4 ms·

Packetisation of energy is an idea that's been bubbling away in my mind for a number of years, and I'm convinced that is the way of the future. It revolutionis

by femto 3y ago

Packetisation of energy is an idea that's been bubbling away in my mind for a number of years, and I'm convinced that is the way of the future. It revolutionised data and in my opinion will (eventually) revolutionise energy, heralding a global "Internet for energy". Imagine if sources could verify the link before sending each energy packet: no more electrocutions, electrical fires or damaged equipment. Imagine if in a domestic situation each energy packet was below the threshold that could damage a person? Electrical installations could become inherently safe, also meaning no need for specialist installers. I get your point about universal time, EM waves and relativity. It doesn't stop data networks though, as packets include synchronisation, effectively establishing a local time for each link. It's fascinating thinking about the analogies between data networks and power networks. What would synchronisation look like for an energy packet? What would a buffer look like, in case packets have to be multiplexed? What are the implications of being unable to copy energy packets? What is an energy packet? So many interesting questions.

flangola7 3y ago
Isn't that just IoT with more steps?
- actionfromafar 3y ago
  Yeah, like Dropbox is ftp with extra steps.
ur-whale 3y ago
> I'm convinced that is the way of the future. What effect would "turning the hose on an d off" have on efficiency?
- devb 3y ago
  Did you just invent alternating current?
- femto 3y ago
  They are the sorts of questions that make it such an interesting problem. Maybe it's not as simple as turning it on and off? Maybe there is something to be drawn from the world of adiabatic logic? Maybe the answer is some sort of resonant system or waveform which "sloshes" energy from place to place instead of a hard on/off (yeah, related to AC). Maybe there is a system to smartly manage the boundaries between packets, to avoid unnecessary switching? I don't think we have the answers (yet), but I think there will be benefits if we can find the answers. For example, we could truly leave the current generator/consumer model behind: just like the web enabled everyone to become a publisher, everyone could become a generator in their own right. No doubt we will then be talking about the Goolalisation of the energy network, where a theoretically distributed system becomes dominated by a small number of monopolies and people talk about how it's not worth running your own generator since Powersoft refuses to accept energy packets from smaller generators.
  - ilyt 3y ago
    > Maybe the answer is some sort of resonant system or waveform which "sloshes" energy from place to place instead of a hard on/off (yeah, related to AC). Most definitely not. You want to limit is as much as possible, as the reactive power (the "bounce back") still goes thru the resistance of the wire, generating losses. In fact big consumers often get charged when their power factor (the amount of energy "received" vs bounced back) is below a certain value.
- actionfromafar 3y ago
  Isn’t that AC current? So, pretty great, probably. I like the idea.
mikewarot 3y ago
Wow... that's an interesting idea Taking a system carefully designed to maximize power transmission across the country, and only sending power in bursts seems rather an odd design choice. You'd need to beef up all the transmission lines and transmission gear by a few orders of magnitude in current and/or voltage to make something like that possible. I'm not sure there's enough steel, aluminum, copper and insulators available to make that happen. How would you store gigajoule "packets" of energy?
- actionfromafar 3y ago
  Maybe a better place for this system is within buildings.
ilyt 3y ago
I think the problem is the physics here. In theory, you could just send higher frequency AC or pulsed DC signal thru the wire (to get to the "one pulse is not all that dangerous" level), and with a bunch of ADCs discover every anomaly VERY quickly. But, now you're pouring high frequency signal into huge antenna, which means you probably need to change cabling to be twisted pair and recabling house is HUGE issue. And you have higher losses which means you'd realistically have to have power converter in every house for it as you couldn't transfer power at long ranges if it ran on several kHz > Imagine if sources could verify the link before sending each energy packet: no more electrocutions, electrical fires or damaged equipment. Imagine if in a domestic situation each energy packet was below the threshold that could damage a person? Electrical installations could become inherently safe, also meaning no need for specialist installers. If we assume sending device and receiving device both communicate how big energy packet would be and fail safe if it isn't, then sure, any interruption would be detected quickly, but it does nothing for cases where end device after the "power interface" has problems. Also at least in US there is like 400 electrocution deaths per year, with only 20% from actual wiring [1], so it is a fight entirely not worth fighting, for residential at least. Where I think it could be interesting is in high reliability equipment, where for example you might have power supply shared between a bunch of subsystems and you don't want failure of one component to cause brownout of whole thing. Sure, fuses do that but they are not exactly very elastic and don't send the signal about nature of the failure upstream. A "power bus", where on top of delivering power subsystems can signal failure condition, or get signals (like "switch into power save mode, battery is low") might be interesting but if USB-PD is any indication it is FAR easier to make it into overengineered mess.. - [1] https://www.nickleelectrical.com/electrical-safety-statistics/ https://www.nickleelectrical.com/electrical-safety-statistic...
Freestyler_3 3y ago
This opens up the grid for trading (PV) electricity directly.

by femto 3y ago

flangola7 3y ago

Isn't that just IoT with more steps?

actionfromafar 3y ago
Yeah, like Dropbox is ftp with extra steps.

ur-whale 3y ago

> I'm convinced that is the way of the future. What effect would "turning the hose on an d off" have on efficiency?

devb 3y ago
Did you just invent alternating current?
femto 3y ago
They are the sorts of questions that make it such an interesting problem. Maybe it's not as simple as turning it on and off? Maybe there is something to be drawn from the world of adiabatic logic? Maybe the answer is some sort of resonant system or waveform which "sloshes" energy from place to place instead of a hard on/off (yeah, related to AC). Maybe there is a system to smartly manage the boundaries between packets, to avoid unnecessary switching? I don't think we have the answers (yet), but I think there will be benefits if we can find the answers. For example, we could truly leave the current generator/consumer model behind: just like the web enabled everyone to become a publisher, everyone could become a generator in their own right. No doubt we will then be talking about the Goolalisation of the energy network, where a theoretically distributed system becomes dominated by a small number of monopolies and people talk about how it's not worth running your own generator since Powersoft refuses to accept energy packets from smaller generators.
- ilyt 3y ago
  > Maybe the answer is some sort of resonant system or waveform which "sloshes" energy from place to place instead of a hard on/off (yeah, related to AC). Most definitely not. You want to limit is as much as possible, as the reactive power (the "bounce back") still goes thru the resistance of the wire, generating losses. In fact big consumers often get charged when their power factor (the amount of energy "received" vs bounced back) is below a certain value.
actionfromafar 3y ago
Isn’t that AC current? So, pretty great, probably. I like the idea.

mikewarot 3y ago

Wow... that's an interesting idea Taking a system carefully designed to maximize power transmission across the country, and only sending power in bursts seems rather an odd design choice. You'd need to beef up all the transmission lines and transmission gear by a few orders of magnitude in current and/or voltage to make something like that possible. I'm not sure there's enough steel, aluminum, copper and insulators available to make that happen. How would you store gigajoule "packets" of energy?

actionfromafar 3y ago
Maybe a better place for this system is within buildings.

ilyt 3y ago

I think the problem is the physics here. In theory, you could just send higher frequency AC or pulsed DC signal thru the wire (to get to the "one pulse is not all that dangerous" level), and with a bunch of ADCs discover every anomaly VERY quickly. But, now you're pouring high frequency signal into huge antenna, which means you probably need to change cabling to be twisted pair and recabling house is HUGE issue. And you have higher losses which means you'd realistically have to have power converter in every house for it as you couldn't transfer power at long ranges if it ran on several kHz > Imagine if sources could verify the link before sending each energy packet: no more electrocutions, electrical fires or damaged equipment. Imagine if in a domestic situation each energy packet was below the threshold that could damage a person? Electrical installations could become inherently safe, also meaning no need for specialist installers. If we assume sending device and receiving device both communicate how big energy packet would be and fail safe if it isn't, then sure, any interruption would be detected quickly, but it does nothing for cases where end device after the "power interface" has problems. Also at least in US there is like 400 electrocution deaths per year, with only 20% from actual wiring [1], so it is a fight entirely not worth fighting, for residential at least. Where I think it could be interesting is in high reliability equipment, where for example you might have power supply shared between a bunch of subsystems and you don't want failure of one component to cause brownout of whole thing. Sure, fuses do that but they are not exactly very elastic and don't send the signal about nature of the failure upstream. A "power bus", where on top of delivering power subsystems can signal failure condition, or get signals (like "switch into power save mode, battery is low") might be interesting but if USB-PD is any indication it is FAR easier to make it into overengineered mess.. - [1] https://www.nickleelectrical.com/electrical-safety-statistics/ https://www.nickleelectrical.com/electrical-safety-statistic...

Freestyler_3 3y ago

This opens up the grid for trading (PV) electricity directly.