Dual-Axis Tracking Generates More Power

Dual-axis tracking systems generate more power than fixed arrays by continuously positioning the PV array so that the incident angle of solar energy is 0°.

Mark Scanlon, CEO of Sedona Energy Labs
November 23, 2010 | 16 Comments

While a tremendous amount of research and funding is going into trying to increase the efficiency of photovoltaic cells by a few percentage points, there is a readily available solution that yields a 40% increase in produced power today - dual-axis tracking [1]. By simply moving the PV array so that it is aligned with the sun throughout the day and seasons, you get a large boost in produced power at a small incremental cost. Of course the cost depends on the design of the tracking system. In today's market, this cost ranges from under a $1.00/produced watt, to around $3.00/produced watt. We are talking about produced watts rather than rated watts.

The key to understanding the benefits of tracking is the significance of the incident angle, the angle at which the sun's rays strike the PV array. To see how the incident angle affects solar intensity and power production, we use the formula Intensity = Constant x cos Θ where Θ is the incident angle measured from perpendicular (Fig. 1). So intensity is at its maximum when Θ = 0−this is when the arriving energy strikes a PV panel perpendicularly. The greater the incident angle, the smaller the amount of energy reaching the panel.

Figure 1. Illustration showing the significance of the incident angle with respect to solar intensity and power production.

Another consequence of a large incident angle is reflection. As the incident angle increases, the glass on the front of the PV panels begins to reflect energy away from the panels, reducing the power produced. The combination of reflection and reduced available surface area is why fixed solar systems produce very little power in the morning and afternoon. Figure 2 is a representational daily energy production graph for a fixed array.

Figure 2. Representational daily energy production graph for a fixed array.

For a fixed array, the incident angle changes throughout the day, from highly acute to highly obtuse. The result is that very little energy is produced during the morning and afternoon.

Tracking Increases Solar Array Performance

An optimally aligned dual-axis tracking system produces about 40% more power than an optimally aligned fixed-frame system [1]. Because it can tilt on both azimuth and zenith axes, a dual-axis tracking system follows the sun daily and seasonally, always positioning the solar array so that it optimally faces the sun.

Figure 3 illustrates the way dual-axis tracking works compared to a fixed frame. The figures were made by rotating a 3D digital model to capture different views of the panels.

In Fig. 3a, we see a fixed-array on the left and a dual-axis tracking system on the right. They are both set to true South and we can see that the lines representing the sun's rays coming into each array are at the same angle from horizontal. (This is a parallel projection so the panel shapes are distorted.)

Figure 3. Illustration of dual-axis tracking: a) a fixed-array and a dual-axis tracking system both set to true South; b) panels tilted toward the sun in two directions, South and East; c) a view of the angle from the tracking array side; d) view from the fixed-array side

In Fig. 3b, we see that the panels in the tracking array are tilted towards the sun in two directions - South and East, and we see that the incident angle is 0° for the tracking array and greater than 0° for the fixed array. Note that changing from a parallel projection to single-point perspective corrects the panel shapes, but distorts the parallelism of the rays.

In Fig. 3c we can see the angle from the tracking array side. The tracking array is tilted toward the South so, again, the incident angle is 0° and the array is generating maximum power.

Figure 4. Representational graph of the power produced by a dual-axis tracking array superimposed over the graph of a fixed-array.

Looking at Fig. 3d from the fixed array side, we see that the incident angle is greater than 0° for the fixed array, so it is receiving less than full solar energy. When we combine the North/South and East/West "errors" for the fixed array, the decrease in the energy the array is receiving is significant.

Figure 4 is a representational graph of the power produced by a dual-axis tracking array superimposed over the graph of the power produced by a fixed array. The area between the curves represents the additional power generated with dual-axis tracking.

Shading and Spacing

Shading, spacing and the aspect ratio of PV panels are interrelated considerations for tracking systems. In the early morning and late afternoon when the sun is low in the sky, the panels in an array would have to stand almost vertically to directly face the sun. As the panels in the array rotate up from horizontal, they create a higher and higher profile. This profile on one panel eventually causes shading on the panel behind it. The closer the rows of panels are together, the sooner shading will occur. While the rows of arrays could be spread far apart, at some point shading will always occur as the sun approaches the horizon.

Figure 5. The InteliTrack Balanced/Frame design. The actuator moves the array in the north/south plane and the pivot bar (attached to bottoms of triangles) moves the array in the east/west plane.

Backtracking is used to mitigate shading. It is done by rotating the panels back from their optimal position facing the sun. This creates a small incident angle, but allows the array to produce more power than if the panels were slightly shaded.

Financial Impact of Dual-axis Tracking

Because dual-axis tracking generates 40% more power from each panel, you can achieve the same power output with 29% fewer panels, frames and so on, which reduces a project's upfront costs and offsets to a great extent the additional cost for tracking hardware [2]. On the other hand, you can use the same number of panels as originally planned and generate 40% more power and higher revenues. This reduces the project's payback time by a little less than half and also increases the overall return on investment (ROI), depending on the financial specifics of the project.

Better Tracker Design

One of the drawbacks of most dual-axis trackers has been the pole-mounted design. Most tracking systems are pole-mounted in a manner similar to satellite or radar dishes. Because the array is hung from a single point, gravity is always trying to pull them over and their wind load factor is quite high. This requires heavy-duty mechanisms to hold and position the array and massive, heavy concrete foundations, heavy-duty earth moving equipment and cranes to set the array in place − all increasing the cost of installation. Another disadvantage to pole mounting is the height which can be anywhere from 12' to 20', forcing the arrays to be widely spaced to avoid shading.

A better design would be lightweight, low wind resistance, and no requirement for an extensive foundation to support it. One solution is the InteliTrack (Fig. 5), which holds the PV panels in balance, like a gimbaled ship's compass, allowing the panels to tilt in two directions around their own axes. This design eliminates the need for a huge foundation since there is no stress from gravity or a high wind load on the supports. It is made of aluminum so it is lightweight and, without a heavy foundation, suitable for rooftops and parking lot shade structures as well as for large ground-mounted utility projects. Wind loading is greatly diminished because of its low profile and the fact that the panels can be louvered to present only a thin edge to the wind.

Acknowledgments

InteliTrack and Balanced/Frame are trademarks of Sedona Energy Labs.

References

1. Percentages derived using the NREL Solar Advisor Model tool by using identical input data, but varying the type of system among fixed frame, single-axis tracking and dual-axis tracking.

2. Panel reduction is derived by the formula: 1 - (100% /(100% +40%)) = 28.57%, where 40% is the increase in power production.

Mark Scanlon received a BA degree in biology from Northern Arizona U. and has done graduate studies in electrical engineering at Northern Arizona U. as well as graduate work in solar and sustaining design at the Wright Ingraham Institute (under full scholarship). He is CEO of Sedona Energy Labs, 2225 N Gemini Dr., Flagstaff, AZ 86001 ph.: 928-863-3245; mark@sedonaenergylabs.com.

16 Comments

William Fitch
December 2, 2010

Hi:

Single axis East - West tracking makes the most sense and from an engineering perspective, its not that hard... The biggest gain in tracking is not the peak energy available, or even the total output. Its the ability to spread the generation over a wider time frame. The output form of the energy over time is just as important as the total energy produced. This has to do with energy engineering, not money. Couple a tracking PV array with wind and you begin to have a generation capacity that moves towards consistency. This is a huge benefit when sizing battery banks, planning loads etc.. Again, none of this has to do with money, if you have your head on straight...

.....Bill

Christopher Lee
December 1, 2010

Many domestic installations around the World form part of the roofing, so you can't choose the angle or orientation, and tracking would increase the complexity of a device that's already too complex and expensive for an unsubsidied client.

'Anonymous' pointed out that light gathering could be improved by relatively simple means. I'm a bit surprised not to see Fresnel concentrators or equivalent optics that could provide some scope for tracking and reduction in reflective losses.

In fact, there seems to be some degree of polaristion between flat panels and solar concentrators that concentrate 100-fold or more. Perhaps there's room for intermediate solutions.

Concerning tracking mechanisms, servomotors and the associated electronics seem to me to be adding yet one more layer of expense, complexity and maintenance issues. Perhaps an old-style thermo-mechanical servomechanism would do the job more cheaply and reliably, with bimetallic devices or those pneumatic pistons you use for automatically ventilating the greenhouse.

In conclusion, the article doesn't do very much to dispel a sense of pessimism about the ability of the renewables industry to provide the required amount of energy on the right timescale and for the right price. Perhaps that's just because potential end-users like me are becoming a bit too impatient.

Lawrence Carroll
December 1, 2010

Rich-Barbarics said, "A warranty - short - intermediate - long term should solve the 'moving parts' reliability problem."

Good point, but warranties don't always work the way they should.

I have a 2003 Honda Civic Hybrid, and when the battery failed at 85000 miles, it caused the computer to fail. Cost to replace both: $5500.00! There went my energy savings for the past 6 years! Don't expect either main or the alternative energy press to reveal that this type of thing happens . . .

The problem (as always) is that businesses don't always either 1) tell the consumer what they need to know, and 2) honor their warranties, even mega-corporations like Honda.

I'm all for alternative energy, but don't screw me (or others) if you want your ideas to really succeed!

Alison Tottenham
November 30, 2010

This article noting the benefits of two-way tracking makes perfect sense. For maximum output the sun's rays do need to hit the panels at an angle of 90deg to the surface of the panel. Therefore, as has been commented on above, the racks do need sufficient space so that they do not shade each other when turned from East to West during the day and angled up and down to match the height of the sun in the sky.
Unfortunately in UK the installers have got hold of the figure of 35deg and seem incapable of realizing that this should be matched to latitude, which in the UK is between 40 and 60degs and will also vary according to the seasonal height of the sun in the sky. Perhaps we could make the guys on the ground aware of this!
Good article.

Paul Lindsey
November 30, 2010

If the panels are mounted in a North-South line instead of an East-West line, then there is no shading as the tracker moves from from sunrise to sunset. Supporting the South end near to the ground (but not so low that the panels can hit the ground at the limits of rotation) and supporting the North end with an adjustable height arrangement would allow quarterly or monthly tilt adjustments. This reduces the dual tracker to a single axis tracker and reduces cost. (This is analagous to piloting a ship or airplane on a set of chords of a great circle, rather than trying to follow the great circle route perfectly, which requires constant course changes unless you are travelling due north/south or due east/west at the equator.)

Also, the weight of panels rotating about an axis on a tracker can be counter-balanced, similar to the way a telescope tracker is counter-balanced in order to reduce the motor size.

Douglas Porter
November 30, 2010

The dual axis tracking product looks much better at resisting wind than previous single pole mounts and glad to see further developments and options using new geometry in this area. In Colorado we don't see many dual axis trackers being done by solar developers but single axis trackers appear to be the norm now for large installations. Earlier single axis tracking systems were challenged by Colorado's snow, wind, dust and ice but the newer generation of trackers appear to have solved some of these problems thus the increase in their use by developers. Look forward to seeing the dual axis systems go through the same evolution.

Carolyn Humphreys
November 30, 2010

Thanks for the quantity and quality of tracker information. I'm frequently asked about tracking and have been using the percentage ranges given. Now of course, I'm looking forward to hearing more about dual axis CPV (thanks Mr. Cook).
Latitude was discussed but site conditions, not so much. Maybe because it's so obvious or because it seemed to be a western dialog, but I'm wondering how much sense it could make to add tracking to a site without clear horizons. We have some hill top scenarios and some coast but otherwise CT has a whole lot of trees, hills/ridges and density. But always curious and with plenty of potential customers being engineers (all of us justifiably in love with all the cool ways we can increase output) I want to be sure to give the most thoughtful responses possible.

Nick Cook
November 25, 2010

>> Continued

Calculating the actual gain from trackers is extremely complex in reality, especially where the sky is overcast for a significant amount of daytime. However if you take the limiting condition by comparing a single panel tracker with no shading against a fixed panel that faces directly to the sun at noon and assume that insolation attenuation at low angles is negligible (both due to panel reflections and atmospheric attenuation) then the theoretical tracker gain would be Pi/2 or approximately 1.57:1. In reality; in order to make a reliable prediction of the energy gain provided by using a tracker, it would either; need to be done by experimental comparison of the two systems or you would need to combine the latitude of the installation with detailed data on annual weather/insolation conditions and the characteristics of the panels.

But to be perfectly honest if you're looking for better ROI and if you're going to resort to using a tracker then this presumably is only really of any real benefit where there are significant amounts of direct sunshine throughout the year. Under these conditions the real gain would come from using CPV (concentrating PV) with cells that not only have double the efficiency of the flat panels, but also use active material (the solar cell) that occupies less than 1%, even as low as 0.1%, of the panel area. The bulk of the panel is made up from much cheaper concentrating optics. Using this technology there is one company that claims they can produce solar electricity at three Australian cents per kilowatt hour, and from some calculations I have done based on production CPV cells I think they are probably right.

Nick Cook
November 25, 2010

As Mr Scanlon has "done graduate studies in Electrical Engineering at Northern Arizona U", he should be able to differentiate between power and energy, both of which are considered in this article but are always referred to as power, even on the graph which clearly shows "energy production" versus "time-of-day" he talks about power production. A simple way to think of the two quantities is; (horse)power is how big the engine is under your bonnet and energy is how much gas you have in the tank, in other words power is how fast you can go, energy is how far you can go. I don't know why it is that so often these sort of articles disregard the difference between power and energy, if you were reading an article in a motoring magazine and they said a car's gas tank held 50 hp that it had a 250 gallon engine under the bonnet you'd think they were talking rubbish.

Alan (comment 7), I see where you're coming from with "trackers don't give you more power" and at the end of the day it is the energy production gain that is important, after all this is what you actually get paid for and in turn pays the costs of the system. However, trackers do in fact give you both more power and more energy, at least apart from the odd instance in the year when the sun is perpendicular to the fixed panel, always assuming of course that the sun is shining and it is not a cloudy day. Incidentally in overcast conditions the optimum angle for maximum energy collection may well be horizontal.

Continued >>

Allan Sindelar
November 24, 2010

Mr. Scanlon has missed - or misled about - two fundamental points. First, as was addressed in one of the above comments, is that a 40% increase is a peak increase, usually achieved in summer, when days are long and a sunpath from NE to S to NW allow tracking to take advantage of the long sunpath. Winter gains are on the order of 15% (at our latitude of ~35 degrees N), leading to an annual production gain of about 32-35%, not 40%.

Second is that trackers don't give you more power, they give more energy. Hummpph!

Drew Haney
November 24, 2010

Anon,
As marston-schultz has suggested, a mirror would probably reflect light in one direction but cast a shadow in another. Mirrors are probably better utilized in thermal solar array-type sites, IMHO.

BTW, and slightly off-topic, if anyone out there has any experience presenting a solar farm site to Duke Energy, I'd love to hear about it at drewomatic@gmail.com

ANONYMOUS
November 24, 2010

Could not a couple of cheap mirrors also double the amount of energy to a panel?
This seems like a simple inexpensive way to get more light to the cells; however, I'm only an intrested observer here.

Chris Bunas
November 24, 2010

The above comments are pretty good. What the article also fails to mention is that the 40% energy gain is also usually above the 40th parallel where the "solar window" has a greater variation of sun angle throughout the year. In northern CA single actual axis tracking systems have produced approximately 15% more power than fixed while the manufacturer claimed 25-30%. After a few years of operation it is clear that it is more like 25% only during peak times. I suspect that the dual axis tracking would have similar slightly, higher, results below the 40th parallel. In the past when module prices were very high, single axis tracking had a pretty good benefit over fixed for system higher than 500kw. Now that module prices are below $2 it does not make sense pay for upfront and long term maintenance costs or bear the risk failure of tracking components. Given the recent challenges of having to provide ten year energy production guarantees on systems, I would rather stick to bullet proof, low maintenance fixed tilt systems and put the money into DC side monitoring and management ("smart-boxes") like SolarEdge or Solstice, in addition to only using modules with + power tolerances. Chris Bunas - Sunterra Solar Inc

Rich Barbarics
November 24, 2010

For a wind system an estimate is made for maintenance, usually 1% of installed cost per year - this could easily likewise be done for tracker systems... and even static arrays require some cleaning during the year so they are not totally maintenence free. The fact the panels 'move' causing some vibration means they should perform better in winter months when snow covered since they will jettison the snow while static panels won't. REC credits favor the greater kwhrs/KW of trackers and improve returns. On horizontal roofs there is always a tradeoff of tilt/spacing and shed effect -- it sounds like backtracking addresses this malady to some extent. If not too heavy relative to ballasted systems, the cost tradoffs (for roof structure work) should be favorable for trackers. A warranty - short - intermediate - long term should solve the 'moving parts' reliability problem.

Marston Schultz
November 24, 2010

What the author has left out is the fact that panels need to be spaced farther apart in order to avoid panels casting shadows on each other. Thus, more land is required.

Distributed generation at the site of use is a more effective strategy, reducing transmission losses.

Drew Haney
November 23, 2010

Interesting article, very well-written. However, I still believe that the economics does not bear out the installation of a dual-tracking mechanism for commercial sites. If the solar panels are permanently aligned to the "best-average" azimuth, you can initially install MORE solar panels, which are getting progressively cheaper and more efficient (not to mention less reflective) in lieu of a complex tracking system. Also, bear in mind every mount is something that will eventually need maintenance/repair. At a 1 MW or larger facility, you'd have to hire a crew just to operate and maintain the mounts and directing apparatus. But maybe that's the point......

Add Your Comments

To add your comments you must sign-in or create a free account.

Photovoltaics World Magazine

This magazine is no longer being published as of May 1, 2012. To subscribe to the similar solar energy content click here. Or, subscribe to our worldwide Renewable Energy World magazine digital edition here.