Solar vs Diesel Pump

A COST AND RELIABILITY COMPARISON

BETWEEN

SOLAR AND DIESEL POWERED PUMPS

Solar Electric Light Fund (SELF)

July 2008

Introduction

In rural and/or undeveloped areas where there is no power grid and morewateris needed than what hand or foot pumps can deliver, the choices for poweringpumps are usually solar or a fuel driven engine, usually diesel.

There are very distinct differences between the two power sources in terms ofcost and reliability. Diesel pumps are typically characterized by a lower firstcostbut a very high operation and maintenance cost. Solar is the opposite, with ahigher first cost but very low ongoing operation and maintenance costs.

In terms of reliability, it is much easier (and cheaper) to keep a solar-poweredsystem going than it is a diesel engine. This is evident in field where dieselengines lie rusting and unused by the thousands and solar pumps sometimesrunfor years without anyone touching them.

The first cost of solar is oftendaunting to donors and project implementers whoare tempted to stretch their budgets as far as possible to reach the greatestnumber of beneficiaries by using a low first-cost option. But most would probablyagree that “quantity over quality” is not a good value if the higher quantity optionis not likely to be giving good service five years down the road and ifbeneficiaries are going to be stuck with interventions they cannot afford tosustain over time.

Solarpumping has had clear advantages for a number of years but thedifferences are becoming more striking in a world of rapidly escalating fuel costs.Not only will some of the world’s poorest people not be able to afford fuel for theirpumps, but living at the end of remote supply chains, they may not even beableto get it in the first place as world demand overtakes supply. (Rwanda hasalready had at least 3 nation-wide fuel shortages in the last 18 months).

In this paper, we offer evidence accumulated by others as well as from our ownexperience showing that solar pumping is the most reliable and cost-effectiveoption for many water pumping applications in developing countries.

Cost Comparisons – HOMER Simulations

The U.S. National Renewable Energy Laboratory (NREL) has developed asophisticated simulation program that optimizes the most economicenergychoices per specific project inputs. In this example, SELF has usedHOMER(Hybrid Optimization Model for Electric Renewables) to model choices for apumping system that is designed to pump 5,000 gallons per day from a totaldepth (head) of around 100 feet. It compares a solar array of 1900 watts againsta 4 kW diesel generator. Both power an equivalent pump of approximately 1horsepower. Several simulations were performed to gauge the effect of the priceof fuel and the fuel efficiency of the diesel generators. These and otherparameters are listed below:

Fuel cost:

Case 1: $1.20 per liter

Case 2: $1.50 per liter (current approximate cost in Namibia, Rwanda and

Benin)

Case 3: $1.70 per liter

Fuel efficiency (consumption) of diesel generator:

Case 1: .3 liters per kilowatt generated

Case 2: .5 liters per kilowatt generated

Case 3: .7 liters per kilowatt generated

Solar resource: Annual average of 4.6 peak sun hours per day

(This figure is low for most African countries and represents

a conservative approach)

Real annual interest rate: 5%

System life: 20 years

Key program outputs:

Initial capital cost: “first cost” for each option – assumes same pump costs

Operation cost/year: Average O&M costs per year. Does not include pump

replacement costs which would be same for both

Net Present Cost: The present value of the cost of installing and operating the system

over the lifetime of the project (also referred to as lifecycle cost).

$ per kilowatt: The cost per kilowatt of electricity per each option.

Simulation 1: “Worst case” for solar: Fuel cost: $1.20 per liter. Consumption rate:

.3 liters per kilowatt

Initial

Capital

Operating

cost/year Total NPC $ per kWH

PVP Option $12,300 $335 $16,472 $0.66

DP Option $2,000 $4,854 $62,494 $2.48

Simulation 2: “Best case” for solar: Fuel cost: $1.70 per liter. Consumption rate:

.7 liters per kilowatt

Initial

Capital

Operating

cost/year Total NPC $ per kWH

PVP Option $12,300 $335 $16,472 $0.66

DP Option $2,000 $12,525 $158,094 $6.27

We can see from these simulations that solar ranges from one tenth to one fourth

the Net Present Cost of the diesel option.

Solar vs. Diesel Cost Comparisons – Recent Studies by Others

One of the most comprehensive recent studies comparing solar to dieselpowered pumps is the 2006 report “Feasibility Assessment for theReplacementof Diesel Water Pumps with Solar Water Pumps”, issued by the Ministry of Minesand Energy of Namibia, prepared by EmCon ConsultingGroup and sponsored byUNDP, Global Environmental Facility (GEF) and the Government of Namibia.

Namibia is an ideal location for a pumping study. Because Namibia has very littlesurface water, it relies on over 50,000 bore-hole wells for its water supply. Manyof these wells are off the grid and are powered by diesel pumps. Solar energyhas been used for pumping in Namibia for over 25 years and from 2001 to 2006,669 solar-powered wells were installed – creating a large field for study.

This report furnishes overwhelming evidence that for small to medium sizedwells, solar (PVP, or photovoltaic pump) is much cheaper on a life cycle costbasis than diesel-powered (DP) pumps. When looking beyond the originalpurchase price, PVP pumping systems cost anywhere from 22-56% of whatdiesel pumps cost and can achieve a payback over DPs in as little as 2 years.

Though called “small to medium”, most village water supply pumps (and SELFpumps used for drip irrigation), fall into this range. The daily output of thesepumps range from 800 to 13,000 gallons per day (3,000 – 50,000 liters). Fiftythousand liters isenough to supply 2,500 people with their daily water needswhen using UNDP standards of 20 liters per person per day. The effective depth(head) for these pumps is up to 400 feet or more (120 meters).

DEFINITION OF TERMS AND CONVERSIONS:

Cubic meter = 1,000 liters

Gallon = 3.79 liters

Meter = 3.28 feet

Hydraulic load: the daily output (cubic meters) x the head (total vertical pumping

distance). A typical hydraulic load would be 18 cubic meters from

a depth of 40 meters = 720

$1.00 U.S. = 7.49 Namibian Dollars in Sept, 2006 (date of report)

$1.00 U.S. = 7.66 Namibian Dollars on July 23, 2008

Life Cycle Costs

The following graph shows a comparison for solar and diesel water pumpsthatincludes a range of pumping heads (10m to 200m) and a range of dailyflow rates(3,000 – 50,000 liters). The life cycle costs (LCC) were calculatedover a 20 yearperiod taking into account upfront cost, operating costs,maintenance costs, andreplacement costs.

Breakeven point

As previously stated, PVPs have a high up-front capital cost but very lowoperation and maintenance cost when compared to DPs. It is useful to knowwhen the solar pump becomes cheaper to run than a diesel pump. The chartbelow compares three classes of PVPs to diesel. The systems in yellow arethesmallest pumps of around 1 horsepower. (As commonly used by SELF). Thegreen systems are double yellow systems (2 pumps in the same hole). And theblue system is a large 4 horsepower solar pump.

The following figure shows the breakeven point for a single case – a pumping

system with an output of 10,000 liters per day from a head of 80 meters. The

study also states that for pumping systems having a hydraulic load of 1,000 or

less, the break even point is less than 2.5 years. Most pumps that SELF has

used for agriculture or village water supply fall in this range.

The effects of diesel fuel price increases.

The rapidly increasing price of petroleum products is of course changingeconomic calculations everywhere and these increases greatly impact bothLCCand break even points for diesel pumps. At the time of the 2006 Namibia study,the price of diesel was 6.70 $N per Liter. At this writing, it is 10.64 $N per liter –almost a 60% increase.

According to sensitivity studies done for the Namibia report, every 10%increasein fuel costs results in a 3% increase in LCC. Accordingly, thepresent day LCCof the DP option is 18% higher than what is shown in these figures. It was alsofound that for every 10% increase in fuel costs, the there is a 5% reduction in theyears to breakeven. So a PVP that broke even in 2.5 years in 2006 now breakseven in 1.75 years. During this period (2006-2008) the price of solar has beenstable so the LCC for PVPs stay essentially the same. As we project into thefuture, the price of PV is surely to come down, while the price of oil is likely to goup.

Solar Water Pumping Cost Comparisons – Older Studies.

The following studies are older and therefore are based on fuel costs that arenotrealistic today. As we demonstrate above, the comparison between solaranddiesel is heavily dependent on the price of fuel. All of the studies citedbelowwere favorable to solar at the time of writing and would be even morefavorable ifwritten today with current fuel prices.

• A Sandia National Lab study of 3 different sized solar pumping systems(106 Wp, 848 Wp, 1530 Wp) in Mexico showed that all had lower life-cyclecosts than diesel-powered pumps. The PV systems vs. diesel hadpaybacks of 2, 2.5 and 15 years respectively when replacing fueledpumps (gas or diesel). 1

Note: At the time of this study, 1998, crude oil prices were $11/barrel vs. $137 today (Source: Energy Information Agency, U.S. Government)

• In a comparison of fueled pumps vs. PV, a German study showed PVpoweredpumps to have the lowest life-cycle costs for PV array sizes of1kWp and 2kWp and the same cost as fuel pumps for power ratings of4kWp. (The largest PV pump SELF has installed to date for village watersupply is 1.9kWp)2

Note: The date of this study is unknown, but it was before 2006, when the price per barrel of crude oil was $68 compared to the $137 price of today.

• A study by GTZ (Posorski, Haars, 1995) in seven countries concluded thatPV pumping systems for drinking water are economicallycompetitive inthe range of small diesel pumps (1-4 kWp solar systems).

Note: At the time of this study, the price of crude oil was $17 per barrel, comparedto $137 per barrel today.

Comparison of Reliability between Solar and Diesel

1. Reliability of diesels.

Unfortunately, we have not found quantitative studies dealing with thereliabilityof diesel engines used in developing countries. However,experiential andanecdotal information abound. The following quotes are indicative of what SELFand others have found during extensive experience in the field:

“ Maintenance and high fuel costs have been long-standing problems withdieselgenerators. The systems are often in remote locations, and the difficulties ofpurchasing imported spare parts and fuel have often made them unreliable.”

ESMAP (Energy Sector Management Assistance Programme)– World Bank/UNDP, “Best Practice Manual Promoting Decentralized Electrification Investment” Report 248/01, 2001

“ Quite often the cost ( of operating a generator) can be very much higher thanexpected because of the need for maintenance personnel and the difficultiesencountered in obtaining fuel and spare parts.”

Intermediate Technology Design Group, “Technical Brief – Diesel”

“ Diesel generators can be used in remote areas, but this requires a constantsupply of diesel and mechanical parts. The cost of the Kwh produced istherefore always considerably higher than with connection to the grid. There alsotend to be manydisruptions, owing to the difficulties that many peopleexperience getting generators repaired promptly.”

Renewable Energy and Energy Efficiency Partnership (REEEP). “CASE STUDY: Concession for rural electrification with solar home systems in Kwazulu-Natal (South Africa).” Dr. Xavier Lemaire, Centre for Management under Regulation, Warwick Business School.

“Conventional diesel generators are not a real alternative solution, as they don’tmeet the demand of rural population and have other negative side effects:

- High cost of operation a susceptibility to break down

- Maintenance servicing is poor and expensive

- Not environmentally friendly

Thus, seeking to reach remote villages with conventional systems is prone to fail,

as the supply to these locations incurs costs and risks that are too high for users

to bear.”

Bremen Overseas Research and Development Association. “Decentralized

Energy Supply: Energy supply for household and small scale home industries in

rural and mountainous areas.”

With our extensive experience working in developing countries for over 17years,SELF has seen first hand, over and over, the inability of poor rural populations tokeep diesel engines running.

In one project we visited in Nigeria, 30 water pumping systems had beendistributed to villages. Most were driven by diesel pumps and a handful weresolar. After 5-7 years of operation, none of the diesel pumps were still working.Although we weren’t able to find out the status of all solar pumps, we were ableto visit a village with a working PVP system. When asked when the last timesomeone had come out to maintain the system, village elders looked at eachotherquizzically and then replied that no one could recall anyone coming out tolook at the system – ever - it had been working away virtually untouched forover5 years. (We later confirmed this with the Government agency thatinstalled thesystem – they said they never had money for maintenance of any kind)

Another village in Nigeria has a large working diesel pump for a central watersupply system. But they can only run it every few weeks, as it takes that long tocollect enough money from the villagers to purchase fuel.

More recently, one of SELF’s health partners in Rwanda, Partners In Health, hadinstalled brand new European-built generators at 5 sites to back-upSELF’s PVsystems. These new generators even came with a one-year maintenancecontract provided by a local supplier. Nevertheless, within the first year ofoperation, at least 2 of the 5 generators experienced serious problems resultingin a loss of service and during the first year of operation most of the sites wereunable to obtain diesel fuel at least once due to shortages in the country.

2. PVP: reliability and maintenance requirements.

Water pumping has long been the most reliable and economic application ofsolar-electric (photovoltaic, or PV) systems. Most PV systems rely on batterystorage for powering lights and other appliances at night or when the sun isn’tshining. Most PV pumping systems do not use batteries – the PV modulespower the pump directly. Instead of storing energy in batteries, water ispumpedinto storage reservoirs for use when the sun isn’t shining. Eliminating batteriesfrom the system eliminates about 1/3 of the system cost and most of themaintenance.

Without batteries, the PVP system is very simple. It consists of just 3components: the solar array, a pump controller and thepump. The only movingpart is the pump. The solar modules are warranted to produce for 20-25 years.The expected life of most controllers is 5-10 years. Pump life can vary from 5 -10+ years (and many are designed to be repaired in the field). Unless the pumpor controller fails, the only maintenance normally required is cleaning thesolar modules every 2- 4 weeks! This task obviously can be done cheaply bynon-skilled local labor.

3. Maintenance requirements for diesel generators.

In contrast to the minimal needs for PVPs, the operation and maintenanceneedsfor diesel engines are both extensive and expensive.

Diesel engines require minor service, major service, and major overhauls atregular time intervals. The following is an excerpt from the Namibian studyquoted above:

“ A minor service includes oil change (topping up of oil included) and air, fuel

and oil filters.

A major service includes decarbonisation, adjustments, oil change and filterreplacements and requires skilled personnel which is assumed to be in theregion or at a close-by service center.

An overhaul includes the tasks of a minor and major service, replacement of

parts (e.g. crankshaft) and drilling of cylinders and requires skilled personnel.

The following schedule has been selected for the replacement intervals ofhigh quality ( Lister) and low quality engines (e.g. of Indian manufacture):”(Note: units are in hours of operation)

The following chart is based from the same Namibian study where the typical

costs (in Namibia) for the various service operations were calculated. In our

chart we have converted the costs to U.S. dollars. (1$ = 7.55 $N)

4. PV reliability assessments from the field.

We can talk all day long in our offices about reliability in the field, but thesatisfaction of the user and the service they receive are what really matters.Thefollowing study goes back and looks at the performance of PVPs in Mexico 10years after 206 systems were installed. The study is from a paper presented atthe American Solar Energy Society Conference in Portland, Oregon in July,2004. It is entitled “Ten-year Reliability Assessment of Photovoltaic WaterPumping systems in Mexico”, by Robert Foster, New Mexico State University; M.Ross, C. Hanley and V. Gupta, Sandia National Laboratories and several otherauthors from Mexico. The following chart illustrates a high degree of usersatisfaction with pumps.

The study summary states that after 10 years 60% of the systems were stilloperating appropriately. The most common cause of failure was pump failure,

which could also have happened if the system was powered by dieselgenerators. There were no failed solar modules. The report offers noinformation on how the systems were sustained, but suggests that the pumpswere all used by private farmers and ranchers and that they were eachresponsible for maintaining their own systems.

We have experienced the reliability of solar pumping systems in the field andhave seen that they have a high “survivability factor”, even with little or nocare atall. However, SELF is a firm believer that there must be sustainability measuresdeveloped for each project to ensure maximum life and service. SELF takesgreat care with each project to make sure there are financial, technical, re-supplyand organizational systems in place for sustaining the pumping systems.

In terms of energy security, it is highly possible that there will be continuingperiods of fuel shortages as richer countries compete for dwindling supplies –leaving the most disadvantaged without the ability to obtain fuel at any price.

Finally, the poorest of the poor should not have to suffer for the environmentalsins of developed countries and be limited in their options by only using thecleanest energy sources for development. But by the same token, the poormaysuffer more than most in the face of climate change and increasingly scarceresources. It’s in everyone’s interest that development and care for the climateand environment happen at the same time – especially when long- term cost andreliability favor the cleaner option.