not exactly going to get cheaper in quantity.
It's not piracy, it's unauthorized crimes.
Wait, the crimes are unauthorized? Are you trying to confuse me?
Remember, it's not piracy, it's "unauthorized copying". Oh, wait...
Okay, I'll give you one more freebie: 512 petabyte solid-state media is on the way.
You could use the same logic to conclude that 512 terabyte solid-state media is on the way.
To run a toaster we will need 40 square meters of solar panel
That's why you tie your system to the grid -- that way the solar only needs to cover your average energy needs instead of your peak power requirements.
No one would ever deploy an off-grid (i.e. battery backed) system and expect it to make financial or other sense. The only reason to deploy such a system is if you have no way to connect your site to the grid. Electricity is unique in that it must be consumed at the same instant it is created. Sharing with the grid allows you to draw your transient peaks as needed, and share your surplus the rest of the time, all at high efficiency.
I think solar is a great idea but a low tech solution makes more sense to me than a high tech solution and [...] our wasteful energy consumption is a very very serious problem
No disagreements there. Both approaches are part of the solution. Energy efficiency is extremely important (and valuable) and energy costs rise. Solar at 1$/watt, if ever realized, will propel mass adoption of a much-needed *clean* resource to handle daytime peak loads. Also a good thing.
we need to have a 'pendantic olympics'
Oooh, can I compete? I guess for starters I'll point out that the word you're looking for is 'pedantic'.
-1 BS.
Seagate drives are the only ones I know of that still have a 5-year retail warranty. Why didn't you replace yours, then?
I looked around for your 6.5 billion per week figure and could not find it.
Actually I wrote 6.5B/month, not per week, and I obtained that number here.
It looks like 1kw is 22,000 (for a non battery backed up system) so that's $33,000 for a non battery backed up 1.5kw system.
No, that's outrageous. (You pulled that number out of some turkey's comments on that blog page instead of researching the actual cost.) Installed (grid-tied, i.e. not battery-backed) PV is well documented to be under $10/watt. The panels themselves are only ~$5/W. Here is some data to start with. Our 3300W system cost $24K, and at least 25% of that went to the installers, and that price is before any state subsidies or tax credits.
Battery-backed PV technology is stagnant, and totally impractical and uneconomical unless you have no hope of obtaining a grid connection. Battery wear-out costs drown any remote chance of saving money. Tying PV to the grid is better for everyone -- the grid is a fantastic virtual storage device.
Solar just converts current energy that was shining on the house anyway.
Absolutely!
I did a napkin calculation a year or so ago and at that time, you could give 100k houses free 1.5mw solar power (with inverters, trackers, and batteries) each year for the cost of the Iraq war.
Sorry, but your napkin math isn't even close.
First, I'm going to assume you meant 1.5 kilowatts instead of 1.5 megawatts, because the latter is just crazy in the context of private homes.
100,000 houses X 1.5kW = 150,000 kW or 150 MW.
Installed solar (unsubsidized) is between $8 and $9/watt. Let's call it $10 to make the math napkin-friendly. That brings us to $1.5 billion for your hypothetical install. The US is currently spending over $6.8 billion each MONTH on the war, and that's just the pentagon's estimate, which ignores plenty of hidden costs.
In other words, we could install 150MW per WEEK for the same cost of the Iraq war.
Nonetheless, my feeling is that there's no time like the present -- to put off a solar installation.
Yeah, that's probably why Wal*Mart is installing so much solar. I mean, it's not like Wal*Mart is famous for being pathologically frugal or anything. Obviously they haven't done their homework -- they're throwing money away on solar and they need you to set them straight.
Summary:
We've just invented a new-and-improved [solar cell|battery|ultracapacitor] and it's really really great but we're not going to quote any actual absolute metrics.
We have units for expressing solar cell performance, but I didn't see any in TFA.
Why not - and turn them into tourist traps?
With hookers! And blackjack! In fact, forget the tourist traps!
I was looking at the Tesla car this morning, I noticed that it has 450lbs. of batteries
Actually, it's closer to 900 pounds, or about 400kg.
then I was think at that time "Imaging a truck that hauls 10 tons of cargo, do I need 5 tons of batteries?
Imagine also an electrified rail system hauling that cargo most of the required distance, using grid power directly. Now you have a lot of liquid fuel left to power vehicles like your truck, for short-haul applications. Realize that this is how much of the world (excepting the US, of course) moves freight. This is precisely why the US is going to suffer more greatly than the rest of the world as fuel costs continue to rise.
Then I checked the energy density of hydrogen and lithium battery, I found that you need 1500kg of lithium ion batteries to contain same amount of energy in 1kg of hydrogen
Your figures are incorrect. A 40kWh lithium pack (roughly equivalent to 1 US gallon of gasoline or 1kg of hydrogen) weighs less than 300kg based on cars like AC Propulsion's eBox. In addition, that energy will power a motor at ~90% efficiency, instead of a ~15% efficient gasoline engine, or a 50% efficient fuel cell. The latter, of course, doesn't account for the inefficiency of producting the hydrogen in the first place. Don't forget to include the weight of a 10,000PSI hydrogen tank in your estimates.
Based on these data, I am not convinced that the battery is a pure better solution for transportation than hydrogen fuel cell hybrid configuration.
Except that, even if you ignore the poor efficiency, hydrogen fuel cells don't exist today in any practical form, and have serious problems that make even their staunch supporters estimate them to be decades away from practical use.
Here is the fundamental flaw with Hydrogen, as clearly as I can explain it.
Hydrogen, as we all know, is not an energy resource -- we have to make it somehow. (There isn't a vast pool of it somewhere just waiting to be tapped.) In this sense hydrogen is a battery. But we already have much better batteries.
To make hydrogen, you can split water, using electricity, in a process called electrolysis. The round-trip efficiency for this process (energy -> hydrogen -> energy) is quite poor -- around 25%. This means you get one unit of energy out for every 4 you put in. If you put the same electrical energy into a lithium-type battery pack, you could drive ~4 times farther using the same energy.
The other practical way to make hydrogen is to reform it from other hydrocarbons -- typically natural gas. The problem in this case is that, if you have natural gas, you're far better burning it directly in a reciprocating engine. Converting the gas to hydrogen is inefficient, regardless of whether you burn the hydrogen (as in this BMW) or convert it to electricity (as in a fuel cell).
In addition to being inefficient, hydrogen fuel cells (which convert hydrogen into electrical energy) have a long list of problems that are presently not talked about much, because they're obscured by more fundamental problems. One amazing dealbreaker is the fact that hydrogen fuel cells only have a useful life of a few years.
This is hardly anything I'd consider "details". As TFA clearly states, Microsoft still hasn't offered any specifics, or evidence, and they probably never will.