That particular storm still caused catastrophic failure of a 500kV 1200MVA transformer at the Salem nuclear power plant in New Jersey. They happened to have a suitable replacement available which was originally intended for another power plant so they were only offline for a few weeks.
Unfortunately, that's not true. The output is AC and the input-referred current will reflect it, i.e. will be quite high when the output reaches its highest (340V). Since the input source has an equivalent internal resistance of 10Ohm and would droop too much, some bulk capacitance is needed to provide these 120Hz current surges.
I should have been more specific but I agree.
There are other ways to reduce input ripple current besides capacitive decoupling. A PFC stage operating in continuous conduction mode would help and may be needed anyway because of the voltage drop across the 10 ohm source resistance. I wonder if such would take up less space than extra bulk input capacitance.
I think the largest problems are physical construction and heat distribution. I could certainly design and build such an inverter but the construction methods needed to achieve that form factor are outside of my prototyping capabilities. This is one of those designs where the power components need to be laid out first with an eye toward thermal management which is integrated as part of the case. Forced air cooling would help a lot but comes at a significant reduction in reliability and operating life so I would not consider it a viable option.
The input is DC so there is no need for low frequency input decoupling. Large value input and output ceramic capacitors are common in high density DC to DC converters although they can make frequency compensation tricky because of their low ESR. I doubt that would be a problem here.
I think the hard part in this case will be the physical design and construction to stay within the thermal constraints. Brick style power converters have power densities and efficiencies in this range but are designed to be attached to a metal surface for heat dissipation and Google's requirements do not allow for that.
Active forced air cooling would help a lot but also make long term reliability poor.
Motorola couldn't manufacture enough of the 68K CPUs, so Apple set up an alliance with IBM and Motorola (AIM). The first generation of the PowerPC was fast and easily manufactured.
I always wondered about this. At the time we had problems getting *anything* reliably produced out of Motorola and the standard procedure became to only design in parts which were available for shipping immediately and sometimes even that was not good enough. A majority of their products appeared to be on eternal allocation and unavailable. So I had a good chuckle about Motorola producing PowerPC chips for Apple and predicted that relying on Motorola would be a problem for them.
The later 68k processors had at least two serious problems; the address registers only supported a subset of the instruction set limiting their usefulness but even worse, the double indirect addressing mode really clobbered the high performance pipeline designs needed to compete with the Pentium and later which is when Motorola threw in the towel.
That was not unusual as up until recently this occurred for all major processor transitions. Later processors were further pipelined to take advantage of newer processes and higher clock rates so the comparison between generations "at the same clock speed" is often misleading because the earlier processors would never be able to operate at significantly faster clock rates anyway. This is especially the case when access times of the register file and first level cache are taken into account.
If one grabs an RPi, sees the specs and tries to run a full blown desktop with heavyweight programs on it, it's going to suck. But that's not the fault of the RPi, that's the fault of the user.
How would it compare to a Pentium III at the same clock speed? They were commonly available with the same amount of RAM.
A women is confronted by a big, strong, stranger. She does not know what he is planning, and she is cautious. Getting away from him is not possible. They are in a room and he is standing in front of the only way out, or she is in a wheelchair - whatever. Leaving the area is not an option.
So now he starts to do things she does not like. He asks her for money. She can try to talk him out of it, just like we argue for lower taxes, and maybe it will work. If it does not, and she gets outvoted, she will probably choose to give in to him instead of getting into a fight to the death over ten dollars. You would probably choose to pay your taxes rather then have police arrive to throw you in jail.
Maybe this big man demands some other things, other minor assaults on this woman's dignity. When should she claw at his eyes or shove her ballpoint pen in his throat? When he tries to force her to kiss him? Tries to force her to let him touch her? Tries to force her to have sex with him?
Those are questions that each woman has to answer for herself. There is one situation, though, where I tell the women to fight to the death. That is when the man pulls out a pair of handcuffs and says, "Come on, I promise I won't hurt you, this is just so you won't flail around and hurt either of us by accident. Come on, I just want to talk, get in the van and let me handcuff you to this eyebolt here, and I promise I won't touch you. I'm not asking you to put on a gag or anything, and since you can still scream for help, you know you'll be safe. Come on, I've got a full bar in here, and color TV, and air-conditioning, great stereo, come on, just put on the cuffs."
I tell women that if that ever happens, maybe the man is telling the truth, and maybe after talking to her for a while he will let her go and she will have had a good time drinking champagne and listening to music. But if she gets in the van and puts her wrists in the handcuffs, she has just given up her future ability to fight, and now it is too late.
How do you spot the precise point where a society is standing at the back of the van and the State has the handcuffs out?
Even the copies that the police lost when they seized your equipment?
Funny, here are 9 disks/books/SD cards labeled "keys" from 1 to 10 and number 8 is missing which makes sense if it was out when the police executed their search and lost it.
The same applies in the US. The Miranda waring "anything you say may be used against you in court" means literally that and anything you say when used in your defense is hearsay. Obviously this is not widely advertised.
Law enforcement memories are perfect when recalling what you said to use against you even when they lie which is just another reason to say nothing. The FBI has perfected this technique in connection with interrogations and local law enforcement learns from the best.
The same places that lack power generation capacity also lack baseline power generation capability. Exporting excess dynamic power to places *with* base load capacity will just make their base load capacity less economical.
This problem of electricity rates going negative has cropped up in Texas with wind power as well.
The larger issue is that dynamic power like wind and solar make operation of power plants capable of base load production uneconomical if the later are required to trim production shifting capacity to the dynamic power sources. That make grid stability worse.
My suspicion is that Germany has ended up outsourcing their base load requirements to France making their renewable electricity numbers look better than they really are.
Self defense. It IS still legal to shoot someone who is threatening to do the same to yourself. Fortunately it is rarely necessary. Simply being seen to have a weapoin is enough to diffuse most situations.
Do you have personal experience with this? Are there any data on that? How many lives are saved per year by the threat of gun violence?
I have 3 times in just over 10 years but never had to actually draw. Just the look of being ready to defend was enough to stave off a more serious confrontation. The first incident convinced me to switch to IWB carry instead of knapsack carry.
Defensive gun use is difficult to measure simply because of massive under reporting. The defender runs the risk of legal entanglements admitting to brandishing a firearm where they will be second guessed for no personal gain when nobody is actually injured which is the vast majority of cases.
The problem with these stats is that they lump everyone in together. Gun-related suicides, for instance, aren't even worth considering in these stats; people will kill themselves one way or another, guns just make it easier and faster. Homicides are the important stat. However, even here most gun-related homicides in this country are likely because of gang-related violence. If you're not a gang member or other violent criminal, you have much less to worry about. So how about some stats which exclude gang members? It'd be interesting to see how the US ranks there.
It is worse than this. The US counts all homicides involving a firearm including justified shootings by civilians and law enforcement. In the case of the UK, a homicide is only counted if there is a conviction. The UK plays other games with their crime statistics like counting multiple individual robberies or burglaries as one when they involve the same perpetrators in the same area and time.
It's not gun proliferation that causes America's problems with crime, violence, and poverty. Somehow liberals seem to think that if we just get rid of the guns, we'll suddenly turn into a gigantic version of Sweden or Norway. It doesn't work that way. Those countries are ultra-safe because of their culture. We don't have that kind of culture.
I would not mind knowing how firearms lead to a proportionally high rate of knife, blunt weapon, and strangulation attacks in the US.
I generally recommend against USB test instruments because of their shortcomings but for a USB based oscilloscope, I would at least get one which supports network analysis which may be particularly useful for students:
Think that a motorola 68000, way back in the day was better than the old 286s it compared to. Imagine that the 68000 took off instead of the 286 - if MS and IBM had built DOS for 68000 instead of x86... today we'd be in pretty much the same position but with a different chipset. But it would be faster and cheaper and more efficient.
While 68000 had a more orthogonal instruction set, starting with the 68020 it also included double indirect addressing which is difficult to implement in a super-scalar out-of-order design.
Wasn't the theory behind the Rise Of Embedded Java that companies would move between suppliers fairly rapidly (which they do) in order to keep BoM to a minimum and that those moves would involve disruptive changes of CPU architecture(which appears to be much more rarely the case) thus driving them to write as much as possible in Java for easy porting?
I think it was a marketing hallucination. They were advertising this concept to replace 8 and 16 bit microcontrollers while JAVA had no support for unsigned arithmetic or the low level access needed for embedded applications.
Slashdot Mirror
That particular storm still caused catastrophic failure of a 500kV 1200MVA transformer at the Salem nuclear power plant in New Jersey. They happened to have a suitable replacement available which was originally intended for another power plant so they were only offline for a few weeks.
I should have been more specific but I agree.
There are other ways to reduce input ripple current besides capacitive decoupling. A PFC stage operating in continuous conduction mode would help and may be needed anyway because of the voltage drop across the 10 ohm source resistance. I wonder if such would take up less space than extra bulk input capacitance.
I think the largest problems are physical construction and heat distribution. I could certainly design and build such an inverter but the construction methods needed to achieve that form factor are outside of my prototyping capabilities. This is one of those designs where the power components need to be laid out first with an eye toward thermal management which is integrated as part of the case. Forced air cooling would help a lot but comes at a significant reduction in reliability and operating life so I would not consider it a viable option.
The input is DC so there is no need for low frequency input decoupling. Large value input and output ceramic capacitors are common in high density DC to DC converters although they can make frequency compensation tricky because of their low ESR. I doubt that would be a problem here.
I think the hard part in this case will be the physical design and construction to stay within the thermal constraints. Brick style power converters have power densities and efficiencies in this range but are designed to be attached to a metal surface for heat dissipation and Google's requirements do not allow for that.
Active forced air cooling would help a lot but also make long term reliability poor.
The ones which accept 48 volt DC inputs are better than 95% efficient but physically larger than Google's requirements.
I am not sure if it is the same thing but I have been using the craptastic Classic Theme Restorer which also fixed some other UI problems.
I always wondered about this. At the time we had problems getting *anything* reliably produced out of Motorola and the standard procedure became to only design in parts which were available for shipping immediately and sometimes even that was not good enough. A majority of their products appeared to be on eternal allocation and unavailable. So I had a good chuckle about Motorola producing PowerPC chips for Apple and predicted that relying on Motorola would be a problem for them.
The later 68k processors had at least two serious problems; the address registers only supported a subset of the instruction set limiting their usefulness but even worse, the double indirect addressing mode really clobbered the high performance pipeline designs needed to compete with the Pentium and later which is when Motorola threw in the towel.
That was not unusual as up until recently this occurred for all major processor transitions. Later processors were further pipelined to take advantage of newer processes and higher clock rates so the comparison between generations "at the same clock speed" is often misleading because the earlier processors would never be able to operate at significantly faster clock rates anyway. This is especially the case when access times of the register file and first level cache are taken into account.
They also ran Breakout, Super Breakout, and Photoshop.
I used to use my IPv6 tunnel to do this until AT&T blocked IPv6 (or at least protocol 41) unless they provide it themselves.
Never would hosting or sending data to NSA's server.
How would it compare to a Pentium III at the same clock speed? They were commonly available with the same amount of RAM.
I like the way John Ross put it:
A women is confronted by a big, strong, stranger. She does not know what he is planning, and she is cautious. Getting away from him is not possible. They are in a room and he is standing in front of the only way out, or she is in a wheelchair - whatever. Leaving the area is not an option.
So now he starts to do things she does not like. He asks her for money. She can try to talk him out of it, just like we argue for lower taxes, and maybe it will work. If it does not, and she gets outvoted, she will probably choose to give in to him instead of getting into a fight to the death over ten dollars. You would probably choose to pay your taxes rather then have police arrive to throw you in jail.
Maybe this big man demands some other things, other minor assaults on this woman's dignity. When should she claw at his eyes or shove her ballpoint pen in his throat? When he tries to force her to kiss him? Tries to force her to let him touch her? Tries to force her to have sex with him?
Those are questions that each woman has to answer for herself. There is one situation, though, where I tell the women to fight to the death. That is when the man pulls out a pair of handcuffs and says, "Come on, I promise I won't hurt you, this is just so you won't flail around and hurt either of us by accident. Come on, I just want to talk, get in the van and let me handcuff you to this eyebolt here, and I promise I won't touch you. I'm not asking you to put on a gag or anything, and since you can still scream for help, you know you'll be safe. Come on, I've got a full bar in here, and color TV, and air-conditioning, great stereo, come on, just put on the cuffs."
I tell women that if that ever happens, maybe the man is telling the truth, and maybe after talking to her for a while he will let her go and she will have had a good time drinking champagne and listening to music. But if she gets in the van and puts her wrists in the handcuffs, she has just given up her future ability to fight, and now it is too late.
How do you spot the precise point where a society is standing at the back of the van and the State has the handcuffs out?
Even the copies that the police lost when they seized your equipment?
Funny, here are 9 disks/books/SD cards labeled "keys" from 1 to 10 and number 8 is missing which makes sense if it was out when the police executed their search and lost it.
The same applies in the US. The Miranda waring "anything you say may be used against you in court" means literally that and anything you say when used in your defense is hearsay. Obviously this is not widely advertised.
Law enforcement memories are perfect when recalling what you said to use against you even when they lie which is just another reason to say nothing. The FBI has perfected this technique in connection with interrogations and local law enforcement learns from the best.
So they trained the police dog to smell out child pornography on digital media. Nice.
The same places that lack power generation capacity also lack baseline power generation capability. Exporting excess dynamic power to places *with* base load capacity will just make their base load capacity less economical.
This problem of electricity rates going negative has cropped up in Texas with wind power as well.
The larger issue is that dynamic power like wind and solar make operation of power plants capable of base load production uneconomical if the later are required to trim production shifting capacity to the dynamic power sources. That make grid stability worse.
My suspicion is that Germany has ended up outsourcing their base load requirements to France making their renewable electricity numbers look better than they really are.
I have 3 times in just over 10 years but never had to actually draw. Just the look of being ready to defend was enough to stave off a more serious confrontation. The first incident convinced me to switch to IWB carry instead of knapsack carry.
Defensive gun use is difficult to measure simply because of massive under reporting. The defender runs the risk of legal entanglements admitting to brandishing a firearm where they will be second guessed for no personal gain when nobody is actually injured which is the vast majority of cases.
http://en.wikipedia.org/wiki/D...
It is worse than this. The US counts all homicides involving a firearm including justified shootings by civilians and law enforcement. In the case of the UK, a homicide is only counted if there is a conviction. The UK plays other games with their crime statistics like counting multiple individual robberies or burglaries as one when they involve the same perpetrators in the same area and time.
I would not mind knowing how firearms lead to a proportionally high rate of knife, blunt weapon, and strangulation attacks in the US.
I generally recommend against USB test instruments because of their shortcomings but for a USB based oscilloscope, I would at least get one which supports network analysis which may be particularly useful for students:
http://www.syscompdesign.com/C...
http://www.syscompdesign.com/C...
While 68000 had a more orthogonal instruction set, starting with the 68020 it also included double indirect addressing which is difficult to implement in a super-scalar out-of-order design.
It's fair because they say it is fair.
I think it was a marketing hallucination. They were advertising this concept to replace 8 and 16 bit microcontrollers while JAVA had no support for unsigned arithmetic or the low level access needed for embedded applications.