Environmental and economic implications of small-scale CHP

Environmental and economic implications of small-scale CHP

Rodney Evans

This paper discusses the possible environmental and economic implications of small-scale combined heat and power (CHP) in the UK. The technology is based on internal combustion engines running on natural gas. Electricity is generated with high efficiency from a fuel that has the lowest carbon to energy content of the fossil fuels. Furthermore, this form of CHP has the potential to be applied very widely. Viable installations could range from 40 kW(e) in small commercial buildings to over 50 MW(e) on large industrial sites. Small-scale CHP therefore offers great scope for reducing emissions of C02, a major contributor to global warming.

Keywords: Environment; Economics; Small-scale CHP

Small-scale CHP can be defined as internal combustion engine (ICE) plant fuelled by natural gas with electrical outputs ranging from a few kilowatts up to about 10 MW or even more, and whose thermal output is used on site. Typically, the engines will be spark ignition or diesel engines, but gas turbines are now available with outputs over 1 MW. Conceiv- ably, fuel cell based CHP may come into use by the end of the century.

In common with large-scale CHP, electricity can be generated at much higher effective efficiencies than is possible with electricity only plant. Whatever the size, the environmental and natural resource implications are very similar given natural gas fuel- ling and high overall efficiencies. The economic implications in terms of competing with central power production also show broad similarity. However, this paper will focus on small-scale CHP as it is a relatively novel technology, whereas large- scale CHP is well established.

However, the application of large-scale plant in the UK has generally been restricted to very large industrial plants where all the heat output can be

Rodney Evans can be contacted at Seven Islands, South Road, Tetford, Horncastle, Lincs LN9 6QB, UK.

used on site. With the advent of plentiful supplies of natural gas, which is an excellent fuel for ICEs, the minimum economic size for a CHP plant has been greatly reduced; roughly speaking from megawatts to tens of kilowatts. This greatly increases the scope for CHP. This is especially true for space heating. A major advantage of small-scale CHP is that the space heating requirements of larger buildings (roughly over 1000 m 2 floor area) can be met by in-house CHP. This avoids the very high capital costs of a city-wide heat distribution system which has been a major barrier to district heating (at least in the UK). The existing gas system is used as the energy carrier instead.

The effect on emissions is also important. Those of interest are sulphur dioxide SO2, nitric oxides (NOx), methane (CH4), carbon monoxide (CO), and non-methane volatile organic compounds (NMVOC). At the very least, small-scale CHP could be expected to reduce SO2 emissions significantly because of the low sulphur content of natural gas. The effect on other emissions, though, is very technology dependent.

With regard to global warming, CHP has a major strategic advantage. As a means of using natural gas (as a coal replacement) to reduce CO, emissions, CHP is an extremely effective instrument, being three times as powerful as modern gas powered central generation plant (combined cycle) in this respect (in terms of CO2 saved in relation to gas used).

Even without allowing for its wider environmental benefits, there is reason to believe that the economic scope for small-scale CHP is substantial. Conceiv- ably, 10% or more of UK electricity demand could be met profitably from small-scale CHP.

This paper will first examine the technical and environmental aspects of small-scale CHP and then consider the economic issues.

Small-scale CHIP efficiencies

The keynote to CHP, large or small, is the very high proportion of the total energy input that can be

0301-4215/93/010079-13 ~) 1993 Butterworth-Heinemann Ltd 79

CHP series - environmental and economic implications of small-scale

Low grade heat) (from the cooling system)

Fuel i n p ~ u t 8-10

1 0 0 /

, , / 10-20

High grade heat (from the exhaust system)

20-40 >Electricity

Heat losses from flue and casing

Figure 1. Energy balance of a typical engine driven CHP unit (% of fuel input).

recovered in a useful form as electricity or heat. The ratios between these two will depend on the efficiencies of the generating and heat recovery equipment.

It should be noted that all the efficiencies quoted in this paper are based on the higher or gross calorific values (HCV or GCV) of fuels. With natural gas, HCV efficiencies are about 10% lower than those based on lower or net calorific value (LCV or NCV) which excludes the latent heat from water vapour in the combustion products (for coal and oil, the respective differences are about 4% and 6%). It has become almost standard practice worldwide to quote efficiencies in LCV terms. This is because the latent heat has been virtually useless as an energy source. However, with space heating and in other low temperature requirements, this heat can now be harnessed with condensing heat exchangers, given a reasonably clean exhaust gas.

Current small-scale CHP plant can produce electric power with efficiencies between 20 and 40%. The highest efficiencies, though, are achieved only in diesel or spark-ignition engines of advanced de- sign (such as lean-burn), and these will tend to be the larger units. A range of 25-33% is likely to be typical of most small plant. A high proportion of the balance of CHP's energy output, which is heat at various temperatures, can usually be recovered at modest cost. Figure 1 shows the energy balance of a CHP unit based on an ICE. The total efficiency of conversion (electrical plus thermal) is usually about 80% (eg with a generation efficiency of 25%, the heat recovered would be 55% of total energy input).

The heat-to-power ratio (HPR) is the ratio of

CHP

useful heat to electrical energy (eg the HPR would be 2.2 in the preceding example). Inevitably, the HPR declines as electrical efficiency increases (given that total efficiency remains constant). HPRs for ICE based CHP range from about 2.5 to 1.0.

Modern gas boilers (non-condensing) also achieve efficiencies of around 80%. If condensing heat exchangers were fitted with small-scale CHP plant, the total efficiency could rise to about 90% because of the extra heat recovered. Where this would be worthwhile, a condensing boiler, also about 90% efficient, would be the appropriate conventional heat source. In other words, the total efficiencies of small-scale CHP plants and the boilers whose output they would displace will be virtually the same, given that a similar level of heat exchange technology is applied to both.

This has an interesting consequence. Compared to the gas boiler, the CHP unit necessarily uses more gas to satisfy a given heat demand because of its lower thermal efficiency (eg 55% versus 80%). The reward for the extra gas consumption is, of course, the electricity produced. It can be demonstrated that if the overall efficiencies of the CHP unit and the conventional boiler are the same, then the ratio of the electricity produced to the incremental gas consumed is equal to the total efficiency of the plant (in the range of 80--90%).This ratio could be termed the equivalent or effective efficiency of generation. In this paper, the efficiency of CHP in its role as generating plant will usually refer to the ratio of electrical energy produced to incremental fuel energy consumed (compared to the conventional boiler that is displaced by CHP).

Efficiencies of 80% or more are far in excess of the generation efficiencies ever likely to be achieved by mechanical prime movers (even fuel cells are hard put to achieve 50%). It is this extremely high effective efficiency that makes CHP potentially so attractive. The effective efficiency can be demonstrated with a simple example. Consider a CHP unit with a total efficiency of 80%, 30% electrical and 50% thermal. If a useful heat load of 100 units is met by a gas boiler also of 80% efficiency, then 125 units of gas will be consumed. If the heat load is met by the CHP plant instead, then 200 units will be consumed. However, the extra 75 units of gas allows the generation of 60 units ofelectricity - hence the efficiency of 80%.

Curiously, the effective efficiency is unaffected by HPR given that the total efficiency of the CHP and conventional plants are the same. None the less, lowering the HPR (by increasing the electrical efficiency) remains a very desirable goal since it

80 ENERGY POLICY January 1993

CHP series - environmental and economic implications of small-scale CHP

Table 1. Gaseous emissions from energy use and their environmental impact.

Emission Formula Carbon dioxide C O 2 Methane CH4 Nitrous oxide N20 Sulphur dioxide S02 Oxides of nitrogen NOx Carbon monoxide CO Non-methane volatile NMVOC

organic compounds

Environmental impact

Global warming

Acid deposition

Photochemical oxidants

Source: N.J. Eyre, Gaseous Emissions due to Electricity Fuel Cycles in the UK, Energy and Environment Paper No 1, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990, Table 1.

permits a greater quantity of electricity (the more valuable energy form) to be supplied from a given thermal base. Furthermore, excess electricity production is much more easily exported (via the existing power network) than excess heat. This implies meeting the heat demands rather than the power demands of the site to obtain the maximum economic benefit (given that exported electricity can find a good market).

In this paper, an average HPR of 2.0 and total efficiency of 80% will be assumed for the environmental and economic assessment, implying generation and thermal efficiencies respectively of 26.7% and 53.3%.

The efficiency of the conventional generating plant whose output would be displaced by small-scale CHP must now be considered. It is reasonable to suppose that CHP electrical output in the UK in the near future would displace output from coal fired power stations as these meet most of the electricity demand, excepting the extreme peaks (now met mainly by pumped storage) and the bottom baseload (met by nuclear and some renewables). Some output may also be displaced from oil fired steam power stations which are used to a limited degree in the UK, but this will not make a significant difference to the analysis presented in this paper.

Since small-scale CHP is well suited to providing space heating loads, its electrical output can be expected to be concentrated in the winter period. At this time, the efficiency of conventional stations at the margin is around 33% (baseload steam stations have slightly higher efficiencies). However, the final efficiency of delivering electricity to the consumer can be significantly lower because of transmission and distribution (T&D) losses.

It has been argued that the marginal T&D loss in winter for power delivered to low voltage consumers (who would be the potential users of most small-

scale CHP) is around 16% .2 This may seem high, but it reflects the fact that resistive losses (mainly in cables) are the dominant T&D loss. It is easy to show that the marginal resistance loss for a small increment of power on the system is double the average rate (this follows from Ohm's law). The introduction of small-scale CHP output would produce a nearly corresponding reduction in load on the T&D system, given that the bulk of the CHP electrical output would be consumed on site or nearby (thereby incurring few losses of its own).

Average resistance losses in the UK are around 6.5% rising to over 8% in winter, hence the marginal loss of 16%. Another consequence of high marginal losses is that extra central generating capacity is needed to overcome them. It has been estimated that the marginal loss rate at peak times can reach 20%. 3 Thus 1 kW of small-scale CHP capacity can replace 1.2 kW of central capacity - which has important economic benefits.

With a T&D loss rate of 16%, the final efficiency to the consumer will be reduced to 28.5% given a

~ O generating efficiency of ~ %. This figure will be used in subsequent calculations of the environmental effects. It should be noted, though, that the loss rates in respect of power displaced by large-scale CHP may be much lower. The users of such plant will inevitably be connected further 'upstream" in the T&D system at high voltages, and so losses will be lower.

Environmenta l impact

The gaseous emissions of principal interest from energy use and their environmental impact are set out in Table 1. Methane is a powerful greenhouse gas, being over 100 times as potent as CO2 on a weight for weight basis. However, its atmospheric lifetime is very short, so its long-term global warming potential is only around 10 times that of CO2. The distinction between methane's short- and long- term effect should be borne in mind when consider- ing strategies for combating global warming (a point which Jackson et al seem to have overlooked). 4 Nitrous oxide is also a powerful greenhouse gas and one with a long lifetime. Reliable data on its emissions from energy-related plant is not yet available. However, levels are generally believed to be modest; so the N20 issue would probably not be important in the context of CHP and will not be considered any further here. Particulate emissions are produced by both conventional and CHP plant, but are generally very low from modern installations, and so neither will these be considered here.

ENERGY POLICY January 1993 81


Table 2. Unit CO2 and SO2 outputs for fossil fuels.

Fuel (and approximate H/C ratio)

Natural gas (4: 1) Heavy fuel oil a'b (2: 1) CoaF (1: 1)

Unit ouputs (g/GJ of fuel consumed) CO2 SO2 50 600

73 300 1300

88 400 1140

tric) of displaced conventional power output; and

• the specific emissions per useful kilowatt hour (electric) of displaced conventional heat output multiplied by the heat to power ratio of the small-scale CHP plant.

Potentially, the environmental advantages of small- scale CHP stem from its very efficient use of primary energy compared to the conventional heat and power sources that it would displace. There is a further advantage in that the predominant fuel for small- scale CHP would be natural gas, which has the lowest carbon content of any fossil fuel and is also virtually sulphur free. The low CO2 output from natural gas is due to its high hydrogen to carbon (H/C) ratio. Table 2 shows the CO2 and SO2 output from the combustion of coal, heavy fuel oil and natural gas (as typically used in the UK).

Furthermore, small-scale CHP would be an extremely efficient means of using natural gas, both for electricity production and CO2 abatement. Or, in other words, if the use of natural gas is seen as a means of reducing CO2 emissions, small-scale CHP would be the most effective technology to employ. This is due to its high total efficiency.

The specific emissions of conventional heat and power sources are given in Table 3. In general, small-scale CHP could be expected to displace gas fired boilers. However, some displacement of larger boiler plant fired by coal and heavy fuel oil (HFO) in the public and industrial sectors could also occur, and so simple average values for these plants are given.

Also included are the principal indirect emissions. These are CO2 from the energy used in producing these fuels, and CH4 from coal extraction and gas wells. Indirect CO2 emissions are about 2.5% of direct levels for gas and coal, but nearly 10% for oil products on average because of heavy refinery energy use.

The indirect methane emissions deserve special comment. Substantial quantities of methane are released in coal mining (about 500 gm/GJ of coal in the UK - of which only 10% is recovered), and thereby constitutes one of the principal hazards in that industry. In UK continental shelf gas production, losses are believed to be very small at 0.2%, and only this figure is included in Table 3. A much larger quantity, 1% or 180 gm/GJ of gas throughput, is thought to be lost in distribution leakages, mainly at the local, low pressure end of the system (although this figure is disputed - see Jackson et al

who argue that that figure may be 5%). 5 However, distribution leakage is irrelevant in the present con-

Notes: aTypical sulphur content of 2.9%. t'The CO: output from lighter fuel oils is much the same. However, the SO_, output from such oils is usually much less. "Typical sulphur content of 1.6% (with 10% sulphur retention in ash).

Source: R.D. Evans, Environmental and Economic Implications of Small-Scale CHP, Energy and Environment Paper No 3, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990, Table 3.2.

The deployment of small-scale CHP would affect the environment primarily in two ways:

• there will be new emissions from the CHP plant itself, but these are then counter- balanced by

• the reduction in old emissions from the conventional heat and power sources (from heat- only boilers on the CHP site and central power plants) displaced by the useful output of the small-scale CHP plant.

Of minor importance, there will be similar effects due to changes in the production and distribution of the fuels used in these plants.

The overall effect will therefore be determined by:

• the degree to which small-scale CHP displaces conventional heat and power; and

• the specific emissions in weight of pollutant per useful unit of electrical or thermal energy from small-scale CHP and conventional heat and power sources.

The degree of displacement is an economic issue which is considered shortly. None the less, the feasible impact at national level will be illustrated by supposing that smale-scale CHP provides 10% (about 25 TWh) of total electricity consumption in the UK.

The specific effect on emissions due to displacement can be conveniently calculated in terms of the electrical output. The thermal output will be the electrical output multiplied by the heat to power ratio. Thus the net unit change in emissions per kilowatt hour (electric) of small-scale CHP output will be:

• the specific emissions per kilowatt hour (elec-


ClIP series - environmental and economic implications of small-scale CHP

Table 3. Emission factors for conventional heat and power sources (g/kWh of useful thermal or electrical output).

Industrial/commercial Coal-fired boilers fired by

Emission power stations a CoalfltFO b Gas ¢ Direct (average)

CO2 1120 390 228 SO 2 14.4 5.8 NOx d 5.5 1.0 0.23 CO 0.15 0.30 0.06 CH4 d 0.008 0.012 0.006 NMVOC 0.02 0.01 0.0015

Indirect CO2 28 23 5 CH 4 5.9 1.1 0.2

Notes: aAssumes 33.0% generating efficiency and 16% marginal loss in transmission and distribution to give a final efficiency to the consumer of 28.5%. UAssumes 75% boiler efficiency. ~Assumes 80% boiler efficiency, dAs NO,.

Source: R.D. Evans, Environmental and Economic Implications of Small-Scale CHP, Energy and Environment Paper No 1, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990, Table 3.2 for direct emissions; and N.J. Eyre, Gaseous emissions due to electricity fuel cycles in the UK, Energy and En- vironment Paper No 1, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990, Table 1 for indirect emissions.

text since it is essentially a standing loss (due to the need to maintain a reasonably constant pressure in the system). Consequently, any change in gas use resulting from the uptake of small-scale CHP will not affect the distribution loss. This loss becomes an important issue only if widespread closure of the local gas distribution system is contemplated (as indeed it is by Jackson et al).

Emission factors (except CO2) for a variety of plant suitable for small-scale CHP are given in Table 4. CO2 emissions per kilowatt hour (electric) are of course directly dependent upon electrical efficiency. Including a 2% allowance for 'indirect' emissions, the CO2 emission factor for CHP, F, is related to electrical efficiency, E, by the following formula:

F = 186/E g/kWh(e)

There will be a similar relationship for specific 'indirect' methane emissions, M:

M = 0.13/E g/kWh (e)

- assuming a loss rate of 0.2% at the well-head as above.

Electrical efficiencies for small-scale CHP can range from 20 to 40%, although 25 to 33% is more typical of the types of plant in Table 4. For illustrative purposes, an electrical efficiency of 26.7% is

Table 4. Emissions from small-scale CHP plant (except COz) (g/kWh of electrical output).

Engine type and air-fuel ratio (AFR) NOx a CO HC h Conventional c

AFR: 15.5 10-16 37-60 2.0-2.3 16.0 12-21 13-22 2.0-3.0 17.0 14-27 1.3-2.6 1.3-2.5

Conventional with catalytic exhuast converter c

AFR : 15.95 (i) Into catalyst 14 11 2.7 (ii) Out of catalyst 1.0 2.7 1.3

Lean burn ¢ AFR: 28.0 2.7 4.0 1.3

Gas turbine AFR: about 50-100 2.7 0.7 0.3

Notes: aAs NO,. bHydrocarbons, 90--95% CH4, remainder NMVOC. cSpark-ignition reciprocating engines.

Source: R.D. Evans, Environmental and Economic Implications o f Small-Scale CHP, Energy and Environment Paper No 3, Energy Technology Support Unit, Ha~'el l Laboratory, Didcot, UK, 1990, Table 3.3.

assumed here, which is consistent with a total efficiency of 80% and a heat to power ratio of 2.0.

The high levels of NOx from conventional reciprocating engines without pollution control would almost certainly be unacceptable if small-scale CHP was deployed widely. Current US and Japanese regulations in major conurbations limit NOx emissions to an NO2 equivalent of about 3 g/kWh(e). Therefore it is safe to assume that the levels for low NOx plant would prevail. A straightforward way to reduce NOx is to equip the engine with a catalytic exhaust converter, similar to those used on cars - hence the precise AFR. In the longer term, lean- burn technology is a more attractive option because it can offer greater electrical efficiency (a relative improvement of 7-9%).

Table 5 sets out the range of specific emissions that may be experienced with future plant and the 'average' values assumed here. From the data in Table 3 and Table 5 (average values) it is easy to calculate the net change in emissions per kilowatt hour of electrical output from small-scale CHP, given a heat to power ratio of 2.0, and the subse-

Table 5. Specific emissions of future small-scale CHP plant output ( g / k W h ( e ) ) .

CO2 = NOx b CO CH4 c NMVOC d Range 465-930 e 1.0-2.7 2.7--4.0 0.8-3.0 0.0-0.1 Average 697 f 2.0 3.3 2.5 0.1

Notes: alncluding indirect emissions at 2% of total. UAs NO,. ~In- cluding indirect emissions at about 0.5 g/kWh(e), dAssuming NMVOC is 10% of direct HC (Table 4). eEfficiency range of 20-40%. fEfficiency of 26.7%.


CHP series - environmental and economic implications o f small-scale CHP

Table 6. Effect of small-scale CHP on gaseous emissions from heat and power production.

Emission C02 S02 NO:fl CO CH4 NMVOC u

National net reduction Estimated net reduction in total emissions % in total emissions in emissions per kWh of assuming that about from all sources with electricity produced by 10% (25 TWh) share of a 10% share of total small-scale CHP total electricity demand electricity demand displacing coal-fired is met by small-scale met by small-scale generation (g/kWh) CHP (t/y) CHP 1000 25 000 000 4

17 430 000 12 4.6 110 000 5 (3.0) (75 000) (1.5) 3.9 100 000 2.3 (0.7) (2 000) (0.1)

Notes: ~As NO,. bSmall-scale CHP would result in a net increase in these emissions.

quent national impact of small-scale CHP providing about 10% of national electricity consumption or 25 TWh. The results are presented in Table 6, together with an estimate of how national changes in emissions compare with present background levels. It is assumed that 25% of the displaced conventional heat production came from coal/HFO fired boilers.

As can be seen, a 10% market share for small- scale CHP can achieve significant reductions in some of the major emissions, particularly of the acid gases, at the cost of very small increases in the smog related pollutant - CO and NMVOC. The very small increase in N M V O C is reassuring as this is much more important, weight for weight, than CO.

It may be objected that the large reductions in acid gases could also be achieved by flue gas clean up at power stations. However, this can be costly, as will be shown shortly.

This analysis is based on existing generation patterns in the UK with its preponderance of coal fired plant. However, the construction of large amounts of combined cycle gas turbine (CCGT) plant is underway in the U K with plans for even more. So, a discussion of the strategic merits of CHP versus CCGT in the context of global warming and resource use is in order.

The greater use of natural gas is an attractive route to reducing CO2 emissions because of its relatively low carbon content, its scope for greater efficiency in use, and low acid gas emissions, compared to coal. C C G T would seem to be a good candidate for deploying natural gas. CCGT would achieve a net efficiency to the consumer of 38%, given a generating efficiency of 45% and marginal transmission and distribution losses of 16%. Allow- ing for indirect emissions of 2%, this would give a total CO2 emission of 490 g/kWh electrical output. This would give a total saving of 660 g/kWh com-

pared to coal (from Table 3). This is two-thirds of the saving that could be achieved by CHP, 1000 g/kWh (Table 6). So, it might seem that the incremental benefit of CHP compared to CCGT is rather modest.

This perspective changes dramatically, though, if the resource aspect is considered. Natural gas is fairly plentiful in relation to current consumption in Europe and the CIS (former USSR) which effective- ly have a 'common market ' in gas with an integrated supply system. However, the reserve to production ratio on this market is only about 50 years. If there is a large swing to gas as a replacement for coal in this area, the pressure on reserves would become much more acute. For example, in the UK a complete replacement of coal by gas for electricity generation (in CCGT) would nearly double current national gas consumption, and this is in a country where gas is nearly the universal fuel for all purposes except generation and transport. So, from a somewhat longer term view - say 100 years - it would be prudent to consider gas as a very 'finite' resource (in contrast to coal whose reserves run into centuries at current consumption rates). And this period of 100 years or so is one in which large reductions in CO2 emissions may be required to prevent major climate change (although the position is still very uncertain).

Consequently, it would be useful to see how effective a given quantity of gas would be when used as a coal replacement in various ways to reduce CO2. The following formula would be useful as it gives the unit CO2 savings (kg/GJ) achieved by gas when substituted for coal:

S = 8 8 . 4 ( E c , / E c ) - 50.6

where


800

700

600i

"E u

500 [ I ] I I

0 100 200 300 I I I

q 0 0 5 0 0 600

Electrical output (kW)

Figure 2. Installed capital costs as a function of size for gas fired spark ignition units.

S = savings in kg CO2 per GJ of gas EG = efficiency of using gas Ec = efficiency of using coal (Ec/Ec) = GJ of coal displaced per GJ of gas; and 88.4 and 50.6 are respectively the unit emissions in kg of CO2 per GJ of coal and gas

C H P series - environmental and economic implications o f small-scale C H P

much electricity. The strategic advantages of CHP in the context of global warming and limited gas reserves are thus quite clear. CHP represents by far and away the most effective means of using gas as a coal replacement for reducing CO2 emissions. The only exception to this would be where gas was used in high efficiency appliances to replace direct electric heating (which would give similar unit savings in kg CO2/GJ to CHP).

Small-scale packaged CHP costs

This section gives a brief overview of capital and operating costs, and the current state of technical development for 'packaged" CHP (under 500 kW(e)). Larger plant is discussed elsewhere. 6 In- formation on typical small-scale CHP capital and maintenance costs is given in two EEO (Energy Efficiency Office) publications, and only the main features will be summarized here. 7 Figures 2 and 3 show how these costs, updated to 1992 levels, vary with plant size. In both cases, 40 kW(e) represents a fairly critical size, in that unit capital and operating costs rise sharply as size falls below this threshold, but fall only moderately as size increases above 200 kW(e). For this reason, the view is taken in this paper that the economically worthwhile potential for small-scale CHP will usually lie in sites that can accommodate the heat output from units of 40 kW(e) and above.

If gas were used as a straight replacement for coal in the boilers of present power stations, the CO2 saving would be about 40 kg/GJ. If used as a replacement for coal in smaller, heat-only boilers, the saving might be 50-60 kg/GJ because of the somewhat higher efficiency of gas in such a plant. When used in CCGT plant in place of conventional coal plant, the saving would be about 70 kg/GJ (given an increase in overall efficiency from around 28 to 38%). The electricity produced by CCGTs would be 105 kWh/ GJ of gas input.

When gas is used in a small-scale CHP plant, displacing not only central generation, but also heating plant, the savings are much greater. At 80% total efficiency and a heat to power ratio of 2.0, a CHP plant achieves a saving in CO2 emissions of about 200 kg and an electricity output of 220 kWh for each incremental gigajoule of gas consumption (compared to the gas used in the displaced heat-only boilers).

In other words, a given quantity of gas routed through CHP could save three times as much CO2 as it would if used in CCGT, and yet produce twice as

1.5

1.3

"~ 1.1

u 0.9

¢-

.c 0.7

0.5

0 I I I I I I I 50 100 150 200 250 300 350

Electrical output (kW)

Figure 3. Average maintenance costs.


C H P series - environmental and economic implications o f small-scale

Typical installed capital costs will vary from £800/ kW or more for 40 kW(e) plants and then down to £500/kW for plants of 200 KW(e) and above (1992 prices). Broadly speaking, the capital costs of the units as supplied by the manufacturer accounts for 65-80% of total capital costs, with installation accounting for the rest.

Small-scale CHP units are composed largely of components that are mass produced, especially the engines and alternators. Other components such as the heat exchangers and electrical control equipment may be customized to some degree, but none the less can take advantage of standard industrial products. For these reasons, the basic costs of a CHP unit in terms of materials and components can be £300/ kW(e) or less. Some of the primary items can be very cheap indeed eg the engine and alternator for a 40 kW(e) unit are respectively £3000 and £1000 off the shelf, compared to a total package price of £30 000.

However, substantial overhead costs are incurred by suppliers of small-scale CHP plant. These include not only the necessary engineering work involved in assembling the bought in components into a complete package, but also marketing, sales and customer support. Because current volumes are so small - perhaps a hundred installations a year in the UK - the unit overhead costs are high. For this reason, the selling price of CHP units is often much greater - by up to 70% - than the basic manufacturing costs. The low volumes also increase the costs of installation and servicing.

The view has been expressed in the industry that a large market - say in excess of 1000 units a year - could reduce overall capital and installation costs by up to a third. Thus the ultimate long-term unit cost for a small-scale CHP installation could be as low as £400--500/kW(e) installed. Given the vast technical potential for small-scale CHP (a notional 14 GW in the commercial sector alone), the market should ultimately be large enough to give good economies of scale, eg even with 1000 40 kW(e) units sold a year, it would take 25 years to reach 1 GW installed capacity.

The maintenance costs charted in Figure 2 are quite comprehensive in that they include not only routine servicing (every 1000 hours), but also major overhaul (10 000-15 000 hours) and the provision of reconditioned engines as necessary (20 000-30 000 hours). Provided this service schedule is adhered to, a unit life of 100 000 hours is realistic. Typically, 'routine' maintenance accounts for 75-80% of the total maintenance cost. This is why unit maintenance costs increase sharply with the smallest units, be-

CHP

cause of labour overheads. It is now normal practice for the supplier of a unit

to undertake the maintenance on a long-term contract with the cost based on a fixed price per hour of operation. This is a very important development, since it ensures that the unit is serviced properly by qualified personnel, thus making it very reliable, and the user knows what his costs will be with a high degree of certainty. Further, this removes a major barrier to the deployment of small-scale CHP in that the user need not have any technical skills in house. The user can thus regard the CHP unit as a black box which is safe in someone else's hands.

The black box philosophy has been aided by the development of remote monitoring systems. These are installed with the unit and connected via a telephone link to a computer at the manufacturer's. The computer monitors the operation of the unit, sometimes corrects minor faults that might other- wise lead to premature shut down (by the built in safety systems), and if necessary alerts the manufacturer that prompt action by a service engineer is needed.

Such measures have brought about very high reliability. Manufacturers aim for an annual availa- bility of 95%, and even 99% has been achieved. For these reasons, small-scale CHP can now be regarded as a proven technology. Prospective users are thus almost entirely free of technical risk.

For the purpose of the economic assessment in this paper, a maintenance cost of 1.0p/kWh will be assumed. This is a fairly conservative figure as it is based on the expected costs for a 40 kW(e) unit with some rounding up. A major supplier now quotes 0.9p/kWh for 40 kW units and 0.7p/kWh for larger units. However, a conservative approach will be adopted here because of the possibly adverse effects of intermittent operation. Current units are generally run for at least 4000 hours pa, but the assessment will concentrate on units geared primarily to space heating loads - where about 2500 hours/pa would be more likely.

With current technical developments, it is possible that the service interval may be extended to 2000 hours or more. This could be achieved with higher quality lubricating oils and longer-life spark plugs (the life of these items is the major constraint on the length of the service interval). Already, 4000 hours of operation have been achieved on an experimental basis using synthetic lubricating oils before change of oil was required.

Because routine servicing accounts for so much of the overall maintenance costs, a doubling of the service interval could reduce these costs significant-



ly, especially for the smaller units where labour charges are dominant.

Economic aspects

This section considers the economic benefits that small-scale CHP could offer. First, it would be useful to give a brief description of the advantages of CHP as a generating source in the electricity supply system.

Although CHP plant would be outside the control of the electricity supply system since its output would be governed by on site heat demands rather than the public requirements for electricity, it can none the less be considered virtually as valuable as conventional plant which is within the control of the system. This is because of diversity or safety in numbers. While the output of any individual plant is liable to be fairly unpredictable, the aggregate output of thousands of plants operated indepen- dently will be predictable with a very high degree of confidence. This follows from basic statistical principles.

CHP plant which supplied baseload heat demands (usually industrial) would have an aggregate electrical output which would be fairly constant throughout the year, perhaps falling somewhat at night and during the holiday season in summer - ie times when public demand for electricity falls anyway. CHP plant which supplied mainly space heating demand (mainly in non-industrial larger buildings) would have an aggregate output which rose in winter with decreasing external temperature and would be at its highest in daytime Monday to Friday (following normal patterns of building occupancy). This would follow very closely the pattern of public electricity demand in the UK which rises significantly in winter with a strong degree of dependence on weather conditions. In other words, the overall pattern of CHP electrical output would synchronize very closely and predictably with overall demand on the supply system, not only seasonally but diurnally as well.

This close synchronicity means that a kilowatt of CHP capacity is as effective in ensuring security of supply as a kilowatt of central plant capacity (possibly even more so since CHP electrical output would incur significantly lower peak distribution losses for the reasons discussed earlier in Section 2 of this paper). Thus the unit value of CHP capacity is closely comparable to the unit cost of central capacity.

Operating costs have to be considered as well, and so the costs of central plant that would compete with CHP have to be considered in some detail. For

possibly the next ten years or so, the bulk of new central generating capacity in the UK will be in the form of combined cycle gas turbine plant (CCGT), judging by the current investment plans of the major electricity companies. So CCGT can be viewed as the competition for CHP.

Firm figures for the capital costs of CCGT are not readily available, but judging from press reports of plans for CCGT plant, unit capital costs appear to be in the region of £400-500/kW. Roughly speaking, then, the capital costs of CCGT and small-scale CHP would be similar once the possible benefits for the latter from mass production and lower peak distribution losses were taken into account.

Although CHP has twice the effective efficiency of CCGT, the advantage in unit fuel costs may be much less or even eroded completely because of tariff differences. It is thought that gas prices for the new CCGT plants are about £1.50/G3, whereas the price for prospective small-scale CHP users would generally be much higher at £2.50-4.50/GJ, depending on size of customer and contract conditions. None the less, it should be noted that British Gas have recent- ly proposed increasing the price of gas for CCGT stations precisely because of concerns about demand out-stripping supply.

However, it should be noted that the highest price would be charged to smaller users whose main use of gas would be for space heating and who would have 'firm' supplies of gas (CCGT and many large CHP users could take advantage of the cheaper 'inter- ruptible' tariffs given that their equipment could be fuelled by light distillate oil if necessary - gas turbines and diesel engines would be the usual choice of plant for larger CHP installations). Conse- quently, CHP sourced electricity produced by the smaller users would be produced principally in the winter months when it can be of much higher value than average.

Thus the overall costs of electricity from CCGT and CHP are probably fairly close at current tariffs. However, the position may be different in the longer term assuming that:

• significant reductions in CO2 emissions are needed in order to combat global warming; and

• gas is a relatively scarce commodity (or would become so if used as a wholesale replacement for coal in electricity generation).

Therefore it would be useful to consider the value that could be placed on gas by CCGT and CHP users respectively in the contexts of scarcity and global warming. Imagine that CHP and CCGT users were


ClIP series - environmental and economic implications of small-scale

rival bidders for a given gas supply. Here, the gas should in theory go to the party with the highest bid, ie the one who can make the most efficient use of the commodity in economic (if not necessarily physical) terms. In general, the value that a business would place on a unit of a commodity (ie the maximum price it would be prepared to pay for it) is equal to the value of the outputs produced with its aid less the total of other costs incurred.

As shown earlier, a unit of gas consumed in CHP can produce twice as much electricity and save three times as much CO2 (compared to coal fired generation) as a unit of gas consumed in CCGT plant. Also, because twice as much electricity is produced, twice as much SO2 would be saved per unit of gas in CCGT (again compared to coal fired plant which can be taken as the baseline electricity source). The electricity production and environmental gains rep- resent the unit outputs.

Although the export or buy back tariffs for inde- pendently produced electricity are no longer speci- fied centrally after the restructuring of the British electricity industry, those in force prior to privatiza- tion give a guide to the minimum unit value of CHP produced electricity (if the CHP output is used to reduce purchases of power rather than exported, the financial benefits are greater). A CHP plant used primarily for heating loads - and which was sized to meet most of the peak heating load - would produce about 2500 kWh of electricity a year per kW(e) of capacity. At the export tariff prevailing in 1990, this would have an average unit value of 4.5p/kWh given that the electricity is produced at the times of highest demand (winter daytime), s

It may seem surprising to suggest that CHP plant should be installed with such low load factors (about 30%) in prospect. This goes against the conventional wisdom that CHP is only economic if it has a high load factor. However, this assumes that the value of the electricity produced is constant throughout the year - which is not so because of the much higher demand in winter. If the electricity tariffs accurately reflect the seasonal and diurnal variation in value (as with the seasonal time of day or STOD type tariff which is gradually coming into use in the UK for non-domestic customers), then it can be shown that the profitability of a CHP installation may be barely affected by load factor provided it is well utilized during the 1000 hours a year (winter weekdays in daytime) when electricity prices are at their highest. 9 The reason why electricity prices should be significantly higher in winter is that conventional generating plant also faces equally unattractive load factors.

The monetary value of CO2 savings is extremely

CHP

difficult to quantify at present. But for illustrative purposes, this could be taken as 2p/kg, which represents the unit cost of the cheapest option for remov- ing CO2 from flue gas (where part of the cost is offset by the benefits from injecting the CO2 into oil and gas wells to enhance their output). This figure is equivalent to a carbon tax of about £70 (or US$120) per tonne of carbon which is at the lower end of the estimates of the economic costs of reducing fossil fuel use that have emerged from some recent studies (eg as discussed in Nordhaus). Such a figure is also only a little more than that implied by recent Euro- pean Community proposals for an energy tax equivalent to US$10/barrel of oil (applied across the three main fossil fuels this would give a cost per ton of carbon of about £45, £55 and £75 respectively for coal, oil and gas).

The implicit value of SO2 savings is equivalent to roughly 0.5p/kWh of electrical output given the costs of flue-gas desulphurization. 1° Further benefits could accrue in respect of NOx removal, but these will be smaller and will be ignored here for sake of simplicity.

Then capital and non-fuel running costs have to be taken into account. For illustrative purposes, assume that capital costs for both CHP and CCGT are the same at £450/kW(e) (given economies of scale from mass produced CHP). At a real discount rate of 6%, the annual capital charge per kW(e) of capacity would be £30--35 depending on amortization period. A kW(e) of CHP plant producing 2500 kWh of electricity annually would use about 11 GJ of gas a year. Therefore, the capital charge associated with each GJ of gas consumed by CHP plant would be about £3. The corresponding figure in respect of CCGT would be half that at about £1.50 (as CCGT would use twice as much gas as CHP per unit of capacity).

Maintenance costs for CHP will be taken as 1.0p/kWh as given earlier. For CCGT, a conservative figure of 0.3p/kWh will be assumed for these costs (a figure of 0.4p/kWh has been estimated by Evans in respect of part load operation). 11 With unit electricity outputs of 220 and 105 kWh per GJ of gas input respectively for CHP and CCGT, the associated maintenance costs per GJ of gas will thus in turn be 220p and 45p.

Table 7 compares the unit outputs and their values, and non-fuel costs per GJ of gas input respectively into CHP and CCGT plants using the data given earlier.

It will be noticed that gas input to CHP could have a net value of over £5/GJ more than if input to CCGT plant. This figure is well in excess of the gas


ClIP series - environmental and economic implications o f small-scale CHP

Table 7. Unit outputs and values, and costs in respect of natural gas consumed in CHP and CCGT plants.

Output and value/cost per GJ of gas consumed in: = Outputs Unit value CHP plant CCGT and values of output Output Value (£) Output Value (£) Electricity 4.5 p/kWh 220 kWh 9.90 105 kWh 4.73 COz savings 2p/kg 200 kg 4.00 70 kg 1.40 SO. savings b 0.5p/kWh 220 kWh 1.10 105 kWh 0.52 Total value of output 15.00 6.65 Costs Capital charges (3.00) (1.50) Maintenance costs (1.75) (0.45) Net value of output 10.25 4.70

Notes: aCosts are indicated as negative values, bso2 savings are directly proportional to electricity output and so they are expressed here in notional kWh (reflecting the displaced output from coal fired plant) for convenience (for calculating their monetary value) rather than in terms of actual quantity of emissions.

price differential that would probably prevail between the two types of plant, which is around £1-3/GJ at current tariffs. Even if the monetary value of the greenhouse benefits (from CO2 abatement) were ignored completely, the net value of gas input to CHP over CCGT would still at least equal the tariff difference. However, the imposition of the EC general energy tax would give CHP a significant advantage.

The advantage of CHP would increase in respect of larger baseload CHP plant (as in industry). Although the unit value of electricity would be less, this would be compensated by the lower gas prices and maintenance costs for the baseload CHP.

If the unit value of electricity production was based on the price for imported electricity, the advantage for CHP would be greater still since the unit value of savings in imported power can be worth up to 40% more than the unit revenue from exports.

It needs to be emphasized that these results depend on gas being viewed as a very finite resource. The economic benefit of CHP would decline if there were a good prospect that gas supplies could be maintained at current costs for many decades even if demand increased dramatically (as would be the case if gas replaced other fuels on a large scale - be it in central power production or other applications). In such circumstances, the price of gas would merit little or no scarcity premium over its present production costs. If gas remains a permanently cheap fuel, then the financial advantage conferred by CHP's greater conversion efficiency is necessarily reduced compared to a situation where gas supplies came under pressure. A full economic analysis is far too complex to be undertaken here, especially as it depends on discount rate assumptions.

However, the tentative conclusion of this paper is

that the use of gas may produce significantly greater economic benefits in CHP than if used in central power production given concerns about global warming and the relative scarcity of gas.

The scope for small-scale CHP

There is a very large technical potential for CHP (both large and small scale) in terms of the thermal base that could usefully accept the heat output from generating plant. This thermal base consists largely (about 80%) of heat needed at a temperature of 100°C or below in the form of space heating, domestic hot water, and industrial process water. The remainder of the thermal base consists of the requirement for process steam (about 15%) and hot gases in various applications (about 8%, although after-firing would often be needed here).

Table 8 gives a breakdown of the UK thermal base in terms of the annual pattern of sectoral usage, and temperature requirements. The figures are necessarily broad brush and should be regarded as no more than indicative for the future given changes in industrial structure, heating standards and energy conservation measures. With a heat to power ratio of 2.0 the potential electrical output can easily be gauged. The seasonal component which is proportional in space heating requirements would be 600 PJ or about 170 TWh (1 TWh = 3.6 PJ) which is well in excess of the current seasonal electricity demand of about 50 TWh. The baseload component would be close to 350 PJ or 100 TWh compared to current baseload electricity demand of 200 TWh. Possibly the baseload potential for baseload CHP could be significantly higher given that the larger CHP units, which have more scope for greater electrical efficiency and hence are lower heat to power ratio, would


CHP series - environmental and economic implications of small-scale

Table 8. Annual useful heat requirements in the UK (P J).

Pattern of heat requirement Sector SeasonaP Baseioad Total Industrial

> 300 °C (hot gases) - 150 b 150 b 100-300 °C (process 350 350 steam/hot water) < 100 °C (space heating) 200 - 200

Commercial 300 50 c 350 Domestic 750 150 ¢ 900 Total 1200 700 1900

Notes: aFor space heating, bAbout only half this requirement could be met by the sensible heat of gas turbine exhaust; after- firing with additional gas (or other fuel) would be needed for the remainder. CMainly hot water requirements.

Source: R.D. Evans, Environmental and Economic Implications of Small-Scale CLIP, Energy and Environment Paper No 3, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990, Table 6.1.

tend to be concentrated in sites with a steady heat load (mainly industry).

The economic potential for CHP, of course, is quite a different matter. Obviously much of the technical potential would be of no more than theore- tical interest because of the excessive costs of ex- ploiting it, although the economic potential could increase slowly over time with technical progress. With current technology, the economic potential is largely restricted to larger premises in the industrial and commercial sectors, and even this would take many years to develop. It is impossible to state with any precision what the present economic potential is but the limited information available suggests an upper limit of roughly 150 TWh or half the total technical potential of about 300 TWh (based on heat requirements for larger premises).

The achievable contribution in the next 20 years or so would undoubtedly be much less than this. None the less, it is believed that supplying 25 TWh or 10% of current total electricity demand in the UK from gas powered CHP plant is not unrealistic in the longer term. This could be in addition to the current CHP electrical output of nearly 9 TWh - almost entirely from plants over 10 MW(e) in industrial premises. It is not possible to draw up a precise dividing line for this new potential for 25 TWh between small scale - says under 10 MW(e) - and larger scale. Clearly, the commercial sector potential would be almost entirely in the small-scale range. A significant proportion of the industrial potential, though, would be best exploited by larger plant, in many cases by combined cycle gas turbines.

This paper concludes with a brief look at the prospects for small-scale CHP in the three sectors listed in Table 8. Industrial sector: a fairly detailed

ClIP

analysis of the prospects for new users of CHP in industry is given in Industrial Combined Heat and Power. 12 It was suggested that new capacity would consist of gas turbine plant with an average rating of 4--5 MW(e).

Commercial sector: spark-ignition engine CHP certainly becomes the appropriate plant for much of the commercial sector. Units with outputs as low as 15 kW(e) have been available, but the economics of small-scale CHP become attractive only for units of 40 kW(e) and above, as noted earlier in the discussion of small-scale CHP costs. Above this threshold, costs reduce fairly slowly with greater size.

With the aid of data presented elsewhere, it has been possible to provide a crude disaggregation of commercial sector energy use by size of premises in terms of floorspace, and from this infer the upper limit to the CHP capacity that could be provided by plants of various sizes.13 As a rough rule of thumb, commercial (and domestic) buildings in the UK (at current insulation standards) need to have space heating plant with a unit output of at least about 80 W(th)/m 2 of heated floorspace to ensure ade- quate comfort in midwinter (when the difference between internal and external temperatures would be around 20 K). With a heat to power ratio of 2.0, the feasible CHP electrical output per square metre of heated floorspace could be about 40 W(e) assuming that the CHP plant were sized to meet most of the heating load. Thus premises would need to have a minimum area of 400 m 2 to accept 15 kW(e) CHP units and 1000 m 2 to accept 40 kW(e) units.

Table 9 presents a summary breakdown of the commercial sectors and its CHP potential by floorspace size. Note that premises over 400 m 2 account for 70% of heated floorspace and 85% of useful heat consumption in the UK commercial sector.

It can be seen that up to 38 TWtgpa of electricity could be generated from premises over the 1000 m 2 threshold, and almost 40% of this would come from premises over 5000 m 2 where the economics can be especially attractive because of the considerable reduction in unit capital and maintenance costs for larger plant (Figures 2 and 3). Premises over 5000 m 2 could accept units of 200 kW(e) output.

It is for these reasons that a total small-scale CHP electrical output of 25 TWh/pa is not unreasonable in the longer term when both the industrial and commercial sector potentials are taken into account.

Domestic sector: although gas engine CHP is technically feasible for individual domestic premises, the disproportionately high costs of maintenance and emission control alone would make it unecono- mic at present. Stirling engine or fuel cell CHP is



Table 9. Distribution of floorspace, useful heat consumption and CHP potential in the UK commercial sector by size of premises.

Floorspace size (m 2) >400 > 1000 >5000 Total <1000 <5000

Total floorspace (km 2) 70 250 140 460 Total useful heat 40 150 120 320

demand (PJ/pa) Upper limit to CHP 2.8 10.0 5.6 18

capacity (GW(e)) Upper limit to CHP 5.6 21 17 45

output (TWh/pa)

Source: R.D. Evans, Environmental and Economic Implications of Small-Scale CHP, Energy and Environment Paper No 3, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990, Table 6.2.

potentially attractive since these sources would have inherently low maintenance costs and emissions levels, but considerable development is still needed to bring these technologies to commercial maturity. In any event, if an acceptable domestic CHP unit could be produced, the gains could be very large as the potential market would run into millions of units which would allow great scope for economies of scale.

However, a small part of the domestic sector could be exploited economically with CHP units of 40 kW(e) or more. This would be in large blocks of flats especially where communal heating systems

were already installed. In new high density residen- tial developments, small-scale CHP might also be considered to heat small groups of houses. The much reduced investment required for heat distribution would overcome some of the barriers to traditional CHP district heating schemes.

1R.D. Evans, Environmental and Economic Implications of Small-Scale CHP, Energy and Environment Paper No 3, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990. 21bid. 31bid. 4T. Jackson, C. Mitchell and J. Sweet, 'Study of leakage from the UK natural gas distribution system'. Energy Policy, November 1990. 51bid. 6Industrial Combined Heat and Power: The Potential for New Users, Energy Efficiency Series No. 7, HMSO, London, 1988. VGood Practice Guide No 1, Guidance Notes for the Implementa- tion of Small-Scale Packaged Combined Heat and Power, Energy Technology Support Unit, Harwell Laboratory, Didcot, Oxon, 1989; Good Practice Guide No 3, Introduction to Small-Scale Combined Heat and Power, Energy Technology Support Unit, Harwell Laboratory, Didcot, UK, 1990. 80p cit, Ref 1. 91bid. l°Op cit. Ref 1. l albid" 120p cit, Ref 6. 13Energy Use and Energy Efficiency In UK Commercial and Public Buildings up to the Year2000, Energy Efficiency Series No 6, HMSO, London, 1988.


Environmental and economic implications of small-scale CHP

Documents

Transcript of Environmental and economic implications of small-scale CHP