Gary's Comments

Ontario has a 60% local content requirement for its solar feed in tariff, though this has resulted in higher prices than would otherwise be the case. This is likely to be less so for low cost thin film technologies, as the lower cost of the panels, the low proportion of cost attributable to labour costs and the proportionately higher cost of transporting them (high weight / $ value) erodes much of the current advantage of manufacture in China.

Regarding keeping the power as DC, this is unlikely to result in much of a saving as there are a wide variety of DC voltages required, and for import / export of power from the home significantly higher voltages are needed than are used in domestic DC appliances. Where an appliance has a good match to the DC voltage output of a PV panel, by all means avoid the DC/AC/DC conversion - otherwise you lose the advantage through DC/DC conversions.

Whilst I do agree that solar power will provide an increasing proportion of energy over time, and recognise that the above is a paper exercise,in the real world, a variety of energy sources will continue to be used.

By blending wind, solar, and small hydro power with the use of biomass such as straw, forestry waste etc. as a "swing producer" a far more secure supply of power is possible with far less power storage.

A further possibility which is becoming feasible in an increasing number of locations is geothermal which is ideal for providing base load power and district heating.

I would criticise the assumptions made in the report regarding projected costs as far too simplistic - the cost of solar will not indefinitely drop as at some point, the technology will become mature and costs will follow those of the material inputs. Also the wage related costs can only drop so far as the systems become easier to fit and PV becomes more powerful per sq metre installed, though on larger systems, robotic installation may play an increasing role.

Regarding coal, the cost is unlikely to escalate indefinitely in real terms as described - especially once carbon capture becomes a mature technology.

There is also an assumption that power consumption will continue to escalate as it has done in the past - much can be done to mitigate this through efficiency measures, though such a scenario is possible if vehicles progressively turn to electricity - in which case, there will be a huge amount of multi-purpose storage in the vehicles.

Some of the projected increase in grid capacity within Kentucky can be averted by dispersing much of the capacity on roofs and near to points of use across the state, though increased interconnection of the electrical grids across the USA into one super grid will allow excess power generated in one state to be utilised in another where at that moment there is a deficit.

Gas / solar hybrid technology isn't new. There is a plant in Algeria

The plant is a "150 MW hybrid solar-gas plant at Hassi R'mel, 420 kilometers south of Algiers. The plant is due to go into operation in 2009 and has a 25 MW solar energy capacity with a parabola trough design"

http://www.renewableenergyworld.com/rea/news/article/2008/03/low-cost-solar-thermal-plants-at-heart-of-algerian-german-research-push-51889

http://www.skyscrapercity.com/showthread.php?t=1283009

Graphine would be great in this application, however we will need a lower cost method of graphine production as it currently is much more expensive than gold

There are essentially two main factors affecting the management of power grids with large penetration of intermittent generation.

1. The risk of excess generation

In extreme conditions, power generated by wind and solar can exceed demand on the local grid even before any contribution is made from traditional generation. This has occasionally happened on windy summer nights in Western Denmark (Denmark has 2 power grids).

2. Variation from prediction

This in the more common problem and relates to errors in predicting wind or solar power, and consequent increases in spinning reserve which must be maintained to ensure grid stability. Usually error bars across a region are around +/- 10% from prediction 4 hours ahead.

Both of the above can be mitigated by one of three techniques

a. Adding power storage to the system

This enables any excess power to be absorbed and released next time there is a deficit.


b. Management of load

In the case of areas with district heating, this can direct any excess electricity to heat pumps (factor 1), or with the same heat pumps act as a swing - scheduling power production up to the bottom of the error bar to the grid and directing any electricity above this production level to heat so eliminating most of the net variation from predicted output (factor 2)

Note:- There are other loads which are not time critical which can substitute for heat pumps on district heating networks

c. Improve grid infrastructure

Develop more and bigger electrical interconnections allowing power to be shared over a much bigger area. This will greatly reduce the risk of excess power production and tend to reduce the percentage size of the error bars on predicted power output (variations tend to even out over a large area).

Thanks for the link, the aerofortis product looks very interesting as a way of maximising yield and easing design constraints.

The product uses a DC to DC converter which as you say manages 4 panels with maximum power point tracking. Several strings can be combined onto a central inverter which receives a constant voltage input. The inverter does not itself have to contain a maximum power point tracker, so can be of simpler lower cost design.

As strings are short, and independently managed, it does not matter if each string has a different orientation, or if the panels are a different size or from a different manufacturer, so eliminating much of the complexity of array design and reducing concerns regarding array losses with partial shading or panel mismatch.

Sorry Alphathermal,

I havn't been able to find the graph I remember seeing showing cost v scale for solar thermal, however I have found some installed costs for large scale solar water heating from Denmark and Germany, and costings for a proposed large scale solar thermal system in Bishkek Kyrgystan.

http://www.solarthermalworld.org/files/District%20heating.pdf?download

http://www.solar.uni-kassel.de/sat_publikationen_pdf/2008%20EuroSun%20Budig%20et%20al%20Solarassisted%20Water%20Preheating%20for%20a%20District%20Heating%20Net%20%E2%80%93%20A%20Potential%20Analysis%20in%20CIS%20countries.pdf

One thing to note is that CSP can be used in hybrid configuration with conventional steam driven fossil fuel systems. Any heat fed into a conventional power plant from a CSP array displaces heat which would otherwise come from fossil fuels, and in doing so, generally achieves a higher efficiency than a stand alone CSP facility. This is because conventional plant operates at higher temperatures and pressures than CSP, so achieving higher levels of heat conversion to electricity as predicted by the Carnot equation

In cities with district heating, solar thermal can make a useful contribution to meeting the city heat load in a highly cost effective manner, particularly when large arrays of solar water heaters are deployed. From some figures I have read, the cost of such large systems can be up to 6 times lower than individual family homes due to

1. Bulk purchase of hardware
2. Savings on the time taken to install and plumb each each collector
3. Savings on the planning and preparation of each collector
4. Substantial savings on scaffold cost per collector
5. Savings on selling, delivery and administration per collector installed.
6. Avoidance of the need for dedicated hot water storage, or if such storage is provided, far lower cost per volume of that storage.
7. Avoidance of "wasted heat" - heat goes into the district system and is more likely to be used by someone - with an individual house, there will be many occasions when not all the hot water is needed such as when the owners go on vacation.

There's a tendency still to confuse energy use with electricity use, and energy storage with electricity storage.

Electricity - or the ability to generate it can only be stored at large scale cost effectively by pumped hydro storage, or compressed air energy storage - both of which are dependent on special geographical / geological conditions only available in a few places.

A better alternative might be to look at thermal storage for thermal loads and dispachable energy use to store or time shift energy which would otherwise come from live electricity use.

How would this be achieved?

District heating - hot water can be obtained from a mix of solar thermal and heat pumps, and stored in a cost effective manner in large tanks or by heating up a large underground block of soil and rock using an array of deep boreholes. Heat from the heat pumps can then be obtained when the grid has spare power to offer - with electricity use regulated to match available supply by the grid operator.

Cooling - likewise with air conditioning and refrigeration loads, there are ways to store cold using tanks of slush ice with or without salt additives (these additives can make slush ice down to -40 centigrade). In a few cities like Stockholm, district cooling has been installed in the business district which makes storing cold more cost effective as storage cost declines with increasing tank size.

Aluminium manufacture from Aluminium Oxide by electrolysis could also be carried out under the control of the grid operator, again giving substantial effective spinning reserve for modest cost - true, limiting the capacity factor of the plant, but at the same time substantially cutting its specific electricity cost.

These are just a few examples among many of ways to shift electricity demand to better match availability - greatly increasing the potential proportion of electricity which can be obtained from intermittent sustainable sources without disrupting the grid.

Regarding input capacitors - if there is a choice between an input capacitor which is so cheap that manufacturers allow for one warranty replacement per unit (needs replacing 5 years into a 10 year warranty), or an expensive input capacitor which lasts 20 years, might there be another way to do things involving redundant components

- building two or more low cost input capacitors into the box only one of which is functioning at any one time, with either a manual or automatic switch to bring in a spare when the first input capacitor fails?

If used this way, would the spare/s deteriorate in the box before it is used, or only begin to deteriorate once it is used?

No mention of vertical axis machines. Whilst vertical axis machines have not so far had much of a role in the wind industry, there are possibly fewer constraints on size with vertical axis as the blades in some designs are in tension. This means that very high stiffness is not an essential design consideration any more than it would be for a sail - so allowing the possibility of far lower blade weight, and construction of blades in multiple sections. The design also allows the generators to be far lower in the tower, reducing crane requirements.

With micro-inverters, the biggest area for caution is reliability. While I would not wish to discourage their use or champion any one brand, a micro-inverter - even for domestic use has to be reliable for around 25 years whilst exposed frequently to high temperatures, possibly up to 90C for on roof and 110C for in roof arrays - harsh conditions for any electronics. The cost of identifying a failing inverter, getting up on the roof, getting to and lifting the affected panel and replacing the inverter will be very high, so lack of reliability will carry a large cost penalty - only part of which (the cost of a replacement inverter) is likely to be covered by any inverter warranty.

Don't forget the real value of power which varies with the time of day. A kWh generated in the early afternoon on a hot summer's day is worth significantly more than a kWh generated 12 hours later when most air conditioning is off or not working as hard, offices are generally closed and most lights are off.

For this reason, the average value of a kWh generated by solar is a little higher than a kWh generated by wind as the generation pattern more closely approaches the demand pattern.

One system proposed in the UK and due for implementation next year is to require the energy companies to loan funds for energy efficiency / renewable energy measures against future energy bills over a 20 year period as a charge against the house and transferable to any new owner. This will allow the customer to repay the loan out of savings, and ensure that any viable domestic energy saving / renewable option is fund-able. The Government is now preparing the required legislation.

An alternative is to have the mortgage companies adjust their affordability of loan criteria to take into account the ongoing running costs of a home. This is already happening to an extent with some lenders in the UK who offer a slightly higher earnings multiple to borrowers buying an energy efficient house or one with renewable energy features to recognise its lower bills.

Note:- All homes sold in the UK come with an energy performance certificate enabling lenders to recognise energy efficient homes in order to make this adjustment to their loan criteria.