Setpember 7 update

My array is on the side of a hill.

To the East is an overgrown lot with wild trees and weeds.

To the West are several evergreen trees over 30 feet tall. Behind the trees and up the hill is a residence. These reduce the sunlight, especially in Winter when the light is most useful.

Down the hill is an eight foot tall wood plank fence in disrepair.

Up the hill is my residence.

I decided to clear off the overgrown brush on my lower lot, about 50ft  by 100ft. I thought I could gain some daylight time in the morning. I did. I bought an electric hedge trimmer and an electric  chainsaw. Clearing off the vegetation would be within the capabilities of the tools.

Before I cleared off the most relevant area, I started getting first strong light at ten o'clock. After clearing off that area, my daylight begins at nine o'clock.

If I can get the Eastern brush cleared, I could get first daylight unobstructed for a mile or two.
In summer, daylight is around six or seven in the morning. In Winter, first daylight is closer to seven or eight o'clock.

If I can get the evergreen trees removed on the Western side, I could extend my daylight length by another hour.

These obstruction issues are important. Especially in Winter when the sun rises low in the sky and is less intense. Oddly, the Sun is closest to the Earth early in January, contrary to what most would think. The Earth's nutation is greatest in Winter. This is the backward tilt of the Earth's axis. Locations below the Equator become closer to the Sun and locations North of the Equator become further from the Sun. This is where the seasons come from.

This brings me to an issue I need to think about and make a decision about. If the array is raised above the ground twenty or thirty feet, many of the local obstructions become null factors.

Of course, raising the array above the ground thirty feet, and maintaining the tracking system, is not an easy task.

The roof faces twenty degrees East of South. That is a workable bearing for roof mounting. The problem is the tracking system. A great deal of energy is lost if tracking is not used. The first approach that comes to mind is building a lumber frame attached to the roof to hold the array and tracking apparatus. This will give me both, tracking, and a South pointing direction for the array.
In Summer, there is no shortage of energy with a fixed tilt system. The array is fixed at forty-five degrees, the average tilt for the year. However, in the Winter, energy is more scarce and reorienting the array for Winter would be a good idea. This would require tilting the array forward, at least, twenty degrees. This brings up the question of the pivot point. Midpoint or top? A midpoint pivot is seesaw-like. A top pivot point fixes the bottom of the array pipe and moves the top or mid region. This is more difficult since the array has to stabilized in the lateral direction. This makes the seesaw pivot more attractive.
A tilt parallel to the roof of the house would be 30 degrees. The seesaw midpoint would have to raised off the roof enough to allow movement of the array to a greater degree.
Of some concern is the manner of attaching the lumber framework to the roof. Holes will leak and cause damage to the structure over time.
One possibility is a triangular frame with the top point at, and attached to, the chimney or the house framing there. The system would weigh enough that movement of the lower portion would not be a concern. In this situation, the house, itself, may become an obstacle to daylight, a factor that needs to be considered.

The array, in a simple form would not face directly South, but follow the roof structure. This is where the house, itself, mayblock sunlight.

Other approaches consist of ground mounting the array using long post or poles either embedded
in the ground, or in concrete blocks. The bigest issues with that approach are the difficulty in system maintenance and the permanence of the concrete blocks. Such blocks would be difficult or impossible to move. They take up yard space and are unsightly. If the blocks needed to be removed, a jack hammer would be needed, or a truck with a strong lift.

After a lot of thought about the matter, the last approach, concrete blocks and all, seems the better choice. No risk of damage to the house roof. No complex system to design and install. Direct Southern exposure is available.
If necessary, a deck could be built around the array, for maintenance purposes.

This idea came from another idea. Building a deck on the back of the house and letting the deck double as the array mounting platform. This is the best choice since an attractive deck is added and the array becomes an add on. Long treated lumber posts could be set into the ground or concrete filled containers, and project up and through the deck floor.

This, then, is the best idea, barring some technology I've not read about yet.
The back wall has two doors that can open onto the deck. The deck can be raised a couple of feet from the doorways to make basement access easy and lawn maintenance possible. Slits between deck floorboards are needed to provide drainage and some sunlight to the grass underneath.

If feasible, two arrays, one at each far corner of the deck, should be included in the design.
This will allow me to double my energy production. Combine a doubling of the number of panels with raising the wattage of the panels results in a potential twelve to twenty-four panels. If the panels are two hundred and fifty watts each, a total maximum production of three thousand to six thousand watts will be available. If the panels' wattage is raised to three hundred and fifty watts, the maximum output would be forty-two hundred to eighty-four hundred watts, or 3kW  and 6kW or 4.2kW and 8.4kW.
Something to think about.