I back VHDL for similar, but more cynical reasons.
As a TA for my school's digital design class, we used VHDL exclusively, but I also learned a little bit of Verilog for another class. Most (all?) of the students in an intro digital design class will have used C before, and when they see Verilog they will think "oh it's just like C." But it's not, and no matter how hard you beat them over the head, half of them will never quite learn. Teach them VHDL. Half of them will hate it and curse you for years to come, but at least they have a chance of learning some hardware from it.
Not sure what you consider to be "high" internal resistance, but one supercap example has a rated internal resistance of 0.29 mOhm. On the model I worked with (since discontinued) the rated current was about 4000 A. Not sure if that was just the rated voltage divided by the internal resistance, or if it was something lower to take heating into account.
I wasn't expecting much, but read the article anyway and I was pleasantly surprised by the stuff they got right. And even more pissed off by the stuff they got wrong.
Yes, switching out windings at low generator speeds will tend to make it more efficient. This applies to PM machines, and could probably apply to induction machines too (though with some other issues).
However, there are many, many other things limiting the power captured from the wind besides the generator efficiency. Blade efficiency tends to drop at low wind speeds, as does the total power available from the wind (it goes with the wind speed CUBED). I'm skeptical that improving the generator efficiency will result in much change in the overall efficiency.
Also, they fail to discuss cost. Very, very few wind turbines today use permanent magnet generators. Why? They're expensive, and you need power electronics rated to the full power of the generator (say, 2 MW) to rectify the output and invert it back to AC. Power electronics of that scale are expensive too. Instead, most wind turbines use induction generators (cheaper) with power electronics controlling only the rotor currents (rated to about 30% of the total turbine nameplate rating). Until they deal with the cost issue, it's really just another trick to get a little more energy out of a PM generator.
And, speaking of energy, the author of that article needs to be locked in a room and made to write "power != energy" on the blackboard a thousand times.
Winds farms don't scale, and do affect the environment some.
Gah. I wish that people would stop throwing around this FUD. There are plenty of studies out there, including one that I'm sure you've heard of that say relatively high wind penetration will work just fine. A lot of people like to recite the knee-jerk reaction that wind won't work, but I have yet to see much evidence that the sky is actually falling.
Actually one of the electric power industry's research shops has already done some studies on the effects of high plug-in hybrid penetration and power plant emissions. I saw a presentation that touched on some of this a year ago, and I think the conclusion was that air quality would get slightly worse in only a few few places (Cleveland, I think) and better pretty much everywhere else.
Now, I can't remember what the story was when it came to CO2...
Other than cost, the main disadvantage to DC is that it's point-to-point. Power goes in at one end and out at the other end, and that's it. With AC you get to build substations along the line and have all sort of options for moving your power around. This is one reason why you usually only see HVDC used for very long distance transmission.*
The cost of the converter stations can sometimes be made up for by the fact that you only have to run two conductors (simpler towers, less aluminum) rather than three. As the line gets longer this can make it look more economically attractive.
*HVDC is also often used for undersea cables, such as the Long Island Cross-Sound cable.
The grid operator can always direct the wind farm to curtail its output if it gets into this kind of situation.
The scarier situation is when you have sustained high wind speeds that finally reach the cut-out speeds of the wind turbines. That is, each turbine is programmed to feather its blades and rotate out of the wind if the wind speed is high enough to cause mechanical damage to the drive train. This would cause a farm that had previously been generating at full power to drop to practically nothing in a fairly short amount of time. Fortunately a wind even that big tends to be forecastable, and you could plan to have other generation available to cover what you lose when the wind farms trip.
The IEEE Spectrum editors had a blog post related to this article that the poster missed:
"To take another example, First Solar, a relatively young company based in Tempe, Arizona, has suddenly been getting a lot of attention with claims that it has figured out a way to make PV material at an installation cost of $1 per wattâ"though the global average for solar installations was in the range of $6 or $7 per watt last year. How plausible is that claim? Well, itâ(TM)s hard to know, because as a feature article appearing in this monthâ(TM)s IEEE Spectrum magazine points out, âoeThe company does not talk to reporters. Not at all.â"
The take-home point here? Be wary of companies that make extravagant claims without details. Especially if the best they can do now is $3/W.
Now I do like First Solar more than some of Slashdot's other favorite snake-oil salesmen (anyone remember EESTOR?), but I'm still suspicious.
Someone might want to correct me if I'm wrong, but I think that a typical solar inverter will have an efficiency of something around 90% (maybe 80%, to be safe). No idea what the efficiency of a typical PC power supply might be.
I think one advantage to going DC-AC-DC is that the AC part is very stiff -- it's connected to the utility and it's not very sensitive to you turning on another machine. Another is that if you want to go DC-DC-AC the DC-AC part needs to be bi-directional, which makes things more complicated and even more expensive.
I have heard that telecom stuff tends to be DC, so getting the DC PSUs might be reasonably easy if you knew where to go.
I'm a grad student in a wind power research group, so I've been keeping a close eye on the fall out from this particular grid event. From what I've heard, the problem was more to do with forecasting than anything else. My understanding is that the wind drop-off that evening was predicted, but at the time the ERCOT (the Texas grid operator) operating procedures didn't take the wind forecast into account. I'm not sure if they've found a better way to deal with the wind forecast yet, but I know they've been working on it.
Yes, wind was the problem, but odds are that if the same thing happens next year ERCOT will have figured out how to deal with it.
How exactly do they plan on making this work? TFA admits that they really have no idea and the cable execs are talking out of their arses, but if you were going to create some sort of 'mobile cable appliance' how would you do it? Where are you going to find enough bandwidth to handle the video?
Slashdot Mirror
I back VHDL for similar, but more cynical reasons.
As a TA for my school's digital design class, we used VHDL exclusively, but I also learned a little bit of Verilog for another class. Most (all?) of the students in an intro digital design class will have used C before, and when they see Verilog they will think "oh it's just like C." But it's not, and no matter how hard you beat them over the head, half of them will never quite learn. Teach them VHDL. Half of them will hate it and curse you for years to come, but at least they have a chance of learning some hardware from it.
1100 MW is a major typo. It's a 11 MW facility.
Not sure what you consider to be "high" internal resistance, but one supercap example has a rated internal resistance of 0.29 mOhm. On the model I worked with (since discontinued) the rated current was about 4000 A. Not sure if that was just the rated voltage divided by the internal resistance, or if it was something lower to take heating into account.
I wasn't expecting much, but read the article anyway and I was pleasantly surprised by the stuff they got right. And even more pissed off by the stuff they got wrong.
Yes, switching out windings at low generator speeds will tend to make it more efficient. This applies to PM machines, and could probably apply to induction machines too (though with some other issues).
However, there are many, many other things limiting the power captured from the wind besides the generator efficiency. Blade efficiency tends to drop at low wind speeds, as does the total power available from the wind (it goes with the wind speed CUBED). I'm skeptical that improving the generator efficiency will result in much change in the overall efficiency.
Also, they fail to discuss cost. Very, very few wind turbines today use permanent magnet generators. Why? They're expensive, and you need power electronics rated to the full power of the generator (say, 2 MW) to rectify the output and invert it back to AC. Power electronics of that scale are expensive too. Instead, most wind turbines use induction generators (cheaper) with power electronics controlling only the rotor currents (rated to about 30% of the total turbine nameplate rating). Until they deal with the cost issue, it's really just another trick to get a little more energy out of a PM generator.
And, speaking of energy, the author of that article needs to be locked in a room and made to write "power != energy" on the blackboard a thousand times.
Winds farms don't scale, and do affect the environment some.
Gah. I wish that people would stop throwing around this FUD. There are plenty of studies out there, including one that I'm sure you've heard of that say relatively high wind penetration will work just fine. A lot of people like to recite the knee-jerk reaction that wind won't work, but I have yet to see much evidence that the sky is actually falling.
Actually one of the electric power industry's research shops has already done some studies on the effects of high plug-in hybrid penetration and power plant emissions. I saw a presentation that touched on some of this a year ago, and I think the conclusion was that air quality would get slightly worse in only a few few places (Cleveland, I think) and better pretty much everywhere else.
Now, I can't remember what the story was when it came to CO2...
Other than cost, the main disadvantage to DC is that it's point-to-point. Power goes in at one end and out at the other end, and that's it. With AC you get to build substations along the line and have all sort of options for moving your power around. This is one reason why you usually only see HVDC used for very long distance transmission.*
The cost of the converter stations can sometimes be made up for by the fact that you only have to run two conductors (simpler towers, less aluminum) rather than three. As the line gets longer this can make it look more economically attractive.
*HVDC is also often used for undersea cables, such as the Long Island Cross-Sound cable.
The grid operator can always direct the wind farm to curtail its output if it gets into this kind of situation.
The scarier situation is when you have sustained high wind speeds that finally reach the cut-out speeds of the wind turbines. That is, each turbine is programmed to feather its blades and rotate out of the wind if the wind speed is high enough to cause mechanical damage to the drive train. This would cause a farm that had previously been generating at full power to drop to practically nothing in a fairly short amount of time. Fortunately a wind even that big tends to be forecastable, and you could plan to have other generation available to cover what you lose when the wind farms trip.
The IEEE Spectrum editors had a blog post related to this article that the poster missed:
"To take another example, First Solar, a relatively young company based in Tempe, Arizona, has suddenly been getting a lot of attention with claims that it has figured out a way to make PV material at an installation cost of $1 per wattâ"though the global average for solar installations was in the range of $6 or $7 per watt last year. How plausible is that claim? Well, itâ(TM)s hard to know, because as a feature article appearing in this monthâ(TM)s IEEE Spectrum magazine points out, âoeThe company does not talk to reporters. Not at all.â"
The take-home point here? Be wary of companies that make extravagant claims without details. Especially if the best they can do now is $3/W.
Now I do like First Solar more than some of Slashdot's other favorite snake-oil salesmen (anyone remember EESTOR?), but I'm still suspicious.
Someone might want to correct me if I'm wrong, but I think that a typical solar inverter will have an efficiency of something around 90% (maybe 80%, to be safe). No idea what the efficiency of a typical PC power supply might be. I think one advantage to going DC-AC-DC is that the AC part is very stiff -- it's connected to the utility and it's not very sensitive to you turning on another machine. Another is that if you want to go DC-DC-AC the DC-AC part needs to be bi-directional, which makes things more complicated and even more expensive. I have heard that telecom stuff tends to be DC, so getting the DC PSUs might be reasonably easy if you knew where to go.
A few links with information about the Feb. grid event: Blog post that sums up the ERCOT operations report.
I'm a grad student in a wind power research group, so I've been keeping a close eye on the fall out from this particular grid event. From what I've heard, the problem was more to do with forecasting than anything else. My understanding is that the wind drop-off that evening was predicted, but at the time the ERCOT (the Texas grid operator) operating procedures didn't take the wind forecast into account. I'm not sure if they've found a better way to deal with the wind forecast yet, but I know they've been working on it. Yes, wind was the problem, but odds are that if the same thing happens next year ERCOT will have figured out how to deal with it.
How exactly do they plan on making this work? TFA admits that they really have no idea and the cable execs are talking out of their arses, but if you were going to create some sort of 'mobile cable appliance' how would you do it? Where are you going to find enough bandwidth to handle the video?