Wind · Vertical axis
A vertical-axis rotor works the same whichever way the air arrives, puts its generator where a person can reach it, and runs quietly. Those are real advantages and they are why the architecture keeps coming back.
It also converts less of the wind it meets, and it is marketed hardest to precisely the households with the least wind to convert. Both halves of that belong on the same page.
Start with the distinction
Vertical-axis turbine describes a shaft orientation, not a working principle. Underneath that single label sit two families that extract energy in completely different ways and behave nothing alike. Comparing a vertical-axis machine against a horizontal one without saying which family you mean produces an argument nobody can win.
What they share is the shaft standing upright, which means wind from any bearing works equally well, no yaw mechanism is needed, and the generator can sit at the bottom instead of at the top.
Drag-driven. Scooped surfaces that catch the wind on one side and present less resistance on the return. Starts easily, turns slowly, converts modestly.
Lift-driven. Vertical airfoil blades that generate lift through most of each rotation. Converts better, starts badly, and lives a harder mechanical life.
The drag machine
Picture a barrel cut down its length and the two halves offset to form an S. Wind pushes harder on the concave half than on the convex one, and the difference turns the shaft. Anyone who has seen a cup anemometer has seen the principle.
The limitation is built into that description. The returning half is always moving into the wind, always producing a force opposing rotation. A Savonius rotor spends its entire working life fighting a portion of itself, which is why its conversion efficiency sits well below what a lifting blade achieves, and why its tips never travel much faster than the wind itself.
Starts turning in very light wind without assistance. Produces useful torque at low speed. Tolerates gusts and direction changes without complaint. Can be built from ordinary materials by someone with basic fabrication skill, which is why it appears in so many workshop projects.
Makes electricity. Low rotational speed is the wrong input for a generator, so it needs either a large diameter, a step-up, or a specially wound low-speed alternator. Its conversion efficiency means it needs substantially more swept area than a lifting rotor for the same output.
Mechanical work rather than electrical generation. Driving a pump, an aerator or a ventilator, where torque at low speed is what the job wants and nobody is counting kilowatt-hours. In that role it is a reasonable and durable machine, and dismissing it entirely misses the point of it.
The lift machine
A Darrieus rotor uses airfoil blades set vertically, either curved into the shape often described as an eggbeater or held straight in the variant usually called an H-rotor. As the rotor turns, each blade meets the combination of the real wind and the wind created by its own motion, and through most of the circle that combination produces lift pulling the blade forward.
Because it works by lift rather than drag, it reaches a meaningfully higher power coefficient than a Savonius and can run at a tip speed above the wind speed. It is the serious vertical-axis machine, and every claim about vertical-axis turbines competing with horizontal ones is really a claim about this family.
A stationary Darrieus blade sits at an angle that generates almost no useful lift. The machine produces very little starting torque and frequently needs help to begin turning, either from the generator driven briefly as a motor or from a small Savonius rotor fitted to the same shaft to get it moving. A machine that needs power to start making power is a design consideration, particularly off grid.
This is the important one. Each blade travels upwind for half the circle and downwind for the other half, so the force on it swings through a full cycle on every single revolution. A horizontal rotor in steady wind holds a broadly constant load. A vertical one never does.
On the downwind half of the circle, each blade flies through air already disturbed by the blades ahead of it. That costs output, and it adds an irregular component to a load pattern that is already cyclic.
Metal fails from repeated loading far below the level that would break it once. A rotor turning at 200 revolutions per minute applies its load cycle twelve thousand times an hour, and a machine expected to last twenty years must survive an astronomical number of them.
That is why reports of premature bearing and coupling failure follow this architecture, and why a vertical-axis machine should be judged on the substance of its bearings, shaft and mounting rather than on its silhouette.
Side by side
Vertical-axis machines are not a failed idea. They win several rows outright. They lose the row that decides how much energy arrives.
| Characteristic | Horizontal axis | Vertical axis |
|---|---|---|
| Wind direction | Must be pointed, by vane or yaw drive | Accepts any direction, no mechanism |
| Driving force | Lift | Drag in a Savonius, lift in a Darrieus |
| Energy conversion | Higher, in clean air | Lower, so more swept area for the same output |
| Behaviour in shifting wind | Loses output and yaws repeatedly | Unbothered by direction change |
| Starting | Self-starting | Savonius self-starts, Darrieus often does not |
| Noise | Higher, rising steeply with tip speed | Lower, and lower in pitch |
| Generator location | At the top of the tower | At the base, reachable from the ground |
| Fatigue pattern | Steadier in clean air, yaw loads in turbulence | Load reversal on every rotation, by design |
| Certified models available | The great majority of the certified market | Check the certification directory for the specific model |
The last row is the practical one. Certification against the national small wind standard is public and searchable, and whether a particular machine holds it matters more than which architecture it belongs to.
The claim examined
The case made for vertical-axis machines in residential settings goes like this. Low altitude air near buildings and trees is turbulent and shifts direction constantly. A horizontal machine handles that badly, yawing continually and never settling. A vertical machine does not care which way the wind comes from, so it handles that air better.
Every sentence of that is accurate. The conclusion drawn from it is not.
The same ground clutter that makes low-altitude air turbulent is what makes it slow. Drag against buildings and trees removes energy from the airstream, and because available power falls with the cube of speed, the air down there holds very little to begin with. A machine that copes gracefully with a poor resource still only has a poor resource to work with.
A vertical-axis machine converts a smaller share of what passes through it, and it is being recommended for a location where much less passes through it. Those two disadvantages compound rather than cancel.
The correct response to a turbulent site is not a different rotor. It is a taller tower, or a different source entirely. Raising the same machine from 30 feet to 100 feet on the worked example in this section increased annual output by about 64 percent, and no change of architecture at low level comes close to that.
This matters because the claim is usually made to households in exactly the situation where wind is least likely to repay them: a suburban or wooded lot, a short mounting height, and often a roof. The architecture is not the problem there. The location is.
Fair conclusions
There are real situations where the architecture wins, and they have nothing to do with the marketing.
Close neighbours, a strict ordinance, or a machine near the house. Lower tip speeds produce less aerodynamic sound, and this advantage is real rather than claimed.
A generator at ground level is serviced with a wrench and a step. Where climbing is not realistic and a tilt-down tower is not practical, that changes the maintenance question entirely.
Pumping, aerating or ventilating rather than charging a battery. A Savonius rotor delivers torque at low speed and asks very little in return, which is the correct trade for that work.
A gate, a camera, a livestock trough heater, a monitoring station. Modest and intermittent demand where a simple machine that never needs pointing and rarely needs attention is worth more than conversion efficiency.
If the goal is to carry a meaningful share of a property's electrical load from wind, and the site has the exposure and the tower height to justify trying, a certified horizontal-axis machine on a tall tower is the answer almost every time.
If the goal is something narrower, quieter, closer to the ground or mechanical, a vertical-axis machine may suit it better. Choose it for those reasons, and not because it was presented as a way to make a poor wind site work.
Next
Whichever architecture a property ends up with, the structure holding it up and the electrical system absorbing what it makes are the parts that determine whether the installation works for twenty years or for two.
The standard machine, how it protects itself in a gale, and how to read a specification sheet.
Read the guideWhere the tower height figures in this guide come from, and how to measure your own site.
Read the guideIncluding the roof-mounted machine, which is where most vertical-axis disappointment happens.
Read the guideSources
Descriptions of Savonius and Darrieus behaviour, including relative conversion efficiency, starting torque and cyclic loading, reflect established engineering practice rather than a single published figure. Efficiency claims for any specific product should be checked against independent certified test data rather than against manufacturer literature.