PoE pi

On a sunny day (Sun, 29 Jul 2018 22:51:34 +0300) it happened Tauno Voipio wrote in :

Be carefull, there will be resistors between those points.

Putting 48V on such an ethernet connector will smoke those resistors.

IIRC the logic in the POE supply tests for this.

You will need to replace connectors for special ones without those resistors if you just want to put a voltage on the cable, when not using an 'intelligent' controller.

No, its far easier especially in the downward voltage direction where a PWM chopper and a teeny choke and a capacitor will do the business.

Cheap semicinductors capable of extremely high speed switching have changed the power game forever.

On chip, a small ferrite choke and a capacitor replace a whole lotta stuff.

Oh dear.

I dont think you have. It all sounds pretty OK to me

It is possble and even likely that HVDC will replace AC grids. It already is used to interconnect different grids. In is inherently lower loss due to capacticance not being a feature. (charging up all that line to line and line to earth capacitance loses energy in the tranmsission line resistance, just as much as in phase currennt does).

But it has a problem: in a normal grid the rotating masses of all the gnetrators snd their turbibes mean that a sudden load change or generator loss can tap this store of spinning mass energy, and a voltage drop followed by a slowly falling frequency is a good signal to use to determine wahat extra power needs to be fed in.

That mecahnism's functionality needs replicating with a DC grid. Maybe big capacitors on the lines would provide some inherent storage.

Completely wrong.

Got nothing to do with compassess and everything to do with te fact that the capacitance to sea water via a few cm of insulation is WAY higher then the capacitance to ground of conductors strung 20 m up in the air.

That wouldnt matter if the cinductors had zero resistance - out of phase current is lossless current - but conductors HAVE resistanbce, and that menas there are losses - serious extra losses - in the cables.

If your original premise is sheer nonsense any facts you deduce from it will be nonsense too.

HVDC is more efficient for long hauls where AC cable losses start to become significant.

HVDC makes sense when you need AC->DC->AC conversion anyway to connect grids that are not synchronised. Or when your generators are not synched

- as in an offshore windfarm.

Or when your solar panels produce DC...

What has happened is that power semicinductors of serious speed and power are now wavilable to replece te original mercury arc valves used in HVDC, and its therefore become a lot easier and cheaper to implement.

And the need for it rises as more and more grids seek to swap problems with each other to make their renewable energy fantasises work.

Of course, the same is true for an AC cable, and twisting the two conductors improves the cancellation in either case.

Batteries seem effective in California, although I'm not so sure that huge batteries maintaining the grid is a better solution than putting them in houses and factories and sticking PV panels on roofs thus making the integrity of the grid far less important.

ROFLMAO!

NOTHING about californias electricity supply is effective, except the links to other states...

I am not talking about maintaining the grid. I am talking about adding some few seconds of 'inertia ' to it.

The cost of a domestic solar panel and enough battery storage for winter well exceeds the cost of a nuclear power station and a national grid.

OK perhaps I should have mentioned Australia - the point is batteries not California.

Batteries can do both at utility scale or at domestic scale make the reliability of the grid far less important.

Why on earth would you want enough power storage for winter ? That would take four or five megawatt hours of storage.

A few hours storage (say ten to twenty kilowatt hours) and a few kilowatts peak of panel (just enough that the total collected annually is close to the total used annually) is enough to make a huge difference, especially if excess power can be sold (or even given free) to the grid when the batteries are full.

Even without solar panels a few hours of on-site power storage is more than enough to cope with brownouts and the vast majority of outages. It's just that once you have the storage feeding it with locally generated solar or wind power makes obvious sense.

Had such a setup been feasible in 1973 it would have made the regular three hour power cuts completely unnoticeable and reduced the impact of the oil crisis to being a minor nuisance.

DROFLMAO!

Even worse there.

Utter total BOLLOCKS.

Exactly. Because dear boy the sun dont shine in winter much, if at all.

More bollocks

Complete s**te.

You'll generally want some form of voltage regulation at the load end of the run in any case (to compensate for losses in the cable) and while in the past that may have been a linear regulator these days switch mode supplies are relatively simple to design thanks to off-the-shelf chips. Given the reduction or elimination in heatsinking and/or active cooling you are likely to find such a solution is actually cheaper and more compact than a linear regulator to boot.

SMPSes work theoretically losslessly by charging and discharging (usually) an inductor so while a linear regulator would take a higher voltage at full current the current decreases with voltage in the case of an SMPS.

yep. its tare to even see a transformer - just an inductor anbd capacitor and a fast switch is all it takes

a 50% duty cycle switch feeding a eries indictor with a cap to te other termninal on te far side will in general give a 50% voltage reduction, etc.

The higher freq. you switch it, the smaller the LC can be.

Buck converters use the flyback voltage of an inductor (think car ignition coil) to generate step up voltages

transformers not used unless isolation needed.

At utility scale they make the grid more reliable because they switch in and out far faster and more efficiently than any generator can. Proven repeatedly in Australia.

At domestic scale they make the householder immune to brown out and outages short enough not to drain them, proven repeatedly by everyone who has domestic power storage (there are a lot these days since Tesla started selling powerwalls even though those things are a *lot* more expensive than deep cycle lead/acid to put in).

Now tell my why those two true and proven statements are "Utter total Bollocks" - or is it that you are reduced to foul language instead of reason.

Sigh - go look up some figures - the energy production of a solar system in December and January is about 20% of that for the same system in June and July - at UK latitudes. So for off-grid winter usage you need a system that produces enough energy for your needs at 20% of rated capacity, and enough power storage to handle long dull stretches. The usual estimate for off-grid usage is at least five days storage.

Of course wind generators tend to work better in autumn and spring and about the same in summer and winter.

But I wasn't suggesting fully off-grid I was suggesting local storage and generation that reduces grid power usage and provides reliable power when the grid does not. For that every little helps.

Every kilowatt hour that goes into batteries from solar panels is a unit that doesn't have to come from the grid with a system as described above every day that produces enough energy to last overnight means nothing drawn from the grid. Power is only drawn from the grid when the batteries are too drained. If the average supply from panels over the year matches the drain for the year then in summer the system will be dumping energy or pushing it into the grid, in winter it will be getting about half the energy from solar and the rest from the grid.

Back that up or shut up. I can back up all my statements with facts.

Three hours of battery storage in the home would have meant no power outages felt. If those batteries could have been charged with solar then that coverage would have come at no load to the grid, and therefore no use of fuel.

Again back up your potty mouthed denials with some facts or STFU.

Initial sales price per nominal kWh of storage or lifetime cost over all cycles until worn out?

I'll accept the former but would want proof and numbers for the latter.

No. 10%

So for off-grid winter usage you need a

So in summer its producing 10 times as much as you can use.

And in winter you need a massive expensive battery. And of coure your coisnumption is higher - need lighst hetaers....

Nope. a part share in a nuke ius WAY chjeaper and uses AY less carbon to make

So can I but I cant be arsed. Believers In Renewble Energy dont like facts

Like Germany is the biggest emitter of carbin dioxide in the whole of Europe, by nation, per capita and per MWh. It is also the most expensive in terms of domestic electricity and it has the highest percentage or renewable electricity. And mote nucler power than the UK.

Initial sales price - hence "to put in".

Li-Ion has a much better long term cost because of the longer life, OTOH I'm not sure it's worthwhile given the amount of battery technology research going on. It seems like pretty much every week someone is announcing something that outperforms Li-Ion in the lab, it may well be a reasonable bet that by the time a new Lead/Acid install is worn out there will be something much better than Li-Ion to replace it with.

You misssed the keyword "anything"...

I wondered if anyone would spot that Catch 22. B-)

IIRC every few seconds the sending end applies a low voltage and current limited pulse of power, not enough to damage anything but enough for the far end to say "hi I'm here". This might just be by changing some chracteristic of the line rather than an ASCII encoded data stream. When the sender gets the correct response it starts to supply the minimum and a real negociation is initiated. Once up and running the load has to ensure it takes a minimum about of power, otherwise the sender will shutdown.

er, AIUI the vast majority of domestic solar PV systems shut down when the mains goes. They do *not* provide any autonomy.

So for 20 kWHr/day you need at *least* a 100kW rated system.

annually

"a few kilowatts peak"? Do the maths ... see above ... 20 kWHr/day at 20% means a 100kW peak array.