# Desing Problem with Single Axis Solar Tracking System

I'm second year Electrical-Electronical Engineering student (It's Electrical Engineering basically but naming issues I guess) and my knowledge about circuits is limited to capacitors, inductors, resistors, basic diodes element-wise and simple circuit analysis using KVL-KCL and Terminal Relations.

TL;DR: I have been getting circuit lessons for the past 2 months, so I am not that knowledgeable about circuits (not considering high school experince).

Anyways, let me describe my problem. Here's my desing so far:

Edit: I forgot to connect DC Supplies to U4. Assume there are.

Notes: U6 detects higher of V2 or V3 and outputs a voltage to change the position of SPDT relay which in return activates Q1 or Q2 to by completing the connection between bases of Q1 or Q2. U5 detects if V1 is the highest or not. If not, SPST relay turns on. If V1 is the highest then SPST relay should disconnect the circuit rendering transistors as off. But the problem is if V1=Vhighest, comparator will behave unpreditibly. So for the time being I'm stuck here. I'll be analyzing the circuit as if U5 will work as I wanted then check if this schematic will do what I want. But I'm uploading revised desing before analysis to get feedback. Thanks for your help.

Let explain my design:

1- Sensing Unit

Positive DC voltage applied stationary LDR's used a light sensing units. Changing resistance values will create different voltage drops across LDR's thus voltage values will be used as inputs.

LDRs are near a point. They're angled wrt surface like this: \|/

2-Control Unit

a)Decision Subunit This unit compares input voltages and gives highest input as output. This unit is incomplete and I'll be explaining what I want to and my question in a second.

b)Function Subunit This unit generates a PWM wave that drives a PWM Servo Motor. As you can see this unit is also incomplete and I'll be asking about this too.

Now, my need:

I need to obtain 3 different Square waves each input. For this purpose I tried using transistor as swicthes. When base voltage meets a criteria it'll turn on, thus changing the duty cycle. (So I want 3 states: Q1-2:off, Q1:on Q2:off, Q1:off Q2:on) But I don't know how I can turn my Vhighest information to a controlling information for transistors' on/off states?

I thought that if I can compare this Vhighest value with V1, V2 and V3 and find out which one it equals to, then find my way from there but I don't know how I can do it, if this method even sounds plausible. So I need advice about how to complete the incomplete parts of my desing, basically.

Notes: If you need further information about project, I can give you this: 3 different PWMs are required for obtaining 3 different states from a motor. If V1 is the highest I need a specific PWM which indicates that light intensity at LDR1 is the highest and changes the angle of motor to a specific value. If V2 is highest then I need another specific PWM and the same for V3.

Also this circuit should completely be analog. Please feel free to inform me if my problem is not clear.

Another Note: The elements that can be used are limited with LDRs,LEDs, resistor, diodes, capacitor, inductors,op-amps, transistors and servo motor without encoders(don't know what that is)

More notes: Thanks to @Trevor, I have found out that my pulse generator is incorrect. I'll be changing that part of my desing in a day or two but please feel free to give advice until then.

• Electronical??? – John D Dec 7 '17 at 18:05
• At the time when this area opened as a subject to study, it didn't cover all electrical engineering topics of the time so they opened 2 different area as "electronical engineering" and "electrical engineering". Then when the economical status of the country allowed a single area to cover all topics, "electrical and electronical engineering" which is equivalent to "electrical engineering". As I said, naming issues. – Samed Yalçın Dec 7 '17 at 18:26
• Oh, and electrical engineering studied high voltage while electronics studied low voltage. Weird stuff, don't bother. – Samed Yalçın Dec 7 '17 at 18:36
• Something sounds off with your plan here. The signal you should be sending to the servo should be a pulse which defines the current position. If there is an offset in the sensors the position pulse should be extended or reduced in the appropriate direction. – Trevor_G Dec 7 '17 at 18:52
• hmm.. ok that changes things a lot.. at least as far as what the sensor information means. – Trevor_G Dec 7 '17 at 21:33

Here is my interpretation of what you need.

U1-U4 compare the LDR values to detect when the left or middle sensors are the maximums. Depending on the outputs of those the voltage divider with three pots is adjusted to generate a demand POSITION voltage. This demand is capacitivly filtered to produce a slower swing between demand locations and make the servo move more smoothly.

The 555 circuit generates a 20mS sawtooth waveform. This is compared with the POSITION voltage to generate the appropriate width servo pulses by U5.

The reset circuit beside the 555 is supposed to ensure all the capacitors have had time to settle before the servo pulses start being generated.

CALIBRATION

1. Monitor servo pulses with a scope and adjust the saw-tooth current with pot X4 to set the pulse period to approximately 20mS.

2. Illuminate the left sensor LDRL, or short it out, and adjust pot X7 to position the servo to the correct left position.

3. Illuminate the middle sensor LDRM, or short it out, and adjust pot X6 to position the servo to the correct centre position.

4. Illuminate the right sensor LDRR, or short it out, and adjust pot X5 to position the servo to the correct right position.