Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror

Comment: Re:Help me out here a little... (Score 1) 533

by savuporo (#49512149) Attached to: Utilities Battle Homeowners Over Solar Power

It is a nice solution in theory. The implementation details however would be very hairy and difficult. You have to agree to the protocol on what the anchor time for zero crossing is, you have to get power plants and large scale grid compensation systems to dance to that beat first. This in itself would be no small feat. And then somehow get all the small inverters GPS enabled, which places extra constraints on their installation locations etc.

And then, your entire power grid will have a single point of failure - GPS outage, or you will have to factor in redundancy protocols.

Comment: Re:Help me out here a little... (Score 2) 533

by savuporo (#49507265) Attached to: Utilities Battle Homeowners Over Solar Power

You do not understand error propagation.
Synchronizing at the point of connection does not mean that you are synchronizing with any particular "power plant", you are synchronizing with a relatively randomly distorted signal that gets its contributions in worst case from multiple neighborhood inverters that are also trying to sync and are all somewhat off, and nearby real world loads with different apparent power and other sources of noise.
 

Comment: Re:Help me out here a little... (Score 4, Informative) 533

by savuporo (#49505573) Attached to: Utilities Battle Homeowners Over Solar Power

Coordinating the power grid needs information, today the information carried for proper coordination ( AC waveform detecion by grid inverter ) is not sufficient to make things work at large scale.
Whether you transmit the coordination signals ( i.e. extra bits of information ) in-band ( over powerline, superimposed on the AC wave ) or out of band on a separate carrier is the question of implementation. Both in-band and out of band would have their upsides and downsides.
Fact of the matter is that todays grid is not built for that.

Simple case, assume there is a dozen or so high power solar installations in my neighborhood, delivering most of the peak power output. The requirement for grid-tie inverters is matching the phase within 1% of the 50/60hz waveform read at the connection point. The question is, _whose_ waveform is mine really following ? Is it the neighbors ? Which one is his following ? How does the error propagate and does it multiply ? That is the gist of the coordination problem.

Comment: Re:Help me out here a little... (Score 2) 533

by savuporo (#49505535) Attached to: Utilities Battle Homeowners Over Solar Power

I agree with most of your post, but i disagree with the assertion that grid "MUST be centralized". It must be well-coordinated, and centralization is one of the ways to achieve coordination, but not the only one.
Power grid currently is built with assumption of central coordination, that's why there are these growing pains with distributed generation.

Comment: Re:Help me out here a little... (Score 3, Informative) 533

by savuporo (#49505123) Attached to: Utilities Battle Homeowners Over Solar Power

Its not just frequency and voltage, there is phase and power factor, harmonics etc. Grid tie inverters are not simple pieces of equipment by any means - they try to synchronize and follow the grid power delivery by following grid input AC waveform at the point of connection, which is a limited bit of information and may not be fully in sync with macro-scale grid need at any given moment.
For a perfectly synched network you would have to have atomic clocks and low latency radio link network with each point of generation, that's obviously not going to happen, so hacks, power factor correction systems, extra reactive loads etc are and will have to be implemented.

Comment: Device drivers ? (Score 1) 161

by savuporo (#48785473) Attached to: Rust Programming Language Reaches 1.0 Alpha

Ok, if this is a systems programming language, where is the first RTOS kernel with all the necessary lowlevel bits and pieces to getting it running on a modest modren 32-bit MCU ? Say, any Cortex-M3 ? Device drivers for basics, register access ?
Because, it would be awesome to have all these theoretical safety guarantees and stuff, while programming hardware.
Is there even a cross-compiler ?

Comment: Ahead of NTSB (Score 2) 20

by savuporo (#48779487) Attached to: Virgin Galactic Test Flights To Restart This Year

Ok, here are claims made without a final conclusion by NTSB anywhere in sight. How the heck can they assume that NTSB does not come out and burn their entire design to ground ? How do they even assume NTSB investigation will conclude before end of the year ?

What a hubris.

Your design grandfather Rutan retired years ago, your chief designer left the company .. and you are still making these grandiose claims ?

Thus spake the master programmer: "When a program is being tested, it is too late to make design changes." -- Geoffrey James, "The Tao of Programming"

Working...