More on the Embedded Microprocessor Solar Targeting System

May 2005

This month we have several very interesting subjects to cover, but first of all, let's review where we are at politically, economically and energy-wise.

Who's got  the paddle?

As we all know too well, by now the U.S. economy should be driven by a large percentage of alternative fuels and energy, but it isn't. The managers of this country have sold us down the river to the energy cartel. Congress and the White House have, in effect, committed treason against the people of the United States by accepting payola from lobbyists. Consequently inflation is out of control, taxes are skyrocketing as prices increase. All the oil drilling in Alaska won't bail us out of the hole we are in -- and the politicians sing lullabies to put the knuckleheads back to sleep.

We are losing the war in the Middle East, spending hundreds of billions each year on war which gains nothing for the world, and we have over 11 trillion dollars in debt. In fact, the debt is so out of control, we can't even pay down the interest, let alone the principle.

Nuclear war is looming on the horizon, mostly driven by madmen in the white house, and where does that leave all of us? -- Again up the creek without a paddle.

In all my life I have never seen the world turn so bad in so little time -- ever since the moron of the white house got in. Then again, the entire congress gave away our Constitution to the little tyrant, didn't they? And not one Democratic senator stood up for the real President in the 2000 election!

We've all been shafted by both sides of the aisle, and they (government officials) don't give a damn if we sink or swim. It's up to us to find and produce a working solution on the grassroots level. -- That's what this club is all about, and that's what we will accomplish this year.

In pursuit of real engineering data

There have been a number of attempts by us and other club members to establish an engineering guideline for disk turbine construction. A big part of the problem has been a lack of standardized tools to collect and process experimental data. This month all of that has changed. We have just recently picked up our first batch of 4.5-inch stainless steel disks, and will begin offering rotor kits within the next couple of weeks.

By standardizing the hot rotor/disk pack, we are able to place all of our experimenters on a level playing field. In the past we have built and tested much larger disk turbines, but they required an enormous amount of inlet energy to fully utilize their capabilities. As an example -- a 10-inch turbine with four plates requires about 10 horsepower of continuous air pressure to spool it up and maintain a constant output of about 4 horsepower.

To make life simple, we have downscaled the size of the turbine to 4.5 inches and four plates, which allows all of our experimenters to gain real engineering data at an affordable cost. Not only will we gain the necessary data we are looking for, but in the end this small turbine will become the core technology for home stored energy systems. Even though the energy output of a small turbine is a fraction of a horsepower, it fits the design philosophy of continuous output, stored energy systems.

I believe a turbine should be run at its peak efficiency point 24 hours a day, seven days a week, and that a continuous flow of electrical power from such an engine should be stored in batteries for later use.

The three subjects we are going to cover this month are:

1. Continuing development of the microprocessor aiming/targeting system
2. Using a recycled engine case for a small turbine
3. Introduction of the new 4.5-inch rotor kit for the PTBC

Embedded Microprocessor Solar Targeting System

As we discussed last month, our targeting system uses a PIC 16F84 processor to acquire the target -- in this case, the sun. I refer to this part of the process target acquisition as opposed to aiming the dish.

As we previously mentioned, Microchip's PIC processors along with ME Lab's PicBasic compiler comprise the best, lowest cost microprocessor development system for any project you may have in mind. The main reason for this position is the fact that PIC processors have general/universal use pins. These general purpose pins may be configured as inputs, outputs, serial communications, A/D converters, etc.

In our application we would like to use four of the cpu pins as inputs for the four phototransistors/LDR's (light dependent resistors). Our four light gates are divided into two sets of two -- one set for azimuth, one set for elevation.

Since the PIC processor pins may be used in either a digital mode or analog mode, we have a lot of design and operational leeway.

If we choose to operate our light gates in digital mode, the function is similar to flipping an electrical switch on and off, whereas analog mode is like using a dimmer switch -- many steps/intensities of light.

So what are the pros and cons of the two systems?

Digital mode

In digital mode we have only two states -- on/off. The light gate is set up to toggle between on/off states -- depending on how much light falls on the "eye". The cpu in turn polls the input pin to determine the state and respond to any changes. In setting up such a system there are two distinct areas to work: mechanical and electronic.

In an analog system the cpu would not only sense the on/off or shadow/no shadow state, but also the intensity of the light/shadow effect. In fact, in an analog system the shadow mask may be eliminated and the light gates set at angles (30° - 90º) to each other. (C)

On the electronic side of things, light gates may be wired for simple digital on/off response or for analog/differential gradient response.

When using an LDR (light dependent resistor), select one with a resistance value of approximately 3k-ohm to 10k-ohm in the on (lighted) state. Wiring for digital mode is shown in figure E, analog mode in figure F.

In digital mode, the LDR is used as a resistive shunt between cpu power and ground. When little or no light shines on the LDR, resistance is high (about 200k), so the cpu pin is held high. When sufficient light falls on the LDR, resistance drops to around 3k-5k, pulling the cpu pin low, and signaling a state change.

In analog mode a charge timer circuit is set up using the LDR and a capacitor. The "Pot" command causes the cpu pin to supply a charge to the resistor/capacitor timer circuit. The timer circuit charge rate will vary directly with the amount of light illuminating the LDR. This in turn tells us the difference in light falling on a pair of sensors.

Phototransistors and photodiodes may also be used as sensors, but their characteristics are very different from LDRs. While LDRs have a maximum and minimum resistance, phototransistors have an "off" state of nearly infinity with an "on" state of nearly zero. Because phototransistors have very little "on" resistance, it is easy to burn them out with excessive current. In all cases phototransistors and diodes must use current limiting resistors. A typical digital mode circuit using a phototransistor is shown in figure G, with a corresponding analog circuit shown in figure H.

In the digital circuit a 10k-ohm "pullup" resistor normally holds the cpu pin high. When light falls on the phototransistor the cpu pin is pulled low, signaling a state change.

In the analog circuit a 1k-ohm resistor is placed between the cpu pin and the phototransistor creating a minimum low resistance value of 1k. Variations in light intensity will add or subtract resistance.

Next month we'll look at computer programs that will read sensor values and download them to our development system screen.