For a given amount of power, if you double the voltage (and thus halve the current), you halve the resistance. OR, you reduce the copper and keep the resistance the same. Or any combination thereof.
Higher voltages = less loss and less copper.
I must have missed something during my electric classes. Here's what I remember:
P = VI = RI^2
The reason why P remains the same while V and I compensate each other should be precisely because R is supposed to be constant. So I still don't see the huge variation you mention. Maybe you could still point me to that scientific reference explaining it.
I never expected a 240V equipment to care about neutral if it had the expected ground (for safety), but that doesn't seem to be the case, although I have no clue as why
Because if you draw 120V on a neutral-grounded 10-30, you dump current to ground.
But that's the thing. Forget the ground for a second. There should not be a need for a neutral on a 240V outlet if the device is going to use 240V. It doesn't matter if one is neutral while the other carries all the 240V, or if they are both hot and opposed by 180 degrees.
If you want to take 120V from there, you should be on a 120V outlet.
We all agree that the ground is not for feeding the device. It's for keeping its metallic parts free of current, for safety.
But if we do use the ground, a neutral-grounded 120V grid works just as well. There's no power going to the ground. It just becomes the reference.
There are safety problems I can imagine with that, tho:
Wrong wiring or "smart" users: if the device relies on neutral to be grounded but the outlet is badly wired or the user manages to reverse the plug, there will be 120V waiting for him on the metallic grounded parts
Wire resistance: the little resistance of the ground wire could make the device's end of it to show some volts instead of being truly grounded
Lightning strikes: the ground close to where the lightning stroke may not be quite the same as in other places and that can cause a surge on devices if their neutral is shorted to the ground.
Our body's resistance is roughly the same regardless of the voltage applied, so we'd get a worse kick from 240V than from 120V.
True, but your common sockets would be rated for half the current. No need to up the average power of the average plug. Double voltage, half current. And either half resistance or less copper.
I get the impression that you fell on the same fallacy as another comment regarding how much current an outlet gives.
When it says 120V/20A or 240V/10A it doesn't mean that it carries that many amperes all the time. It does carry that many volts, but the amperes are directly proportional to the resistance applied. Our body.
Huge? Why? Can you point to some scientific references?
V=IR.
I thought we were talking about power, P = VI, where when you increase V and decrease I in the same proportions, P remains the same. Isn't this usually the case?
No, it uses more copper. It's a case of one live, a neutral, and a ground, or two lives, a neutral, and a ground, all to carry the same amount of current. It used to be considered acceptable to wire the neutral and the ground together (NEMA 10-30), but no longer is (hence NEMA 14-30 is the new standard).
Ok, I see my error there. I never expected a 240V equipment to care about neutral if it had the expected ground (for safety), but that doesn't seem to be the case, although I have no clue as why. Must be to optimize the current consumption by profiting from this design directly in the electric motors?
I'm saying that our current system restricts you for no good reason. It's not "needlessly complex" in the least to have all of your sockets on 240V and have plugs with a lower current plug into higher-current sockets. What's needlessly complex is having a mix of incompatible sockets with two different power delivery configurations (single phase and split phase).
Yes, it would be simpler to design a grid on 240V in that case. The case for 120V would be safety.
Our body's resistance is roughly the same regardless of the voltage applied, so we'd get a worse kick from 240V than from 120V.
Since the huge majority of appliances don't need the extra volts, we have a boost in safety with a 120V grid.
It's amazing! 9 out of 10 people they stop after spotting them with the rod actually have bombs! What that says about the people that aren't being stopped is a mystery...
Oh, don't get me started on 120V. It's a huge waste of copper and power
Huge? Why? Can you point to some scientific references?
that we only went with because it's easier to make a 120V incandescent lightbulb than a 240V.:P
I'll have to take your word on that.
Anyway, you won't find a 120V 30A outlet in your average US home because we run most higher power devices at (the more reasonable) 240V (although we do it in a weird way -- split phase).
There's a reason for using split phase and it's not that weird. It creates two independent single phase circuits providing 110V and it can be used as one 220V circuit by replacing the neuter with the other phase. It saves copper, as opposed to your very first statement.
But even in that case, it'd still be useful. Homes usually have both 30A and 50A 240V sockets. But you can't plug a 30A into a 50A without an adapter, even though there's no reason why you shouldn't be able to. And you should also be able to plug a 120V plug into one phase of your 240V/30A or 50A sockets, but you can't do that without an adapter, either.
It's not clear to me what you're asking. You want a socket to provide both 110V and 220V and be able to plug such devices according to their needs? It seems a bit complicated in my opinion.
Usually when the system is defined to be 110V for most appliances and 220V for exceptions such as the dryer or maybe the stove, one gets 110V outlets everywhere and exception outlets where such appliances are supposed to be installed. Why would you want to make each and every outlet more complex to be able to support exceptional cases?
OTOH 110v 20A is a lot more common, and you can plug a 110v 15A plug into a 110v 20A outlet;
Only if you have a special 15A/20A hybrid outlet. The standard NEMA 5-20R doesn't have a T-shaped slot; it only has the horizontal on that side. This is the US trying to correct a weakness in our outlet system after it was discovered; it's a bandaid on the problem of having entirely different pin layouts on each socket. The Australian standard of having different pin *sizes* deals with this problem automatically.
You lost me. If I understand the T-shaped slot correctly, it's supposed to be a higher current slot that still allows you to plug lower current devices in it. What's the problem with that?
You can still do it if you want, it's just aesthetically arguable to have a huge box (as opposed to this or this), protruding of the wall with plugs all around, so I'm not sure that's the point.
It's huge size is certainly not an asset. Just browse other comments for examples, especially how it becomes an unwanted volume for laptop power bricks.
The only pro might be the extra security. I won't argue that, but I can argue if all that security really pays for the inconveniences or if it's just redundant.
I'm kind of surprised the place is still standing... I really sort of thought it would have burned down from an electrical fire by now.
That's because not being able to run a microwave along with a stove without tripping a breaker usually means that you have working breakers, not just that your wiring is inconvenient.
AFAIK, in most cases the difference between ground and neutral is irrelevant. It's a nice security measure, but it hardly means that your building will catch fire just because it's missing.
US: I like it, especially when it's not grounded and doesn't have horribly different sizes. And I'm from a country where the Europlug is the standard, and it's not France ; ) It's tendency to unplug only depends on the outlet. If it bends, you can unbend, so.. no problem there either. In fact, you can compensate for bad outlets by bending.
Europlug: It's ok too, but can take a little more space than the US one.
Australian: haven't tried it.
Italian: haven't tried it.
Dannish: haven't tried it.
Brittish: hate it. It's enormous, makes it impractical to fork one outlet into many.
When a socket is for higher current, it doesn't mean that it always provides such high current. It means that it supports it if the device you plug requires it. So when you have a low current plug (and hence a low current device), you might appreciate being able to plug it into a higher current socket.
The opposite is not true, tho. When you have a low current socket, it can melt and cause fire if you try to use it with high current devices. Of course, your breakers will probably disarm first, but in any case you don't want to try that.
That's why both directions is not an option, while one direction is.
Doesn't seem like he missed the point. By insisting on the pi detail, he is pointing out that imprecision between Math and real world has so little to do with the speedometer's innacuracy that it's irrelevant. And this, my friend, has nothing to do with things working in theory and not in practice.
The theory and the practice in this case are: if we know the radius, we know the speed. Problem is, we don't know the radius, so let's take the minimum standard and add a buffer.
That's where the real innacuracy probably comes from.
If I understand it correctly, the rotation of magnets generates electric current directly related to the angular velocity, which is then extrapolated to the velocity.
I believe it's something like min wheel radius x pi x angular velocity, the later being represented by the electric current.
It's not the same as bicycle speedometers, which really measure the time between two passes of the magnet during one wheel revolution.
I no longer use Windows at home. I like my Ubuntu, even though:
- it took me days to have it working on my tablet PC, installing driver versions that are not part of the distro - takes me a custom script to rebuild drivers at every kernel update (because of first point) - had to disable ACPI (blame Microsoft and HP if you want: its ACPI works according to Vista, not according to standard. So Linux kernels refuse to talk to it) - takes a rmmod ehci_hcd/modprobe ehci_hcd to make my webcam work 1 out of 3 times - doesn't support screen rotation - battery life sucks (yes, I've been through all battery optimization tricks) - and so on.
I still have fun with it and enjoy having paid in time rather than money to have whatever is working, working. I also love to be able to workaround problems with scripting and shell commands, which is not the strongest quality of Windows.
Yes, M$ is evil. Yes, M$ pays to have their drivers. Yes, Linux is great. But... it still lags in support for new hardware. That's the point.
We'll see about holding up on court when the Psystar affair ends.
Meanwhile, Apple's business seems to be selling complete solutions, hardware and software. I think they made it perfectly clear by now that they are NOT interested in allowing people to play with half of their solution and mix it with another half that they DON'T WANT TO SUPPORT.
They want their full solutions to just work and gain such a fame. When people insist in using halves with unsupported parts, it creates a new type of user that will be bound to have a worse user experience and potentially hurt the "just work" image they want to nurture.
So it seems reasonable to assume that they DON'T WANT you as a customer unless you buy the product as THEY want to SELL IT.
What I understood from the article is that the program uses something in the likes of a mix of neural networks with fuzzy logic to detect what the normal behavior of a certain software is.
Example: it runs for a certain number of iterations in "analyze" mode, creating rules that link input with output.
Then it's triggered when an abnormal behavior occurs and tries to find a patch to the software so that the cause of abnormal behavior becomes normal without impacting the previously detected rules.
Example: (obviously simplified) It identifies a rule that links typing something in "url" field and the following actions: 1. identify protocol 2. identify host name 3. resolve host name 4. connect to host 5. ask for path 6. parse/show response
If at some point something is written in "url" field but breaks the above rules, it will create a patch that will force this faulty input to behave nicely.
If my understanding is correct, there's no magic or perpetual motion involved. It's a matter of using learning followed by inference. Maybe both at the same time.
I can only guess... Linux may be the easiest way to get a free OS and tweak it to your needs, since it already runs on everything from your generic PC to your electric toothbrush, then they probably held the opinion that C# was the current fashion in programming languages.
Slashdot Mirror
Geebus! Put a disclaimer next time! Some people browse from work places!
I thought we were talking about power
For a given amount of power, if you double the voltage (and thus halve the current), you halve the resistance. OR, you reduce the copper and keep the resistance the same. Or any combination thereof.
Higher voltages = less loss and less copper.
I must have missed something during my electric classes. Here's what I remember:
P = VI = RI^2
The reason why P remains the same while V and I compensate each other should be precisely because R is supposed to be constant. So I still don't see the huge variation you mention. Maybe you could still point me to that scientific reference explaining it.
I never expected a 240V equipment to care about neutral if it had the expected ground (for safety), but that doesn't seem to be the case, although I have no clue as why
Because if you draw 120V on a neutral-grounded 10-30, you dump current to ground.
But that's the thing. Forget the ground for a second. There should not be a need for a neutral on a 240V outlet if the device is going to use 240V. It doesn't matter if one is neutral while the other carries all the 240V, or if they are both hot and opposed by 180 degrees.
If you want to take 120V from there, you should be on a 120V outlet.
We all agree that the ground is not for feeding the device. It's for keeping its metallic parts free of current, for safety.
But if we do use the ground, a neutral-grounded 120V grid works just as well. There's no power going to the ground. It just becomes the reference.
There are safety problems I can imagine with that, tho:
Our body's resistance is roughly the same regardless of the voltage applied, so we'd get a worse kick from 240V than from 120V.
True, but your common sockets would be rated for half the current. No need to up the average power of the average plug. Double voltage, half current. And either half resistance or less copper.
I get the impression that you fell on the same fallacy as another comment regarding how much current an outlet gives.
When it says 120V/20A or 240V/10A it doesn't mean that it carries that many amperes all the time. It does carry that many volts, but the amperes are directly proportional to the resistance applied. Our body.
Here's a very interesting article about that.
Huge? Why? Can you point to some scientific references?
V=IR.
I thought we were talking about power, P = VI, where when you increase V and decrease I in the same proportions, P remains the same. Isn't this usually the case?
No, it uses more copper. It's a case of one live, a neutral, and a ground, or two lives, a neutral, and a ground, all to carry the same amount of current. It used to be considered acceptable to wire the neutral and the ground together (NEMA 10-30), but no longer is (hence NEMA 14-30 is the new standard).
Ok, I see my error there. I never expected a 240V equipment to care about neutral if it had the expected ground (for safety), but that doesn't seem to be the case, although I have no clue as why. Must be to optimize the current consumption by profiting from this design directly in the electric motors?
I'm saying that our current system restricts you for no good reason. It's not "needlessly complex" in the least to have all of your sockets on 240V and have plugs with a lower current plug into higher-current sockets. What's needlessly complex is having a mix of incompatible sockets with two different power delivery configurations (single phase and split phase).
Yes, it would be simpler to design a grid on 240V in that case. The case for 120V would be safety.
Our body's resistance is roughly the same regardless of the voltage applied, so we'd get a worse kick from 240V than from 120V.
Since the huge majority of appliances don't need the extra volts, we have a boost in safety with a 120V grid.
It's amazing! 9 out of 10 people they stop after spotting them with the rod actually have bombs! What that says about the people that aren't being stopped is a mystery...
Oh, don't get me started on 120V. It's a huge waste of copper and power
Huge? Why? Can you point to some scientific references?
that we only went with because it's easier to make a 120V incandescent lightbulb than a 240V. :P
I'll have to take your word on that.
Anyway, you won't find a 120V 30A outlet in your average US home because we run most higher power devices at (the more reasonable) 240V (although we do it in a weird way -- split phase).
There's a reason for using split phase and it's not that weird. It creates two independent single phase circuits providing 110V and it can be used as one 220V circuit by replacing the neuter with the other phase. It saves copper, as opposed to your very first statement.
But even in that case, it'd still be useful. Homes usually have both 30A and 50A 240V sockets. But you can't plug a 30A into a 50A without an adapter, even though there's no reason why you shouldn't be able to. And you should also be able to plug a 120V plug into one phase of your 240V/30A or 50A sockets, but you can't do that without an adapter, either.
It's not clear to me what you're asking. You want a socket to provide both 110V and 220V and be able to plug such devices according to their needs? It seems a bit complicated in my opinion.
Usually when the system is defined to be 110V for most appliances and 220V for exceptions such as the dryer or maybe the stove, one gets 110V outlets everywhere and exception outlets where such appliances are supposed to be installed. Why would you want to make each and every outlet more complex to be able to support exceptional cases?
OTOH 110v 20A is a lot more common, and you can plug a 110v 15A plug into a 110v 20A outlet;
Only if you have a special 15A/20A hybrid outlet. The standard NEMA 5-20R doesn't have a T-shaped slot; it only has the horizontal on that side. This is the US trying to correct a weakness in our outlet system after it was discovered; it's a bandaid on the problem of having entirely different pin layouts on each socket. The Australian standard of having different pin *sizes* deals with this problem automatically.
You lost me. If I understand the T-shaped slot correctly, it's supposed to be a higher current slot that still allows you to plug lower current devices in it. What's the problem with that?
You can still do it if you want, it's just aesthetically arguable to have a huge box (as opposed to this or this), protruding of the wall with plugs all around, so I'm not sure that's the point.
It's huge size is certainly not an asset. Just browse other comments for examples, especially how it becomes an unwanted volume for laptop power bricks.
The only pro might be the extra security. I won't argue that, but I can argue if all that security really pays for the inconveniences or if it's just redundant.
Don't forget paper cuts! These are deadly!
I'm kind of surprised the place is still standing... I really sort of thought it would have burned down from an electrical fire by now.
That's because not being able to run a microwave along with a stove without tripping a breaker usually means that you have working breakers, not just that your wiring is inconvenient.
AFAIK, in most cases the difference between ground and neutral is irrelevant. It's a nice security measure, but it hardly means that your building will catch fire just because it's missing.
Here's what I think of the mentioned models:
When a socket is for higher current, it doesn't mean that it always provides such high current. It means that it supports it if the device you plug requires it. So when you have a low current plug (and hence a low current device), you might appreciate being able to plug it into a higher current socket.
The opposite is not true, tho. When you have a low current socket, it can melt and cause fire if you try to use it with high current devices. Of course, your breakers will probably disarm first, but in any case you don't want to try that.
That's why both directions is not an option, while one direction is.
Doesn't seem like he missed the point. By insisting on the pi detail, he is pointing out that imprecision between Math and real world has so little to do with the speedometer's innacuracy that it's irrelevant. And this, my friend, has nothing to do with things working in theory and not in practice.
The theory and the practice in this case are: if we know the radius, we know the speed. Problem is, we don't know the radius, so let's take the minimum standard and add a buffer.
That's where the real innacuracy probably comes from.
How about measuring the speed with which your wheel turns and extrapolating to the speed of the vehicle?
If I understand it correctly, the rotation of magnets generates electric current directly related to the angular velocity, which is then extrapolated to the velocity.
I believe it's something like min wheel radius x pi x angular velocity, the later being represented by the electric current.
It's not the same as bicycle speedometers, which really measure the time between two passes of the magnet during one wheel revolution.
What's your point?
Just that Linux doesn't have that.
I no longer use Windows at home. I like my Ubuntu, even though:
- it took me days to have it working on my tablet PC, installing driver versions that are not part of the distro
- takes me a custom script to rebuild drivers at every kernel update (because of first point)
- had to disable ACPI (blame Microsoft and HP if you want: its ACPI works according to Vista, not according to standard. So Linux kernels refuse to talk to it)
- takes a rmmod ehci_hcd/modprobe ehci_hcd to make my webcam work 1 out of 3 times
- doesn't support screen rotation
- battery life sucks (yes, I've been through all battery optimization tricks)
- and so on.
I still have fun with it and enjoy having paid in time rather than money to have whatever is working, working. I also love to be able to workaround problems with scripting and shell commands, which is not the strongest quality of Windows.
Yes, M$ is evil. Yes, M$ pays to have their drivers. Yes, Linux is great. But... it still lags in support for new hardware. That's the point.
Hopefully the shit it got wasn't false positives...
Actually, that should have been modded -1 Stupid. It was a honest mistake :P
Sure, remember the GM x Microsoft Analogy War?
We'll see about holding up on court when the Psystar affair ends.
Meanwhile, Apple's business seems to be selling complete solutions, hardware and software. I think they made it perfectly clear by now that they are NOT interested in allowing people to play with half of their solution and mix it with another half that they DON'T WANT TO SUPPORT.
They want their full solutions to just work and gain such a fame. When people insist in using halves with unsupported parts, it creates a new type of user that will be bound to have a worse user experience and potentially hurt the "just work" image they want to nurture.
So it seems reasonable to assume that they DON'T WANT you as a customer unless you buy the product as THEY want to SELL IT.
See? I can use caps too ; )
Since I can't mod, I'll just say it: thank you sir for your sober analysis.
What I understood from the article is that the program uses something in the likes of a mix of neural networks with fuzzy logic to detect what the normal behavior of a certain software is.
Example: it runs for a certain number of iterations in "analyze" mode, creating rules that link input with output.
Then it's triggered when an abnormal behavior occurs and tries to find a patch to the software so that the cause of abnormal behavior becomes normal without impacting the previously detected rules.
Example: (obviously simplified) It identifies a rule that links typing something in "url" field and the following actions:
1. identify protocol
2. identify host name
3. resolve host name
4. connect to host
5. ask for path
6. parse/show response
If at some point something is written in "url" field but breaks the above rules, it will create a patch that will force this faulty input to behave nicely.
If my understanding is correct, there's no magic or perpetual motion involved. It's a matter of using learning followed by inference. Maybe both at the same time.
Maybe the Bayesian SPAM Filtering algorithm could be used as a real life far-cousin example?
I wonder why they went with C# on Linux?
I can only guess... Linux may be the easiest way to get a free OS and tweak it to your needs, since it already runs on everything from your generic PC to your electric toothbrush, then they probably held the opinion that C# was the current fashion in programming languages.
Thank you for the informative post, Steve ; )
Every time I heard about Tesla Motors I thought I was hearing about that scam company Spark EV.
So I decided to post this clarification in case any other incautious reader did the same mistake.
And thus another chair is thrown in Redmond.
That kind of people was bound to think that of EFF regardless of this new effort. Just look at all the mischief they do ; )
Your premise seems to be unfunded.
For everybody else's luck, EFF was impartial and also added a same silence attempt from the same-sex marriage opponents.