Vexatious cost of photovoltaic power and how we beat it
Cost of all PV products is one sore point we must tackle. It affects the economics of all solar power applications, esp. in in India. Here is why:
- The global prices of crystalline silicon solar cells are centred about USD 2 per Wp or Peak Watt, the average power generated
during those sunshine hours as can be tapped (called peak insolation
hours)
- Most of India, for example, has a yearly mean of 4.5 peak insolation hours. Hence a 2Wp cell/panel generates 2 X 4.5 = 9 Watt-hours of energy. Hence, while the cost 'per watt' is high, a typically
tropical country, with its higher insolation hours, produces more energy and
makes for a lower cost per watt-hour, say as against a European or
Polar location.
- However - In case of crystalline silicon cells, these need to be opened in a vacuum chamber, mouned on panels, silver-soldered with all interconnections and external leads and then encapsulated in resin. This process makes a panel in a Western country anywhere up to USD 8 per Wp, while in India, TBP, BHEL, Udhaya and others sell their panels at Rs 200-250 or 5 USD per Wp. (Previously, 5USD/Wp used to be the import cost of cells alone.)
- BUT these panels start at 35 Watts, i.e. Rs 7000-9000 each. These make for half the cost of a typical SPV-based power system comprising charge regulator, battery, load-side controller, load equipment (50 Hz invertor / Electronic ballast + PL lamp / Fan / combinations thereof) This makes for the high cost of the solar streetlights and domestic light-fan systems that you see being advertised by Tata BP, Udhaya, Ritika, SELCO and so many others.
- Due to the selling costs increasig by the day,
especially in this business where Government subsidies are involved, manufacturers are steadily increasing the power ratings per system instead of reducing them. In fact they have dropped manufacture of 9/10Wp panels and extended their ratings to 75 and 125 Watts!
- However, to drive a DC motor as in a pump, the power required is DOUBLE the HP rating of the motor. This is why a 1.5 KW pump
makes for an astronomical cost.
- For the above set of reasons, such systems are only justified in cases where the cost of not having power is greater. Large
parts of Andhra should come within this parts.
- The MNES and SNAs or State Nodal Agencies also buy
these systems from manufacturers and install them at their full/part cost in areas selected by them. Ridiculously so, in states like Haryana and Punjab which are supposedly 100% electrified.
- Subsidisation is availed in no-power zones, where the panel cost is borne by the Central Govt and some State Govts too subsidised. But this subsidisation is gradually being withdrawn.
- Anyway, there are a lot of "contingencies" involved in this business. Manufacturers naturally have to pad selling margins.
- About 40-60% of India's PV production, as per individual ministry's allocation goes to the Non-Subsidy sectors of Telecom (the P&T Microwave network, etc) Railways (communication, signalling and remote area station lighting) Oil & Gas (field radio and pipe-line protection, a kind of plating process) as well as Defence (radio and camp lighting) ... obviously, their costs of not having any power are much larger than the cost of these equipment, which are further marked up in price due to the higher QC & Reliability Engg quotients here. So a good time is had by all!
- These concepts seem to be emulated by other developing countries as well, though Indian products become somewhat cheaper
due to the low labour quotient as well as export subsidies. I believe that some silicon cells are still imported against an export obligation.
For this reason, the industry does not appear to be
in a hurry to go beyond the tender-driven product lines and high unit prices. Last year Tata BP made Rs 720 million from domestic sales and another Rs 1.28
billion from exports, using its BP connections in Africa, East Asia, etc.
in a hurry to go beyond the tender-driven product lines and high unit prices. Last year Tata BP made Rs 720 million from domestic sales and another Rs 1.28
billion from exports, using its BP connections in Africa, East Asia, etc.
On the other hand:
- Amorphous silicon is about 1 USD per Wp.
- Its handicap is that it takes a 6"X6" square
surface to deliver 1Wp as against about 1"X1" sq in a mono-crystalline cell.
- Scientists call it poor efficiency. Manufacturers
of crystalline cells claim an efficiency range of 14 to 18 %, (converting that fraction of all solar radiation falling on a unit area into electricity) an amorphous silicon cell converts at 6 to 8 %.
- Earlier, amorphous silicon could be deposited in
very thin layers, grealty hampering current carrying capacity and the crystalline lobby claimed it was only suitable for small watches, calculators, etc.
- The truth is that crystalline panels have their losses caused by heating up of cells in the sun and due to current, also due to the heating up of soldered joins and interconnection leads, being covered by insulating resin. This reduces the effective efficiency to 10-12 %. On the other hand, improvements in amorphous technology have brought these too effectively in the 10% conversion range.
- Current and reliability limitations have been largely overcome, by being able to coat amorphous silicon in vapour form on glass sheets in strip form. The coatings are covered with an aluminum laminates.
- aSi or amorphous silicon panels are now mass-produced this way.
- This makes it possible for us to cut amorphous silicon or aSi coated glass panels to a desired voltage and current in a way as to choose the length of each strip by the current and number of strips by the voltage.
- So even if a panel is supplied in large ratings of several Watts and a Metre in length/width, as from some European and Japanese manufacturers, we can literally cut our cloth to our needs. Hence the range from about 300mWp to 2Wp.
- This way, we keep the costs low.
- To arrive at this low power rating, we study each energy usage application to -
- find the lowest amount of energy units
needed to do a task, say illuminate ground, light a room, lift water from a tank, charge a radio's batteries in a day, ...
- Energy being Power X Time,
we then decide if the charge needs to be complete in 1 day or more;
- For
each day of charging we assume Time = 4.5 Hrs;
- Hence power required
becomes Energy in Watt-hours/ 4.5 Hours.
Herein I have let you into another round of our secrets. ;))
The next post will be in finding out the lowest
energy solution. You will be surprised at what we have found ... questioning so
many absolutes ... like the mini pump application that Graham
Knight uses in African & Asian projects ... perhaps at 1/1000th the cost of what anyone provides here!
energy solution. You will be surprised at what we have found ... questioning so
many absolutes ... like the mini pump application that Graham
Knight uses in African & Asian projects ... perhaps at 1/1000th the cost of what anyone provides here!
Regards
Udit Chaudhuri
