Learn about site assessments for the alternative energy potential of homesteads, including renewable energy areas of solar, wind, natural factors and advice on harvesting.
Whether you already own property or are simply in the
market for it, the alternative-energy potential of a
particular piece of turf ought to be as much a part of your
thoughts as are access, flooding, septic-field percolation,
the soil’s bearing capacity, and the view. Good solar or
wind potential can return thousands of dollars per year in
energy savings–or even in income from power sold to
your area utility–and can offer your family security
in a future of uncertain energy availability.
In the simplest sense, an energy site assessment to evaluate the alternative energy potential of homesteads is just a
matter of figuring out how much, when, and where. As you’ll
see over the next few pages, the first part of the question
is a straightforward matter of measurement, whether you
perform it yourself or consult tables prepared by someone
else. Determining at any given instant the amount of power
available isn’t difficult, when compared to figuring “when”
and “where.”
Predicting performance at some point in the future (“when”)
is much more difficult because solar energy–and wind
is essentially a solar-driven phenomenon–is by its
nature variable.
Likewise, “where” can introduce large uncertainties into
the estimation of energy potential, particularly with the
more ephemeral sources, solar and wind.
Thus a useful alternative-energy site assessment should be
composed about equally of’ careful measurement and an
understanding of the limitations on accuracy. Do your best,
but don’t overestimate the reliability of your best. And
when you use the numbers, err on the side that will keep
your lights lit and your house warm (or cool).
SOLAR ENERGY SITE ASSESSMENTS
In a solar site survey, “how much” is insolation (including
that which shines directly, that which is reflected, and
that which is scattered by the atmosphere-called diffuse);
“where” is local cloud cover and site shading; and “when”
consists of seasonal and annual variations in both how much
and where. Enthusiasm for solar over the last 15 years has
produced a great deal of information about this resource .
. . and those data can prove valuable to you.
More than is the case with wind, the way that solar energy
arrives has much to do with the success of attempts to
develop it. As you do a solar site assessment, then, keep
in mind that you’re looking for more than just whether it’s
worthwhile to develop your site or how much energy might be
available. The nature of the resource will suggest the best
hardware. For example, different sorts of collection
systems are best for capturing direct or diffuse radiation,
for coping with extensive shading from trees or buildings,
and for making do in areas with a high degree of
uncertainty in cloudiness, degree-days, and expected
sunshine. It’s beyond the scope of this mini-manual to tell
you how to design a system from the data you acquire, but
we will try to suggest some of the directions in which the
information might lead you.
CONSULTING THE DATA BASES
For a residential solar site survey, there’s no real need
to actually measure the amount of sunlight available in a
particular locale. Radiation available on a horizontal
surface is listed for 248 locations in the Insolation Data
Manual (see the “Access” information at the end of the solar section of the article), and these data are
reprinted in numerous solar energy references, such as The
Passive Solar Energy Book (hereafter referred to as TPSEB).
Because the numbers include the sum of direct and diffuse
radiation, corrected for time of year and atmospheric
blocking and scattering, they can be used directly for
predicting the performance of a solar collection system. A
sample listing for Indianapolis, Indiana, is provided in
Figure 1.
You may be disappointed to see that there’s much less solar
energy available in Indianapolis in December than in June
(about 78% less, in fact). Sad but true. First of all, the
sun is much lower in the sky in December about 27 degree altitude (its angle above the horizon) versus 74 degree at
solar noon (0 degree azimuth) on June 21-so its rays are
spread farther across a horizontal surface, reducing the
concentration. What’s more, the December sunlight is
scattered by its less direct (and therefore longer) passage
through the atmosphere.
Note, however, that these insolation figures are on a
horizontal surface. In December, vertical or angled
surfaces will be much closer to perpendicular to the
incoming rays, and the concentration per square foot will
be higher than that recorded in the charts. (We’ll get back
to this in a few paragraphs.) Likewise, ground that slopes
down to the south will receive a higher concentration than
that which is flat or sloping north. Though you can adjust
to a north-sloping lot with properly angled windows or
absorbers, a south-sloping one is still preferable, as
shadows from trees, hills, and other objects will be
shorter.
Another important factor enters into December’s
comparatively low insolation value in Indianapolis. The
Global KT Cloudiness Index, as listed in Figure 1, is the
fraction of the solar energy available above the atmosphere
that’s reaching the ground: 0.335 for Indianapolis in
December. Though the total radiation figure includes the
effect of cloudiness, this number tells us more: that 66.5%
of the radiation arriving from the sun at the outer edge of
our atmosphere is being intercepted in the air by clouds,
haze, etc., before it hits the ground.
When you compare this figure to those of other months and
other locations, it’s obvious that it’s pretty cloudy in
Indianapolis in December. And because of the clouds, much
of the solar energy that arrives will be diffuse, rather
than direct. This can affect collector orientation (since
diffuse radiation can, in effect, be considered to come
straight down) and collector choice (since high-temperature
absorbers make little use of diffuse radiation).
The National Climatic Data Center’s Comparative Climatic
Data for the United States offers more clues about average
weather conditions in Indianapolis: Figure 2 tells us that
39% of the days are sunny in December, while June has 66%
possible sunshine further confirming that clouds obstruct
much of the solar energy in Indianapolis in December.
Thumbing on through Comparative Climatic Data, you’ll find
that, on the average, December in Indianapolis offers 5
clear days, 6 partly cloudy days, and 20 cloudy ones. These
numbers offer clues to the size of storage needed for a
solar collector to get through cloudy spells in
Indianapolis. (Unfortunately, these data don’t tell us
whether it’s clear for 1 day, partly cloudy for 1 day,
cloudy for 4 days, and so on through 5 cycles . . . or
clear for 5 days, partly cloudy for 6, and then overcast
for 20 consecutive gloomy, chilly days. Collector and
storage sizing would be very different for these extreme
examples.)
Another section in Comparative Climatic Data gives us
snowfall averages-potentially useful figures because the
reflectivity (or albedo) of snow can dramatically increase
the amount of radiation striking a collector. Snow often
increases ground reflectance by a factor of four. A
vertical collector can capture a great deal of radiation
bouncing off the snow, but one that’s angled upward, toward
the sun, will reflect (rather than absorb) most of this
indirect radiation.
Unfortunately, Comparative Climatic Data doesn’t tell us
the actual number of days that the ground is likely to be
snow-covered. But by taking the 4.9-inch snowfall figure
for December and fudging against other tables that give the
mean number of days with a minimum temperature of less than
32 degrees Fahrenheit (25), the normal daily maximum temperature
(39.2 degrees Fahrenheit), the normal daily minimum temperature
(23.7 degrees Fahrenheit), and the normal daily mean temperature
(31.5 degrees Fahrenheit), we can see that there’s a fair likelihood
that the month’s total gain would be increased by
reflectance off snow.
On sunny days with snow cover, gain would be enhanced by
about 40%; there would be a minor increase on cloudy days.
(See TPSEB, professional edition, for more exact figures.)
From a design standpoint, the presence or absence of snow
cover adds variability to the “when” part of an assessment.
Unless you live in an area where you can depend on snow
cover, this factor makes it more difficult to pick the
right size collection and storage system.
EXTRAPOLATING TO YOUR SITE
If you happened to be considering buying a piece of
property a mile or so west of the Indianapolis Airport,
where the measurements in the figures were made, the
government data would be dandy to use in making
alternative energy judgments. The area is fairly flat, so
you’d be unlikely to be caught in a pocket where fog, haze,
or smog would obscure the sun much more than at the
airport. Likewise, you wouldn’t be near a body of water
that might encourage morning fogs.
If, however, you look for land over near Terre Haute, 70
miles away, cautious extrapolation becomes important.
Indianapolis is still the closest measuring station, but it
would be worth looking at the data for Evansville, Indiana,
and Springfield, Illinois, to see if there are big
differences. A large variation (say, more than 10% in total
radiation) for the three sites should make you cautious
about assuming that Terre Haute’s climate is the same as
that of Indianapolis. (It’s possible, even likely, that the
Indianapolis Airport has significant periods of smog or
haze.)
Next, you need to look at the regional and local features
of the Terre Haute property. There’s very little altitude
difference between the two areas, so there aren’t likely to
be big differences in insolation based on reduced
atmospheric density. (Likewise, the altitude shouldn’t make
it colder.) What about the site’s topography? Is it in a
valley that might trap fog, smoke from woodstoves, etc.? Is
there a significant stream nearby to add moisture to the
January chill? At the very least, frequent morning fogs
might lead you to orient your collector a little west of
south to face the more abundant afternoon sun. Here you’ll
have to depend on observation, estimation, and
instinct-bearing in mind that most insolation and weather
data stations are at airports, where the terrain is usually
flat and unobstructed.
SHADING
The heart of the work you’ll do on your feet when
performing a site assessment is to determine the pattern of
shadows cast by anything that might get between the sun and
your collector. Owner surveys and performance monitoring
tell us that the most frequently encountered problem with
solar energy systems is shading. The importance of clear
access to the sun’s rays can’t be overestimated. Even the
popular notion that deciduous trees are OK for winter
performance and beneficial for summer shading-because they
shed their leaves-is seriously flawed: The average bare
hardwood knocks out 40% of the incoming rays, and the
leafless branches of some species may intercept 45% or more
(see Figure 3). Transmission levels are lowest, of course,
when the trunk of a large tree is in the way. but even the
smaller peripheral branches can block from 20% to 40% of
the sunlight.
Even a preliminary solar site survey should include a rough
determination of the shading patterns on a piece of
property, and shading should be thoroughly diagramed before
you build anything. Here again, though, the sort of solar
collection scheme you’re planning will determine the
information you need.
Water-heating systems must have unrestricted access to the
sun 12 months of the year . . . while a passive heating
system would, ideally, be unobstructed in the coldest
months, partially shaded in the spring and fall to limit
gain, and fully shaded in the summer to prevent
overheating. Likewise, though you’d be interested only in
the southern exposure for a photovoltaic panel, you’d want
to look at potential summertime shading on the east and
west walls of a planned house. In the summer, a great deal
of the cooling load can come from. gain through east and
west windows that intercept the rays of the low morning and
(in particular) afternoon sun (see Figure 4).
What you must do is draw a profile of the landscape against
the horizon from exactly where you plan to locate the
collector, and compare this diagram to sun-angle charts such
as those available in TPSEB-to see if the site will be
shaded. In the case of large systems, this may mean making
repeated measurements at the various corners of the
collector. When applied to roof-mounted collectors or high
windows that will be installed on an as-yet-unbuilt home,
this can be cumbersome (if you try to position yourself at
the correct height) or complicated (if you try to project
shadow patterns to places you can’t reach). Unfortunately,
in this case there are no simple shortcuts or substitutes
for diligent work.
A very accurate profile of the southern horizon can be
developed using a surveyor’s transit. A compass and an
Abney level will allow you to prepare an acceptable shading
profile. The basic procedure is to plot the altitude angle
(height above the horizon in degrees, when horizontal is
0 degree and vertical is 90 degree) of the skyline at compass
point increments of 5 degree between 60 degrees and 360 degrees (120 degrees east and west azimuth of south). To use the
skyline profile you develop, you pick a sun-angle chart for
your latitude-TPSEB offers them in 4 degree increments
between 28 degree and 56 degree north latitude -and compare the
profile to the chart. The procedures are clearly described
in TPSEB, and an example is shown in Figure 5.
It’s less apparent, however, what one does to determine how
shading affects the actual total solar gain. To figure this
out, you have to first find the percentage of shading for
each hour by counting altitude boxes. Then each hour’s
shading percentage–from half hour to the next half
hour–must be weighted by the amount of maximum gain
available during that hour. If 50% of the 9:00 A.M. hour
(8:30 to 9:30 A.M.) is shaded, the total loss of gain will
be much lower than if 50% of the 12:00 noon hour (11:30
A.M. to 12:30 P.M.) is shaded.
From the example shown in Figure 5, you can see that during
November, December, and January the collector will be
unshaded; during February and October, however, it’ll be
shaded from 7:30 A.M. until 8:30 A.M.; March and September
shading stretches from 7:30 until 9:00 A.M.; and April and
August shading is from 7:30 until 8:00 A.M. This isn’t a
great deal of loss: about 5% of total direct radiation on a
horizontal surface in February and October, 10% in March
and September, and 3% in April and August.
As you can see, a guess as to the extent of these morning
losses probably wouldn’t have been far enough off to cause
problems. Nonetheless, you can’t make an accurate estimate
of shading losses without knowing the percentage of daily
gain that falls in each hourly period. You can calculate
percentages from hourly radiation charts offered in books
such as TPSEB, or you can send an SASE to MOTHER EARTH NEWS, attention
Readers’ Service, Hendersonville, NC,
and we’ll send you a chart covering latitudes 28 through
52, in 4 degree increments. An example for Indianapolis is
shown in Figure 6.
A reasonable alternative to developing your own shading
charts is to purchase a site survey tool. There are several
excellent professional models on the market, but the
$80-plus price tags probably aren’t justified for
preliminary site assessment. On the other hand, the Solar
Card is, at $12.95, a worthwhile tool that simplifies
shading assessment and offers a grid for determining
shading percentage (see MOTHER EARTH NEWS NO. 77).
PUTTING IT ALL TOGETHER
Once you have insolation figures for your locale from
reference books, and have adjusted them for any site
peculiarities and the percentage of shading, you can
determine the overall energy input. From this, you can
estimate the performance of various types of equipment
based on manufacturers’ claims or, in the case of passive
systems, calculated gains.
Let’s say that you’re thinking about installing
photovoltaic panels on a house near Indianapolis. You’ve
determined that the site isn’t significantly different from
the Indianapolis weather station. But it will be shaded by
a grand old white oak that has an altitude of 30 degree and
is located between 55 and 85 degree east azimuth. You don’t
want to cut the tree down, and there’s no better location
available for the panels. The losses are 5% in February and
October, 10% in March and September, and 3% in April and
August. The actual horizontal radiation, with shading
amounts subtracted, is shown in Figure 7.
You may have noticed that up until now we’ve always spoken
of the available solar energy as that which falls on a
horizontal surface. Naturally, you probably don’t intend to
mount your PV panels flat (and you certainly won’t have
horizontal windows in your passive solar home). Use the
annual graph to find what annual average fraction of the
radiation on a horizontal surface will be available on an
angled collector (or the monthly graph for vertical
windows). Just multiply the horizontal radiation (corrected
for shading) by the number for latitude and collector
angle. Indianapolis latitude is about 40N, and the optimum
angle from Figure 9 is 40 degrees. (Note: An annual fraction of
energy captured by an angled collector should be sufficient
for most people, but if you need monthly fractions for
surfaces other than vertical or horizontal, send an SASE to
MOTHER, attention Readers’ Service, at the address above.
Please include your latitude.)
When you multiply the aggregate annual horizontal
radiation–the daily factor for each month times 30,
summing the months–by the angle factor, the annual
available energy for Indianapolis comes to about 515,000
Btu per square foot per year. (Note that you can’t simply
multiply the angle factor by a given month to determine
performance for that period. The factor is an average of
the yearly optimal values, weighted for maximum system
performance.)
This number–arrived at after all this research,
calculation, and fudging–is a pretty good
approximation of the amount of solar energy available to a
south-facing solar collection system at a particular site.
Orientations other than true south result in
different total amounts of gain and different percentages
of the total being allocated to various times of
day–a complicating factor beyond our scope here.)
Of course, what use you put the energy to will also have a
great effect on the useful output. The PV system we’ve
talked about will have an efficiency of between 5% and 7%
(And such systems are usually rated at an energy input of
317 Btu per square foot.) So for every 1,000 Btu per hour
in, you’ll get back between 15 and 20 watts per hour (at
3.41 Btu per watt). An efficient solar water heater, on the
other hand, might capture somewhat more than half of the
available Btu.
Now it’s up to you to put the numbers to work: The data you
develop in a site assessment are the takeoff points for
system design and performance estimation.
Solar Access
Climatic Atlas of the United States , U.S.
Department of Commerce, 1983. Available from the National
Climatic Data Center, Asheville, NC, $15.00.
Comparative Climatic Data for the United
States , National Climatic Data Center, Asheville,
NC, $4.00.
Insolation Data Manual, by Knapp,
Stoffel, and Whitaker, Solar Energy Research Institute,
1980. Available from U.S. Government Printing Office,
Washington, DC. Stock No. 061-000-00489-1, $8.50.
More Other Homes and Garbage, by Jim
Leckie, Gil Masters, Harry Whitehouse, and Lily Young,
Sierra Club Books, 1981, $14.95 paperback.
THE MOTHER EARTH NEWS Plans, Hendersonville,
NC. Wind-Driven 2,000 Watt Electrical Generator
(Stock No. 84033), $15.00. Dollar-a-Watt Windplant (Stock
No. 84039), $10.00. Crossflow Turbine Plans (Stock No.
84019), $15.00. Please add $1.98 shipping and handling
with each order.
The Passive Solar Energy Book, by Edward
Mazria, National Association of Home Builders, 1979,
$24.95.
Wind Energy Resource Information System,
52 tables for 975 sites from the National Climatic Data
Center, address above. You’re interested in WERIS Table
11(Percent, Frequency of Occurrence, Wind Speed vs. Month),
Table 4 (Average Windpower by Hour and Month), and Table 14
(Percent, Frequency of Occurrence, Wind Direction vs.
Windpower). Table 10 (Significant Weather Parameters and
Events by Month) will provide useful information about
thunderstorms, tornadoes, hail, freezing rain, etc.
Photocopying of these tables from microfiche costs about
200 per page.
WIND ENERGY SITE ASSESSMENTS
Though the national wind data base consists of a variety of
information from 975 stations in the U.S., Puerto Rico, and
the Pacific islands, the fickle nature of wind makes
interpreting these numbers more difficult than analyzing
similar data for solar energy. To correctly extrapolate
from a weather station to a particular site without
actually measuring the wind requires quite a bit of
guesswork-even after the most meticulous site survey. For
that reason, many times it’s wise to correlate your own
measurements of wind speed, duration, and direction with
the published data before investing much money in
equipment.
DECISION TIME
As you’ll see in a few paragraphs, average wind speeds can
be very misleading if used to estimate the energy available
at a wind power site. They will, however, allow you to
decide whether it’s worth devoting much time to studying
the wind energy possibilities on your land.
An airport near you should be able to provide you with the
average annual wind speed at that location. If not, the
National Climatic Data Center’s publication Comparative
Climatic Data (see “Access” section at the end of the solar section of this article) lists average and maximum wind
speeds for 285 locations.
Should your site be close to the measuring station,
comparable in terrain, and devoid of obstructions–trees,
buildings, etc.–within a distance equal to 15 times the
object’s height, you can use these rules of thumb: When
annual average wind speed equals 8 mph or less, look for
another source of power. When annual average wind speed
equals 12 mph or more, plan to develop wind power. When
annual average wind speed is between 8 and 12 mph, take
measurements at the site, correlate them, and proceed
cautiously.
A BIT OF THEORY
What may seem like small differences in wind speed numbers
can make a big difference in the amount of energy
available-a situation you can understand only by looking at
what makes up the formula for wind energy:
Energy = 1/2 X air density X area of machine X wind
velocity 2
Air density affects the number of molecules in a given
volume of air striking a turbine’s blades; area considers
the rotor’s size; and the cube of the wind velocity
(velocity X velocity X velocity) takes the standard
momentum equation (v2) and multiplies by velocity one more
time to allow for the volume of air passing the turbine’s
blade,
Look at what happens if the wind speed changes only
slightly: Let’s say that your closest measuring station has
an average annual velocity of 11 mph, but the average
annual velocity at your site turns out to be only 9 mph.
The cube of 11 is 1,331 and the cube of 9 is 729. Though 9
mph is only about 18% less wind speed, the energy available
is 45T( less. To look at it another way, a 13 mph average
wind has more than twice as much energy as does a 10 mph
wind! Do you begin to see why little errors can make such a
difference?
The cubic influence of wind speed on power has another
important effect on the amount of energy available: Average
wind speeds may not give an accurate picture of available
energy. Consider the example of two sites that both have a
15 mph average wind speed. On the face of it, they would
seem to have the same wind energy potential. But do they
What if one site has constant 15 mph winds, and the other
has 10 mph winds half the time and 20 mph winds half the
time? Energy at the first site is proportional to the cube
of 15 (3,375); but at the second, it’s proportional to the
cube of 10 (1,000) plus the cube of 20 (8,000), or 9,000.
There’s more than twice as much energy in the wind at the
site with half 10 mph and half 20 speeds.
BEYOND AVERAGE.
To get an even clearer picture of wind energy
possibilities, you need to use NCDC’s more detailed data.
And before we go any further, you might as well get used to
the idea of dealing with the metric system; it’s the
language of wind energy. From this point on in the article,
we’re going metric, too. If you want to switch to English
units, find a calculator that will make the conversions or
multiply by the following conversion factors:
As we already mentioned, NCDC has wind measurements from
975 stations. These data are presented through the Wind
Energy Resource Information System (WERIS), and, as you’ll
soon see, they’re extensive and detailed. Start by writing
to NCDC and requesting a list
of WERIS stations in order to determine which one is
nearest you. There are 19 different tables–accounting
for 52 pages–available for each station, and there’s no
point in ordering more photocopying than you need. We’ll
show you how to use the more basic of the tables in the
following paragraphs.
Because there’s no need at this point to consider the size
or performance characteristics of a particular wind
machine, the figure we’ll be looking for is power
density the amount of power available per unit of wind
machine rotor area in watts per square meter (w/m2).
FREQUENCY OF OCCURRENCE
NCDC WERIS Table 11, “Percent, Frequency of Occurrence,
Wind Speed vs. Month,” divides up wind velocities into
convenient bins-showing the percent of the time that the
wind speed falls into that range and time period. By
calculating the power of each bin and summing those
numbers, you can get an accurate picture of the power
available. An annual summation gives a picture of total
power density, and a monthly analysis tells you whether the
wind is likely to be sufficient to satisfy your energy
needs in a given month. An example (back to Indianapolis)
is shown in Figure 1.
You can’t simply use the speed itself to determine the
power from each speed class once again, because of the cube
effect. The power midpoint of each speed class is an
average of the powers available at the lowest and highest
speed in the class. For example, the speed range of the 3
m/sec class is 2.5 to 3.5 m/sec, and the power
midpoint–the cube root of half of the sum of the
cubes–is 3.08 m/sec.
Lucky for you, the NCDC computers have already done all
that ciphering for you. Figure 2 includes some of those
figures, taken from WERIS Table 4.
Without the labor of calculation, we find that Indianapolis
has an annual average power density of 76 w/m2, with highs
and lows of 117 w/m2 and 32 w/m2 in January and August,
respectively. This power was determined at an anemometer
height of 6.1 meters and will vary at other heights.
ALTITUDE AND TEMPERATURE
Way back in the section on theory, the formula for power
contained an element for air density. Since then, we’ve
ignored it . . .but it’s finally time to correct that
omission. Because air is less dense at higher altitudes and
higher temperatures and more dense at lower altitudes and
cooler temperatures, power density needs to be corrected
for any differences in altitude and temperature that exist
between the site and the measuring station.
For every 1,000 feet above sea level, density drops by 3%.
NCDC figures have taken the Indianapolis air density into
account in the annual power density figure of 76 w/m 2 . If
you’re at a significantly higher or lower altitude than the
measuring station, however, the power density has to be
corrected for the difference. Likewise, significant
temperature differences–which might be experienced at
different altitudes–should be factored in by
multiplying power density by the correction numbers in Figure
3.
WIND DIRECTION
NCDC’s Table 14 is another invaluable aid to wind power
site assessment. It tabulates the direction of wind by
power density, allowing you to determine what might be the
most crucial upwind direction for a machine. Figure 4 shows
the distribution of power directions for Indianapolis.
Though wind power at the Indianapolis station is fairly
well distributed by direction, you can see that the
south-to–west quadrant is particularly important.
Assuming that your site has similar prevailing winds, it’s
important to make sure that there are no obstructions close
to the windplant’s location toward the south and west.
EXTRAPOLATING TO YOUR SITE
Relating wind data from an NCDC station to a site you’re
studying is far more difficult (and chancy) than relating
weather or insolation information. This is because wind
velocity, direction, and turbulence are profoundly affected
by local topography, vegetation, and buildings.
When you choose a station from which to use data, read the
fine print to see what kind of location was used. Many
stations are located at airports, which are typically flat,
low-lying, and often unobstructed by nearby trees or
buildings; others are in municipal areas with lots of
buildings. Furthermore, locations for airports and cities
are generally selected for a lack of wind, so data
collection stations are seldom optimal wind power sites.
Compare the area around the wind station to that around the
site you’re studying. There are three basic factors to
consider: roughness, topography, and barriers.
Surface roughness influences the way and the degree to
which upper-level winds are reduced by friction with the
ground. Different sorts of roughness characteristics and
their effects on wind speed at various heights are shown in
Figure 5. Bear in mind that NCDC data is taken at different
heights at different locations and that the velocity (and
therefore power) must be corrected to take into account the
height at which the wind was measured. The formula for
calculating increase in power for additional height over
the measuring point is this:
Power (original) = (Height (original)) 3a
Power (new) (Height (new))
The symbol a is the roughness factor, and it serves as the
exponent in the equation. For our example case of
Indianapolis, we’ve standardized the figures to the listed
anemometer height of 6.1 meters. If your .station is at a
different height, you’ll have to work out the power
increase for yourself, using the equation above and the
roughness factors listed in Figure 5. (In the case of tall
crops, woods, or buildings, be sure to consider the surface
to be at the level of the top of the obstacles.) Just
multiply the original power density by the appropriate
number for the height above the effective surface to arrive
at the corrected w/m2.
If the roughness of the ground upwind from a site is
consistent, it’s fairly simple to determine an acceptable
tower height based on the information in Figure 5. That’s
seldom the case, though, so you’ll have to make educated
guesses about the overall effect of the combinations of
roughness at the site. Keep in mind, too, that when the
roughness changes–at the edge of a woods, for
example-there is a transition zone of intermediate values.
As a very rough guide, this might extend for a distance
equal to twice the height of the change upwind and 10 times
the height downwind.
Topography can have a dramatic effect on both the pattern
and the speed of wind. First, simply positioning a site on
a hill is roughly equivalent to having a taller tower; the
terrain lifts the machine up into the prevailing winds. The
ideal hill site is on the top one-third of the upwind side
of a ridge perpendicular to the wind, as shown in Figure 6.
In this location, not only will the machine have the higher
winds available at altitude, but it will be in an area
where wind is accelerating to ride up over the ridge. This
effect can increase wind speed as much as 100% on the top
one-third of the upwind face of a ridge with an ideal slope
of around 30 degrees 70. By the same token, sites on the lower
two-thirds of the leading face or on the trailing face of a
ridge perpendicular to the wind may have significantly less
than the predicted wind.
Disruptive turbulence will be found on the trailing face of
perpendicular ridges. And at the top of ridges with flat
peaks or at the top of a cliff, there’s likely to be an
area of wind shear: a zone where the normal increase in
wind speed with height above the surface is disrupted,
resulting in a sharp difference in speed over a short
vertical distance. Wind shear can place the upper and lower
parts of a wind machine’s rotor in drastically different
wind speeds, which would exert tremendous stress on the
equipment.
Terrain can also redirect the wind’s prevailing pattern.
For example, in the case of the ridge we were just
discussing, wind may funnel around its ends or through a
pass. On the upper one-third, such sites may be excellent
for wind power. Similarly, valleys may direct wind down
from mountains into basins in the cool morning and back up
in the heat of the afternoon. At the lower end of these
valleys where there’s not enough constriction to cause heavy
turbulence–a machine on a sufficiently high tower may
get a boost from the daily cycles.
Unfortunately, topography seldom cooperates by fitting
textbook descriptions. Unless the situation is quite
obvious, in areas of significant hilliness you’d be wise to
make direction and power measurements of your own and
correlate them with the data from the nearest measuring
station.
Barriers upwind from a wind machine produce wakes that are
more turbulent and have lower speed than the prevailing
wind. As a rule of thumb, avoid any site where a barrier of
a given height will be twice that height laterally downwind
or 15 times that height upwind, as we show you in Figure 7.
The width of a barrier also plays a role in wind
disruption. Amazing as it may seem, a single tree can
reduce wind speed by 10% at a distance downwind as great as
30 times the tree’s width. In other words, a 25-foot-wide
tree could reduce wind speed by 10% at a point 750 feet
downwind! To avoid this sort of problem, plan on using a
tower that will put the wind machine 25 feet above any
barrier that’s within 500 feet in the direction from which
winds of significant power come.
These same guidelines apply for ridges or tree lines (such
as shelterbelts) upwind from a wind power site. And in the
case of a shelterbelt, the density of the growth has an odd
effect: The most disruptive tree lines directly upwind are
comparatively porous–that is, they have low
vegetation density; medium porosity leads to the greatest
lateral disturbance; and very dense growth in a shelterbelt
produces the fewest problems.
As you can see, analyzing the similarities between a wind
power site and a nearby measuring station requires, at
best, quite a bit of subjective judgment. Before making any
major investment, you should put any doubts to rest by
measuring the wind.
MEASUREMENT
If you have concerns about roughness, topography, or
barriers at a site, you should, as a minimum, determine the
power-wind directions) so you can estimate how serious the
effects of the site’s characteristics will be. A simple
wind sock (or weather vane) on a pole and a notebook may
give you enough information, as long as you’re
conscientious about making observations and recording them.
A wind-direction and–speed meter is more convenient
and will give you a better idea of where the strong winds
are coming from. (TJ Byers described how to build such a
device for less than $20 in MOTHER EARTH NEWS NO. 68.)
However, you’ll need more sophisticated equipment to
correlate power density at a measuring station with that at
your site. A wind odometer is an improvement over an
anemometer that merely measures instantaneous speed. (More
Other Homes and Garbage describes how to build a wind
odometer from a pocket calculator.) It will provide you
with an average wind speed, which can be corrected to a
power density range by using the factors shown in Figure 8.
The only way to get more accurate numbers is to use a
recording anemometer, a device that differs from a standard
anemometer or wind odometer in that it tallies wind speeds
(and directions, if you like) and drops the frequencies of
occurrence of various velocities into the appropriate
registers-producing a total for each speed group over the
period. Unfortunately, recording anemometers are
prohibitively expensive for our purposes.
You’d actually be better off buying or building a small
wind machine and equipping it with a watt-hour meter.
(MOTHER offers plans for several different inexpensive wind
machines–see page 100 for information.) Such an
approach would give you numbers equal to the power density
times the efficiency of the machine–nearly 10% for the
Blue Max and 20% for the 2000–watt windplant, two
designs MOM offers–and would leave you with a useful
piece of hardware once the measuring was done.
If you go on to purchase a sophisticated wind power system,
you’ll want to know how it will respond to the wind
characteristics at your site. The machine’s efficiency curve
will influence what percentage of different speeds it can
make use of, and its cut–in and cutout speeds will
determine what portion of the available winds it can
capture. By judiciously applying the manufacturer’s or
dealer’s specifications to your site assessment, you should
obtain a performance estimate that will leave you with few
surprises.
