Power    and

Building-integral  On-site

Solar  and  Wind Power   Systems

This brief tutorial explains how to estimate power and energy yields, of building-integral on-site solar and wind power conversion, storage, and management installations; analyzes past and new systems; and compares their cost.

No suitable photos of existing building-integral installations, combining both solar and wind power, were found.  Perhaps none exist, because conventional power storage devices needed for power on demand do not afford acceptable system options.  For now, this illustration provides the best available means to visualize this concept.

Briefly stated, PV (photovoltaic) and thermal solar panels can cover  much of a building's exterior surface; including walls, windows, roofing, and skylights.  PV panels (shown as light blue surfaces) can be used instead of conventional building enclosure materials, for competitive total building and power equipment life-cycle cost.  Proximity to other parts of the building's electric system is not critical.

However, solar heating panels (shown as dark surfaces) need to be nearer to the other components of the building heating system. They can be interspersed with PV panels as shown here.  Heating system insulation, pipes, and ducts can thus be minimal.

Building-integral wind turbines can be mounted so the building channels wind to them, to increase effective windspeed at the turbine by a factor of at least 2 and possibly 4, and the building provides a high and sheltered mounting (with no need for towers and the uneven windspeed patterns they cause to spinning blades, which causes noise and blade fatigue).  It also affords practical protective grills, to safeguard people from possible blade disintegration, keep birds away from the spinning blades, and protect the turbine from weather and sun damage.

Doubling wind speed at the wind turbines can increase their output power by 8X.  Although that increase appears easy  to achieve, with no compromises in building cost or function, a building's aerodynamic properties at its intended site would need to be analyzed, in addition to wind data available for the site. Average wind speed records are available for many locations.

Whatever combination of solar, wind, and utility power is available, the entire power system would be greatly enhanced by a reliable, zero maintenance, ultra-long life, lower life-cycle cost  power storage and management system.

Stand-alone solar and wind powered buildings might include amenities that don't need power constantly, such as pumps for pools and pool filters, exterior lighting, etc.  These discretionary loads may be progressively enabled when stored power exceeds 90% of maximum capacity and is increasing.

To appreciate advantages of  building-integral power systems and UPS (Uninterruptible Power Supply) over others, consider these specific comparisons:

PV that's not building-integral, windmills on towers, fuel-burning generators, conventional lead-acid batteries and flywheels -- practically all other solar, wind, and UPS -- use up costly real estate.  Building-integral UPS, including the proposed underground flywheel battery, is part of the structure it serves, and does not need additional real estate.

Conventional UPS needs a sheltered off-limits enclosed area. Building-integral PV panels, conversely,  provide a sheltered enclosure, and it's not off-limits.

Home Power Magazine  covers on-site power technologies and practices. Earthsystems  maintains a  library of  environmental websites.   Special Housing  maintains a site dedicated to special needs of  building occupants, energy efficiency, renewable energy, and the environment.

Solar Power

Steam-cycle solarthermal generators have been operating for over a century; but are not commonly used for on-site electric power, perhaps due to their special plumbing needs.

Building-integral PV panels are rapidly becoming dominant solar power conversion means.  The Solarex building (left) illustrates contemporary PV panel use.

PV panels between glass plates can be produced in several translucent shades.  So they can substitute for conventional tinted glass skylights and windows.

DC output voltage vs. current of a PV panel is illustrated below (left image).  For maximum yield from the panels, interface circuit conditions should be set to operate the panels at  P=max on this curve.  It results in a maximum volts-amp product.  Power is zero at end points P=0.

When sunlight on the panels is negligible (e.g., at night), the bottom end point of this curve shifts practically to the origin (0,0) and the top end point shifts a bit up.  At maximum sunlight, the bottom end point shifts right, and the top end point shifts down a bit.

Connecting the panel's DC output to an inverter, which converts power to sinusoidal output voltage and current as illustrated above (middle image), has become a popular but questionable option.  Some even recommend that the inverter's self-synchronizing output be plugged into any 60-Hz 120vrms AC wall outlet in the building served.  It adds another input power source.  When it exceeds loads, it can reverse utility power.  But let's take a closer look:

Apart from possible "hot" plug-in synchronization problems and "live load" hazard to utility line workers, PV yield would be only about 47% (assuming 95% inverter efficiency) of yield directly available from the panels.  This conclusion is evident from examining the panel's voltage vs. current curve, and the inverter's input current waveform (which, without huge holdup capacitors, resembles the PV Power curve).  Panel output power will cyclically vary between P=0 and P=max.   Area under the PV Power curve (shaded green) corresponds to yield, and is about half  the amount achieved by drawing constant current at P=max.

Some inverters turn off solar powered output when grid power is off.  So then you have no power when the grid is down.

We can do far better. We can double the power from the same PV panels, protect the grid from "live loads" and provide uninterruptible power.  Here is how:

Contrast the middle image's cyclic PV panel current, with the constant current at P=max, drawn by the 2-phase system illustrated by the right image.  Phase 1 draws current Imax sin2(377t).  And Phase 2 draws current Imax cos2(377t).  So total current sum is Imax and it is constant.

Industrial buildings usually have significant loads from 3-phase machines.  A 3-phase inverter to power them would likewise draw constant DC current, without significant ripple.

Accordingly, we propose:


Low internal heating, high reliability, and zero maintenance facilitate safe underground siting for a DC flywheel battery enclosure. Safety is further enhanced by electronics which initiate power-down if abnormal rotor assembly vibration is sensed.  However unlikely, flywheel disintegration would release uncontrolled energy.  The underground site -- not the vacuum enclosure --  is designed to absorb it, without damage to the building where it is housed or injury to people.  The power electronics and status monitor display is designed to be mounted on a wall where its status display will be conveniently visible.

A flywheel battery we are developing, and its applications, are described and illustrated in my  overview webpage.  It is shown in more detail, and analyzed (noting differences between it and those being developed by others) in my  comparison webpage.

Integration of the flywheel battery with utility, solar, and wind power, and on-site loads, is illustrated by the simplified schematic diagram shown below:

This diagram differs from a similar one in my  overview webpage, where grid power is connected to on-site loads by switchgear that switches loads to on-site power during grid outages.  That circuit does not prevent line spikes and drop, but is a few percent more efficient.

This circuit provides better on-site power quality from grid power (rejecting line spikes and preventing line drop); and can draw grid current at unity power factor and only a few percent distortion,  with a PFC (Power Factor Correction) as shown in the circuit.  The bridge rectifier transmits power from the utility to the on-site 170 vdc buss, but it blocks reverse power flow (from the dc buss to the grid).   Besides long-term power storage, the flywheel battery draws current from solar/wind sources and the central grid, and supplies on-site power on demand, so as to maintain constant, ripple-free 170 vdc at the dc power buss.  Along with the filter/holdup capacitor shown, the flywheel battery also helps accommodate possible current ripple from the wind-driven generator, need for ripple-free PV current, and substantial 120-Hz ripple to supply the 60-Hz sinusoidal power inverter shown.

In a grid-connected system, during grid power outages, the bridge rectifier shown essentially disconnects utility lines from building power, to protect utility line workers.  Note that building power is not interrupted by grid power outages.   A system like this accommodates usually lower cost off-peak grid power schedules, and circumvents need to negotiate utility buy-back of  power generated on-site.  Mainly, it enables zero-maintenance, safe, non-polluting, sustainable on-site power generation, and carefree building UPS.

PV Panel Output Power & Yield Computations

Radiant power from the sun, to most of  the populated earth surface, after atmospheric reflection and absorption, averages 644 watts per square meter.  It's available for over 5 hours per day on stationary PV panels, installed so they receive maximum sunlight.

So a building like the top image, with 1000 square meters of  PV panels, having 10% power conversion efficiency, can generate about 64 kw for 5 to 8 hours a day.  That's an electrical energy  yield well exceeding 10,000 kwh per month.  Most PV panels cost less than $4 per watt output, and are guaranteed for over 20 years.

Total PV panel cost, in this example, would be $256,000.

So their lifetime electric power output costs about
($256,000) / (10,000 kwh/mo)(12 mo/yr)(20 yr) = $0.10 per kilowatthour.

PV Panel  Cost Comparison to Utility Power

PV panel installation cost will surely be lower than building materials for which they can provide attractive and durable substitutes, once standardized mounting and connection hardware is available.  Since those building materials would not be needed, for simplicity, we have not included PV panel installation cost in this example.

A building using 10,000 kwh per month would use an average (10,000 kwh) / (720 h) = 14kw.

This system would need 5 networked RPM flywheel battery systems, each rated 15-kw 50-kwh.  Installed cost, after production ramp-up, with a 20-year warranty, cannot be quoted yet because none are in production.

The 5 x 50kwh energy storage this system can provide, could power this building for 250kwh / 14kw = 17 hours. That would enable power on demand and 24 hours per day use of power from this PV panel installation. It would also enable purchase of grid power during off-peak hours at lower price, and provide immunity from grid power blackouts and voltage sags.

This building-integral solar power installation can enable 10,000 kwh per month solar electricity; plus uninterruptible building power, immune from grid blackouts and voltage sags, and off-peak grid power purchase.

Grid power now costs about $0.10 per kwh, and its price is likely to continue rising.  Environmental damage, from burning the fuels needed to generate it, are not included in this price.

Comparisons for solar are even more favorable when buildings are remote, in areas not served by utilities, and especially if environmental costs are included.

Wind Turbine Details

Power that can be generated from wind is very sporadic and far less predictable than solar.  Nevertheless, probable power and yield for a given system can be computed from statistics. Consider an axial-flow wind turbine with variable-pitch blades (images below).

This type is proposed because it achieves 45% windmill efficiency, is feathered at high windspeeds, is self-starting, and lends itself  to  building-integral installation.

A "ring mount" with a down-wind vane to align heading to the wind, is shown at right.  It differs from usual tower mounting in that it does not have a broad column near the blades.  Such a column, like any object which disrupts uniform air flow, subjects blades to cyclic wind forces, which can cause blade and stem fatigue and early failure, plus annoying "swooshing" sounds while running.

This assembly is shielded from direct sunlight.  So fiber composite or wood blades, if used, would be subjected to minimal ultra-violet and other weather damage.

Darrius windmills, that resemble a vertical-axis egg-beater, and achieve 30-40% efficiency, are not proposed, because they are not self-starting and do not accommodate a wide range of  wind speeds.

Savonius windmills, often built from 2 oil-drum halves mounted on a vertical axis, have only 15% efficiency, and can't withstand high windspeed, are also not proposed.

Wind Power and Yield Calculations

Electric power from a windmill  =  (.0049)(windspeed)3(swept area)(efficiency)
where windspeed is in miles per hour
           swept area is square feet swept by windmill blades
           efficiency is %/100 for windmill, generator, and PWM regulator.

Since wind is so variable, average windspeed is used to predict duration at each speed, which follows the statistical distribution:
     Probable time at windspeed V   =   V /  Vavg2  *  exp(-.785 * V2 / Vavg2)
where total time is 720 hours/month
           Vavg is the average wind speed at the windmill.

Power (kw at V), probable time at windspeed (hours at V), and their product (kwh at V), are plotted (for Vavg= 10 mph) vs. windspeed V (mph) in the figure shown at right.

Blades begin to feather at windspeeds above 30 mph, causing generated power to remain practically constant at feathered windspeeds.

Yield (green area in this figure) is not appreciably lessened (red area) by blade feathering. However, compared to fixed blades (dashed blue curve), peak power that must be handled by the entire system is only about 20% as high.  Also, feathered blades with self-limited spin speed greatly reduce mechanical stress on all windmill parts (compared to fixed blades, with turbine shaft spinning at essentially one speed, or locked during low and high winds).

For a windmill that sweeps 300 square feet, with 10 mph average windspeed (5 mph average for locale, doubled by building aerodynamics), and overall efficiency of  40% for windmill, generator, and PWM regulator, we get:

Power generated at 10 mph = 588 watts from each windmill
Yield from wind power = 729 kwh/month from each windmill
Maximum generated power (limited to level at 30 mph windspeed) = 16 kw from each.

If the building aerodynamics can result in 20 mph average windspeed at the same windmill, we can get yields of about 5,000 kwh/month, from the same windmill, having the same 16 kw maximum.

On-site Wind Power Cost Estimates and Comparisons

Estimated cost of the 2 windmills proposed above, their generators, power electronics, and mounting, is about $30,000.  This windpower system would require 2 more 50-kwh 15-kw flywheel batteries, to provide power on demand and UPS.  So the total cost would be about $50,000 for wind powered generators and UPS.  This equipment would not need substantial maintenance over a 20-year lifetime.

Energy yield from wind power, over 20 years, would be (2)(5000 kwh/mo)(240 mo) = 2,400,000 kwh.

So wind power from this installation would cost about ($50,000) / (2400000 kwh) = $0.02 per kwh.

On a life-cycle cost basis, these estimates indicate that on-site wind power systems can provide power on demand plus UPS at far lower cost than utility service and conventional UPS equipment.

Again, as with the on-site solar power cost analysis, advantages of this "green power" option also include:

Combined On-site Solar and Wind Power Systems

Both sources provide clean renewable power. Combining them provides a higher probability that power will be generated at any given time (i.e., source power is spread over a longer time). This permits somewhat less on-site power storage need.

On-site solar and wind power systems will have profound environmental benefits. It will also benefit special housing projects.

Visit these website pages, to learn more about RPM's flywheel battery:

Overview              Flywheel  Battery Tutorial             Comparison: RPM & Others

     RPM's Resources                                    RPM Business Plan Abstract

RPM's flywheel battery & EVs? Yes, but not onboard!

Flywheel  Facts  and  Fallacies

Urban EV with Onboard Batteries, Charger, PV, Regenerative Motor, Pedals

Technology:  Public and Business Policy

Future cost-competitive clean green care-free sustainable electric power

RPM's Broad-speed-range Generator

If you have comments or suggestions, email me at

Edited   January 2014