h a l f b a k e r y
A few slices short of a loaf.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
a grid of millions of capacitors each one individually and
digitally controlled to release or accumulate power, can
be strung along and around the mains lines, built into
and roofs of any existing or new structures, and releasing
power in the correct direction, quantity
and phase so as
emulate the AC. Then gradually lower the power needs
peak hours to use the accumulated power from off-peak.
Until the large utility companies buy into the technology
buy it off), small companies can use it to accumulate off
night-time power, or to store night-time wind or marine
current energy, or to run an off grid solar power station
[Edited - the following section was added]:
Sets of capacitors are set to be connected or
disconnected via controlled switching, so that they are
charged during different
times of the AC cycle, with some charging from the
current one way, and others set to be charged when the
current is the other way. When needed, they are then
discharged in sequence in a controlled manner so as to
emulate the AC current on the mains.
The control system could be distributed, or made central
for every group of capacitors, using PLCs, distributed
DSCs and other advanced low cost industrial control
Efficiency of regular capacitors can be extremely high,
with almost no loss of power or energy (see link)
An industrially manufactured product in these lines can
presumably cost much less than comparable battery
Raising capacitor efficiency (reducing energy loss)
[pashute, Oct 01 2014]
AC power storage
shameless self promotion [xaviergisz, Oct 07 2014]
||So these are magic capacitors which work in alternating current then?
And they're cheaper than normal capacitors but don't have any charge leakage?
||Capacitors are hellishly inefficient at storage. An order of
magnitude worse than batteries. Supercapacitors are better,
but then you are limited to low voltage and the losses
associated with voltage conversion AC/DC problems.
Batteries are expensive
and less efficient than pumped storage. Which is why we use
||I wouldn't worry about charge leakage much but, while caps are really really fast and very robust and long lasting, they aren't very efficient. Also they're rather expensive per kWh in terms of both cost and size/weight.
||In terms of energy conversion efficiency, caps can
be pretty good. On an AC system, to charge a cap,
just connect it between hot and neutral just as
the voltage crosses 0 then disconnect it when the
voltage reaches the peak. To put the power back
on the grid, connect it when the voltage is at it's
peak (matching the voltage on the cap), and let
the cap dump its charge until the voltage reaches
0 before disconnecting it. With a large number of
caps you could connect one (or more) each 60Hz
cycle to have fairly fine control of how much
power is being added or removed from the grid.
||This results in a very simple control circuit for each
capacitor, but the problem is that you need to
have a whole lot of capacitors. Say for example
that during a typical day you want charge for 6
hours, discharge for 6 hours and be idle for 12
hours. To accept charge for 6 hours, you need
one cap for each cycle = 60Hz * 60s/min * 60min/hr
* 6hr ~= 1.3 million. There needs to be a high
voltage switch with low leakage for each cap and
each needs to be able to handle the peak current
that an equivalent battery system would use. I
suspect those 1.3million transistors will be quite a
bit more expensive than the electronics to
convert between AC to DC and back for a battery
||To benefit from off-peak electricity, simply buy a long
extension lead and find a like-minded friend who
lives 180° away from you.
||great, now I need a friend in the mid pacific who's also on
||Well, maybe you and a friend could share a 230V donation from the rest of the world?
||Hmmm.....aha, reversible lightning.
||[Scad mientist] clarified what I thought I had described in
enough detail. No magic here [Loris], but actually some
quite simple electronics. Capacitors can be dirt cheap.
Regular capacitors, and that's what I'm talking about.
Yes they take up much space, but if distributed as I
described it can be practical.
||The cost of control (transistors) can be lowered by using
PLCs, extreme low-cost DSCs, and other industrial control
||Of course to get this going a serious investment would be
needed, but that doesn't mean the idea is not worth its
||And about capacitor efficiency - it can be set to almost
100%, with any regular capacitor. See link.
||I'm editing the idea for clarity.
||I'm not saying it won't be possible to use capacitors to store useful amounts of energy in future, if
better capacitor technologies come along.
||However, distributing current technology capacitors like you propose just has ridiculous issues.
||The specific energy capacity of capacitors compared to batteries is a joke. They're great for power -
they can dump their energy quickly - but they can't hold much energy per unit weight.
Large, cheap capacitors leak like sieves.
Controlling circuits - well, you seem to have missed the bit where scad mientist pointed out how
expensive even his simple system would likely be. I'll get back to this later.
||But hey, let's suppose for the sake of argument that we get a really efficient energy storage system.
Why the hell would we run it along power-lines? I can imagine that this might be handy for everyone
to have a bit of storage in their property.
If it's really small, we could have it all in one compact unit. Fine. Spread out through the walls and
roof? Nope. We've already decided it holds a lot of juice, so ... well, maybe you could mull on why
some electrical devices have warnings about opening them up even when the power is off (hint -
they're the ones with big capacitors in).
Alternatively, what if they're cheap and efficient, but really large? Wouldn't it make more sense to
have them all in one facility? What do you gain from strewing them along the power lines? I mean,
other than disadvantages in maintenance, control, security and safety.
||Now, controlling circuits. You explicitly want each one to be controlled independently - fine, but
that's going to cost extra. Both financially and in efficiency. Does each one have a clock so it knows
when to charge or discharge? Let's assume not. So how do you propose to ensure that they play
nicely with the rest of the power system? A side-channel? Or were you hoping to communicate with
them via mains frequency changes? That's not going to be easy when every single other capacitor
you've installed is investing its energy fighting your control.
||Are you sure I have not given adequate answers to all your
||It is switched off the mains with a PLC. Not fighting with
anyone for anything. Modern switching can run dirt cheap
for millions of switches. And take a minute amount of
power to run.
||Simple standard modern capacitors have almost zero
leakage current when switched off from the circuit, and
storing electricity, mind you, if you have not heard, is one
of the major challenges that any electric company has, let
alone when managing a distributed (renewable energy) grid.
So this could be a better solution than CAES or RESS. No big
capacitors, all are relatively small and harmless (say
standard 50 volts each, 2200mf caps).
||I will dispute your claim of dirt cheap. While it is
true that you can get millions of transistors for less
than the cost of a bag of topsoil, those are on a
VLSI chip and can't handle anywhere near the
voltage or current you would need. If you get
switches big enough, but only use them for 16ms
out of each day, you'd be better off making an
inverter with switches that are 1000x larger, but
are used for many hours each day.
||I started doing the calculations to see what switch
you would need for a system with a million 2200 uf
caps, compared to a inverter system with lead-
acid battery bank with similar power and capacity,
but then decided I didn't have that much time to
burn right now.
||Now I would not be hugely surprised if someone
actually did (or has already) found a cost effective
way to build capacitors designed for stationary
operation. The key would be making them ultra-
cheap and not worrying about weight or size. But
even if that worked out, using them in an ultra-
distributed system like this won't be as cost
effective as using a fairly standard inverter but
enhanced to handle the higher voltage range of a
capacitor compared to a battery.
||So Bath County Pumped storage can run for about 16
hours on a full reservoir. At 3003 MW. That's
||Total cost 1.6 billion in 1985. About 3.5 billion now.
||Capacitor: 75 Joules per $
Car battery: 170,000 Joules per $
Pumped Storage: 50469 Joules per $
||I'm surprised, and writing a letter regarding car batteries
to the government.
||I'll probably get a letter back regarding longevity,
charging efficiency, and there not being 103 million spare
||isn't charging low voltage batteries pretty horrific
efficiency wise? Inverting it afterward? hmm, web
suggests transformer rectifier can be 95%, lead acid
charge/discharge can be 60-95% and inverters 94-96%.
||So 1,700,000 in, assuming
Transform/rectify> charge> discharge> invert at
95>90>90>95% gives 1,700,000> 1,615,000>1,453,500>
1,308150> 1,242,743J. about 73% of the original.
||Pumped storage is up in the 80% region. Still ballpark.
You get some free when it rains too, if you have
||I think they should be coupling power stations physically.
i.e. Fuel>Steam>turbine> pump. Then let it out when you
need. Cuts out steps Fuel> steam> turbine> generator>
transformer> transmit> transformer> pump.
||So we conclude that pumped storage facilities are less economical than lead-acid batteries for starting your car in the morning?
||//So we conclude that pumped storage facilities are less
economical than lead-acid batteries for starting your car in
||And people say we're wasting our time.
||//Are you sure I have not given adequate answers to all your questions?//
||//It is switched off the mains with a PLC. Not fighting with anyone for anything.//
I don't think you understand the issue.
How are you telling the capacitors to store or release their energy? There are basically three options.
1) By timer.
2) Through the mains. Changing either the amplitude or frequency of the AC.
3) Through a side-channel.
||(1) means as many clocks as you have capacitors, and is not responsive to supply and demand. (3) means you need all the infrastructure of the side-channel, and all that entails. (2) is how load is already regulated to some extent, so we know that method works for a decent fraction of power -sinking- devices (apparently some behave in the opposite manner to what you'd want). However, if you have millions of very relatively stupid power -sources-, I think there might be issues.
||And you haven't explained why you want to string them along power lines, or in peoples walls or roofs despite all the disadvantages of doing that.
||isn't the issue more to do with capacitors being ideal for
very short term storage/release <1 second. And the storage
demand being on a different scale >1 min. Then add in the
capacity issues. The occasional computer controlled
capacitor bank could be useful for local conditioning, say at
the sub-station level. Where grid spikes could be absorbed
and minor local brown-outs smoothed.