Households can supplement network

2012-02-19 16:30

Cape Town – Solar power could for most of the year provide the bulk of an ordinary household’s electricity requirements, but this would currently prove expensive.

Municipalities are still resisting the practice – already commonly adopted in Europe – by means of which home owners can connect their renewable energy systems to the power grid and be compensated for the excess electricity they upload to the network.

Three Cape Town homeowners now form part of a one-year test project, together with Cape Town municipality, which will compensate them on a one-to-one basis for any electricity they generate.

This means that when the households' photovoltaic solar panels generate more electricity than the household can use at that point, the electricity is sold to the municipality at the price the households pay for electricity when their solar panels generate insufficient power.

Dr Anthony Keen, a retired academic from the University of Cape Town’s medical school, is one of these homeowners who has been experimenting with domestic solar energy for more than three years.

Between 2007 and 2009 Keen spent R262 329 on constructing his system. Based on this outlay, the electricity that he generates himself costs R2.17/kWh.

But, as Keen says, in 25 years' time it will still cost R2.17/kWh – and what will Eskom charge then?

Moreover, photovoltaic solar panels are becoming cheaper as sales increase and technology improves.

A tariff incentive for selling excess green power to municipalities would make this type of long-term investment even more attractive.

Keen has mounted 20 solar panels with a total generation capacity of 3.8kW on his roof. A bank of 24 2V batteries and a single-phase 6kVA inverter, which converts direct current into alternating current, have been connected to the panels.

In his experiment he simulated two scenarios. In one he aimed to manage, as far as possible, without power from the grid. That was by using battery power when solar power did not suffice. Electricity from the grid was used only when the batteries produced too little power.

In the other scenario net-metering was simulated and the sequence for using electricity was first from the solar panels, then from the power grid and then from the batteries.

Keen said net-metering is easier because everything is automated, but this does not use solar power as efficiently as in the first scenario.

From May 2010 to April 2011 Keen’s household bought an average of only 86kWh/month from the grid.

The remainder came from solar power. Electricity consumption is managed so that most of the power is used when the solar panels are generating most of the electricity.

In the year from March 2009 to April 2010, Keen generated 6 431kWh from his solar panels and bought 2 001kWh from the grid.

According to Keen, the plumbic acid batteries used in his system are his weakest link – they are only 54% efficient.

He wants to invest in lithium batteries. These cost five times as much, but in the long run seem economically attractive.

Another alternative for the future provision of energy is to use surplus electricity to manufacture hydrogen to deliver electricity through a fuel cell.


According to Keen, people have various reasons for tackling a project such as this.

Some want to wean themselves from dependence on Eskom owing to the risk of power disruptions, some want to help Eskom by reducing the burden on the electricity grid, some want to adopt a greener lifestyle and help the planet, and some are simply fascinated by the technical challenges.

It's often a combination of all of the above. Keen decided on solar power because he reckoned it was the most practical and easiest application in an urban environment. He looks upon wind power in such a residential area as being too risky and unpredictable.

  • Clive - 2012-02-19 18:21

    Potentially every dwelling in South Africa can generate enough power to cover its requirements. The cost of battery storage is a costly component. Potentially every dwelling in South Africa can generate enough power to cover its requirements. I suspect that the need for large arrays of batteries could be eliminated with the householder merely selling excess capacity to the grid during the day and buying capacity from the grid at night the model might look a lot more positive. It potentially could have a positive impact on smoothing out the spikes in the demand. The problem will turn out to be Eskom who will see this as a threat - hence their reluctance to allow roof generated power to be fed back into the system.

  • Skia - 2012-02-19 21:29

    "A bank of 24 2V batteries ..." really? and it has always been eskom that has torpedoed the idea of people supplying power to the grid.

      Michiel - 2012-02-20 06:26

      Those 2v moldular batteries can supply up to 800 Amps - each. They are extremely expensive.

  • Rob - 2012-02-20 03:28

    I don't think Eskom is the actual problem here. As the article indicates above, most electricity users purchase their electricity from municipalities who are resellers and they don't want to reduce their income by having to pay back to those who feed back into the grid.

  • Gideon - 2012-02-20 07:26

    I am also very interested in this topic, having implemented just about all the other energy saving suggestions; I also want to get into solar power generation. My thinking has been not to rely on battery storage (due to the losses) but rather over generate during the day ("selling it to the grid") and buying back during the evening. In my calculations, I should be able to do this with a 5-6 KVA system. My aim is to lessen the burden on the environment but I am looking for a return over a 10 year period. This seems to be possible providing that you can sell power to the council. I stay in Pretoria and is wondering of the Tshwane municipality would support me in such a venture?

  • Ariado - 2012-02-20 08:33

    Think the battery replacement costs and life on the battery bank. Not worthwhile.

  • errol.wagner - 2012-02-20 08:39

    Why are there no benefits for those who use gas instead of electricity?

      Rob - 2012-02-20 12:16

      You use less electricity. With the way the price of oil is increasing, the cost of using gas will reach parity with electricity which is more convenient than gas.

  • Andreas - 2012-06-06 11:50

    Very similar comment as the one I posted on News24's coverage of the Vodacom installation. The best is a grid tied system, i.e. no battery backup. Here you can work with prices of around R60k for 2.5kW, (which is enough for a normal household by the way). A 2,5kW system produces around 4500kWh per year. At an electricity price of R1.20, this works out to around R5400 / year and thus a ROI (based on the current electricity price) of around 11 years. If you however take a standard 20 year PPA into consideration, and take the R60k over 20 years, your current electricity price from your PV system (1:1 without any escalations or interests) would be R0.66/kWh (compared to R1.20/kWh currently) and this doesn't mean that the PV system stops working at 20 years :-). Prices have come down considerably since Dr. Keen did his installation. If you take a 5kWp PV system with 10kW of battery inverter (with a 6kVA backup genset to limit the days of autonomy vs. bad weather) and 1000Ah of lead acid, look at around R300k. For independence? Calculate against that the cost of a NEW Eskom connection (transformers etc.) or the cost of diesel and maintenance ... PV will always win hands down as its fuel is free! You must calculate against the total cost of ownership and this, by the way, is also why nuclear will fail the "cheapest power" statement. PV can do for Africa what the steam engine and printing press did for Europe!

  • bob.hull.984 - 2012-09-06 15:26

    Regarding the Return on Investment :- the most cost effective way to go is grid tied inverter connected to solar panels.If you have rotating disc meter turn off everything except your solar, & if your solar capacity at that moment is greater than +/- 250W you will see your meter turning backwards. Further when your solar power < power being used @ that moment you are being "paid" the going rate, because you are saving electricity at that price, so grid tied inverter+no battery= way to go. I have paid R16K for +/- 1KW 20panels (700W @1KW/sqm 25*C) & taking the escalation @ 14.6%. The system will pay for itself in 39months based on 6KWH`s/day for 250 days /year. The crunch is LIGHTING it can damage the inverter in microseconds.I have incorporated Bi Filar Xfms GAS DISCHARGE TUBES,SILICON AVALANCHE DIODES & M.O.Varistors in my surge suppression between inverter & grid. USA DEPT.OF Energy give panel life = 25 years but inverter only up to 10years & our lightning (Gauteng) is worse than theirs.Is there somebody with a storage scope who would like to collaborate with me on this ; to see let-thru pulse power ? I have 30KV pulse generator.

  • pages:
  • 1