Best systems for electricity generation?

[Jules_Theone]

I've seen recently about 'smart' electric meters in Britain are going to be in every home by 2020 costing at least £7 billion*. They are designed to measure electricity usage as before but give more accurate break-down of where the energy is being used, at what time of day, average over a week etc. and electricity will cost more during peak demand times. Supposedly everyone is going to do their washing at 1am (off peak) to save money and help the power companies?

 

[Nigel Goodwin]

I've seen recently about 'smart' electric meters in Britain are going to be in every home by 2020 costing at least £7 billion*. They are designed to measure electricity usage as before but give more accurate break-down of where the energy is being used, at what time of day, average over a week etc. and electricity will cost more during peak demand times. Supposedly everyone is going to do their washing at 1am (off peak) to save money and help the power companies?

More importantly it saves the companies employing meter readers - so the huge investment will save them money in a very short time.

 

[dougy83]

Micro grid-tie is very cool... I was looking at 150w inverters now the size of a soda can for $30 USD and fully solid state so I guess efficiency wont be too bad. It's not a big stretch to modify one of these and pump your 80 watt solar panel straight back into your house mains (after checking local laws).

I'm sure the power company would love you trying to pump modified squarewave into their grid.

 

[Mr RB]

From my understanding these little cheap units are not modified squarewave. They dont use a 50Hz transormer at all, they use a high freq switchmode supply to generate 50Hz 240v half-sine, then use 4 output devices IGBTs etc to act as a commutator bridge to make the half sine a full sine AC output. Ripple is very low and so is manufacturing cost...

Modified squarewave is kinda last century technology.

 

[tcmtech]

I have heard about that set up as well. I have not found one yet to take apart and study further though. But it is not a hard circuit to design.

As far as Grid tieing goes, even with a modified sine wave, a standard properly designed LC choke type line filter takes the square wave edges and will round them off to the point that your not going to have any problems with line noise. Its basic filtering 101.

Also in a grid tie setup your GTI unit does not need to be small. It's not intended to be used as a portable device! What your after is input to output conversion efficiency not power to physical volume efficiency.
I dont care if it takes up a cubic foot for every KW of capacity or weighs in at 50 pounds per cubic foot too.
If it does the tie in and conversion efficiently thats what counts.

If you cant fit another cubic foot or two of equipment onto your property you really need to start tossing stuff out!

 

[Jules_Theone]

From my understanding these little cheap units are not modified squarewave. They dont use a 50Hz transormer at all, they use a high freq switchmode supply to generate 50Hz 240v half-sine, then use 4 output devices IGBTs etc to act as a commutator bridge to make the half sine a full sine AC output. Ripple is very low and so is manufacturing cost...

Modified squarewave is kinda last century technology.

I'm in the process of reverse engineering one of these 150w switching inverters, almost have a complete schematic which I will post here when it's done. As far as I can see, it's just a 12v to 360v DC-DC flyback supply, controlled by a TL494, then a H-bridge is used to 'chop' the DC to a squarewave with 50% deadtime added. Most of the inverters I have analysed use two TL494, one for each part and op amps for voltage / load sensing. Another one i have dismantled used a PIC to control the analogue circuitry, but I didn't look at it in too much depth as I can't 'see' whats in the PIC.

I think using a high frequency circuit to generate the high voltage then chopping it to make AC is going to be more efficient than chopping low voltage and feeding it into a bulky transformer to step it up.

 

[Mr RB]

Absolutely. I thought of making a "proof of concept" micro grid tie with a PIC and just generate the high-freq PWM to 50Hz 1/2 wave 240v sine, and use one IGBT (or even one SCR?) to feed the 1/2 wave sine back to the grid. I've built high-freq switchmode supplies for many years and figure 12v->240v half sine should be easy enough at 85%+ efficiency. It could even be high-freq current controlled so it feeds controlled current into the grid 240v sine, so the waveshape is determined by the grid anyway.

Being micro grid tie of a few watts it won't matter feeding power back only half the time (on every 1/2 wave). And of course if there are more than 1 of these micro devices you can just connect some of them reverse, so some feed in on the A half wave and some on the B half wave.

I figure this is simpler and slightly more efficient than a full bridge commutator on the AC side... If the goal is to be able to pump power from a 12v 20W panel directly back to the grid at the best efficiency, best reliability and least cost.

 

[tcmtech]

I think using a high frequency circuit to generate the high voltage then chopping it to make AC is going to be more efficient than chopping low voltage and feeding it into a bulky transformer to step it up.

I am hoping for a rainy day soon!
I plan to follow up on some independent research projects relating to the switching at the line side. I just built a big power supply so I have a good isolated and variable 0 - 200 VDC 20 amp power source.
That should be able to push enough power in reverse to get any big line noise to show up on my O-scopes.

I had very high winds yesterday and was able to hook up my O-scope to my latest experimental GTI unit. 12 -22 volt input.
The high winds were pushing peaks of 60 + amps!
I did see some line hash form when directly monitoring the AC Line coming right off the GTI (I had to zoom in considerably to see it clearly though) but I couldn't pick it up any place else in the shed on that line.
I suspect it was from grossly over driving the line filters. I know the one I have on it was out of a 500 watt UPS and is about a third of the capacity I should have. At about 800 watts returning the little fuzz showed up at the switching points.
I have snubber caps and MOV suppressors mounted in my breaker panel so I suspect they were doing the final clean up so thats why the line noise could not get any further.

I am sure that line side switching will pick up some added efficiency and reduce the overall size more too! But it will still require a fair filter to get the switching point noises cleaned up.