Trackers

[PaulR]

Andrew

Just to add to what Ga Rick said. We shall look into the feasibility of adjusting these arrays to different angles.
.
However I did want to make some comments on the concept of a manually adjustable array. The biggest issue that I would see is that by moving from a fixed structure to an adjustable one you potentially lose a lot of array frame strength. This is particularly an issue for wind loading. I have seen such frames before and they were flimsily done and took at least two people to adjust to avoid causing adverse stress on the fixed modules. I have no doubt that these issues could be overcome by a well engineered design. But this would likely come at some cost. It would then be a matter of a cost/benefit comparison between the two options.

[PaulR]

Just wanted to add an image I took in India of a three stage one axis adjustable frame.

IMG_0028.JPG (257.3 KiB) Viewed 1270 times

[c*p]

The discussion about 2- and 3-position fixed arrays has set me thinking about tracking systems as well. I am wondering about the details of the tracking systems installed at DKA, and have noted the following details from the "Technologies" section of the Home Page (excuse the formatting!).

1 Large scale tracking, 2-axis, ADES: Remote control by PC; moves E to W, tilts with height of Sun
2 Large Scale 2-axisTrackers, DEGERenergie, Trina: Light Sensor; goes horizontal if overcast
5 Solar Forest 1 axis, programmed, poly, Kyocera: "Active" system, preprogrammed, 6 steps of 15 degrees
6 Solar Forest 2 axis, sensor, poly, Kyocera: Light sensors, less adjustments if cloudy
22 Concentrating (mirror) PV (CPV) Array, 2-axis, Solfocus: Accurate 2-axis, little detail given

There is no (explicit) detail whether the mounts are "alt-az" or equatorial, to use telescope-mount terms. For the non-technical, here is an explanation. The "alt-az" mount is like a gun mount: it rotates horizontally to give the azimuth, or compass point direction required. Then the second axis allows movement up and down in altitude, or elevation. An equatorial mount has the first (called polar) axis mounted North-South inclined at the latitude angle (South end highest) so it is parallel to the Earth's axis. This gives movement East or West (hour angle or Right Ascension in astronomical terms). The second axis (declination) then allows tilting North or South. Both mounts have pros and cons.

The alt-az is simpler to construct. If you take the 3-position mount shown in this thread and modify it so the tilt is adjusted by a motor, then mount the whole thing on a (horizontal) turntable, you have an alt-az system. It suits a rectangular array of panels because the bottom edge of the array is always parallel to the ground (and the altitude or tilt axis). It should be easy to engineer this system for wind resistance. The disadvantage is that to track the Sun, you have to continually drive both axes in a non-simple manner (but easily controlled by PC).

With the equatorial mount, because the main axis is angled, it suits a triangular or trapezium shaped array with the widest part at the upper end of the main axis. Otherwise, unless the whole axis is mounted high up (wind risk), it may be possible to drive the lower corners of the array into the ground. This system requires more engineering to withstand severe wind. The advantage is that one only has to drive the main axis continually to track the Sun. The second (declination) axis only needs a slow seasonal adjustment as the Sun moves higher or lower in the sky through the year. This could be done by motor or a manual multi-position system as discussed in this thread.

The best one-axis tracker for PV output is an equatorial mount with the declination fixed at zero so the panels are parallel to the polar axis. I presume system 5 is mounted this way. One may choose to tilt the declination to favour Summer or Winter depending on the seasonal energy needs.

It would seem better for the sun-sensing arrays to use an alt-az mount for its simplicity of construction (system 6?). The preprogrammed systems would probably use equatorial for the simpler control required: continually move the polar axis (15 degrees per hour) to follow the Sun through the day (with a reset East at the end or start of the day), and only slowly adjust the declination in a simple annual variation. I suspect this is so for systems 1,5,6.

From pictures I have seen, I believe system 22, Solfocus, is alt-az. Could the system descriptions please be made clearer on the mount types for all the trackers?

There is a further point to this post. With the manual multi-position systems, it was debatable whether the gain in going to 3-position was justified. Here I am wondering if there might be simpler mount options than equatorial, which capture most of the gains without the compromises in complexity. I have not had a chance to do the calculations; for example a fixed-tilt alt-az might capture the majority of the gains of a tracking system. This would amount to simply putting a normal fixed-mount array onto a turntable. Further, the drive system might be simplified too. Perhaps start it facing E, and rotate through N to W over 12 hours as the most basic option? This ignores complexities such as the Sun rising in the NE in Winter but the SE in Summer, and a sun-position correction called the "Equation of Time". An even more crude system could be like a see-saw, with the axis N-S and the see-saw length E-W. The system would tilt to the E in the morning, and at noon flip over to tilt W for the afternoon. I suspect this is simplifying tracking (and sacrificing output) too much! Or is it that once you introduce an axis and a way of moving the array, you might as well go to the 1-axis equatorial system?

The reason a crude tracking system should capture most of the gains (except for concentrating systems which need to track accurately) is that the output depends on the cosine of the angle between the Sun and the panel normal (perpendicular). Thus even if the panel orientation is off by 20 degrees (a full handspan at arm's length), output should still be 94% of optimum. Also, one would lose less output for greater inaccuracy in the early morning and late afternoon because the output is lower anyway when the Sun is low in the sky.

I hope others may contribute thoughts and have time to do some calculations on these ideas. c*p