Working on a new Site

Hi, as I promised, I'm working on a new site which also has the facility of discussion and user article submission. Won't say much, go check out

Monday, April 17, 2006

Knowing Sun Tracker

Sun Tracker, as the name says, is a device that helps a Solar Device { it can be a cell, panel, or dish...etc.} keep track of the Sun from morning to evening.

But there is lot more to know than just this!
Sun Trackers are generally not used for simple solar panels but they're absolutely essential for a solar power plant.
But, if you believe me, Sun Tracker can help increase overall efficiency of a solar installation by over 40%
And I proved it and made a simple device to show the effect. It helped me bag Intel Science Talent Award 2004.
So, lets start with the basics;

How it increases efficiency?

Common sense tells us that when an object faces the sun directly, it receives more heat. Try it! Face your left palm in the direction of sun on a sunny day, you'll feel heat, and now turn your hand clockwise, what do you feel?
Similarly, when a Solar cell faces the sun, intensity of light falling on it is maximum. There is a mathematical law associated with this, called Lambert's Cosine Law, which says that intensity of light falling on a plane is directly porpotional to the cosine of the angle made by the direction of the source to the normal of the plane. See below;
Lambert's Cosine Law
From the Equation, it can be understood that when the angle is '0 degree' i.e. when sun is directly overhead, magnitude of the equation is maximum as "cos 0 = 1"
It follows that if we somehow keep the panel aligned with the sun, we'll operate our solar panel on maximum light possible all the time. Thus, compared to a fixed panel, we'll get more power from a tracker equipped panel. This is what I proved in the Science Fair, ~45% increase in power.

How does it work?

I made an Intelligent Electronic 2-Axis Tracker. 2-axis means in addition to normal morning to evening movement, it also takes into account seasonal movement of the Sun i.e. north-south movement.
I'll explain the basic circuit here;
To sense the position of Sun in the sky, two light sensors {called LDR} are used, in the following configuration; left side of the picture shows two orange colored sensors mounted on sensor-board. Right side describes the case when Sun's position shifts.

Clearly, you can see that when the position of Sun shifts, one sensor comes in the shadow of the opaque object. When there is less light on the sensor, its electrical resistance increases. Now here, left side sensor's resistance increases and the right one remains the same. This change is detected by an Electronic Circuit which then moves the sensor board {here clockwise} so that once again both the sensors are equally exposed. This way, it corrects itself continuously. And the output from the circuit is used to control a Solar panel, using DC Motors.
As I made it... (click for larger view in a new window)
As I made it...

How much does it cost?

You'll be surprised to know that actually you save quite a lot of money by using this. As I've told earlier, there is 40% gain in efficiency, that means you'll get 40% more power by paying the same price.
Let's suppose that I want 1000W power from Solar installation,

Case I: Cost(I)
Rs.200/W($4.50) [multiply] 1000 = Rs. 200000 ($4500)

Case II: I buy a Sun Tracker worth around Rs.5000 ($ 112) , and get 40% more power. To get the same energy output as earlier, I need a solar panel with 40% less power output, i.e. 600 Watt power panel.

So, Cost(II) = Rs 200/W [multiply] 600 = Rs. 120000 ($ 2700) [plus] Rs. 5000 ($112) = Rs. 125000 ($2812)

Following Case II, I save Rs. 75000 ($1688) , These savings include the cost of tracker!
Isn't that lots of money??

But why it isn't widely used then?

Because, it is difficult to install on roof tops, rotating space is required! But, it can be used on Street Lights. I proved in the science fair that it has potential and moreover, for small applications like street lights, tracker needn't be too much accurate, cost of tracker can be kept down.
See, even if we take worst case scenario, the cost of tracker can never come anywhere near the huge savings we get by reducing power of Solar panels.
For home systems, there is a new idea see here

Why these are essential in Solar Power Plants?

Solar Power Plants generally use concentrating mechanisms to get intense sunlight used for producing power. Concentrators focus sunlight on a point. To maintain focus, a concentrator needs to be accurately aligned with Sun, all the time. Without sun trackers, they cannot focus sunlight and are rendered useless.
Didn't understand something? Need more data? Do you know more? You like it?
Tell me! Click on 'comments' below, provide your feedback, Now!


Anonymous said...

wind and solar on oil rig could generate power ,transport the power via laser to shore base receiver for distribution.

maybe far into the future

Mridul said...

Nah not "that" far into the future as already there has been some tests of using microwaves to transfer electricty. And it can be done!

daniel said...

Hello!Can you hep me please?
I am working on a project on free energy source, and I need a way to warm up a metal plate using the sun's energy.
The problem is that I need to find a material that can absorbe as much heat from the sun as possible, and I need some figures, in terms of W/sqm, but unfortunately so far I didn't find any figures :(

elango said...

hi i am elango,me too interested in science and i am doing my engineering in electrical and electronics
i had thought of ideas of transmitting power wirelessly and creating energy independent world
if you have any info. about this topic
please share with me

Mike said...

Hi Mridul,
That'a a fine project, and a good idea...
But think of this: Suppose I want to make
a tracker for a 20KW Solar installation,
isn't it difficult to apply your system to it? Any suggestion on creating a cheap solar tracker for that case?
mailme at:

Mridul said...

Hello Kastrinos,
Thanks for the comment.

I feel glad to tell you that yes it is possible. 20 kW solar installation means big panels or concentrating systems, the only thing that need to be changed in the system is the mechanical design, the electronic circuit can remain the same. And mechanical system is no trouble, rotation speed of the motor required is very less (less than 5 rpm) so a small low cost motor will suffice, you need good quality gears to reduce the speed.
I've seen a 25 kW solar installation using parabolic trough system. So in all, not a trouble some issue at all!

Best Regards,
Mridul Kashatria

Mridul said...

oh sorry, I meant, Hello Mike, sorry I got the wrong name from the e-mail address

chalacuna said...
This comment has been removed by a blog administrator.
Anonymous said...


It is quite a thoughtful idea. I am quite impressed. Especially using LDR sensor configuration. I would like to share my thoughts. May be you have addressed all of them already.

1) Calibration of different LDRs, i guess 4 of them, is an issue. This has to be done along with associated circuits.

2) The leakage thru the circuitry makes it not very useful for small panels.

3) Current taken in running those motors and friction in gear system.

4) May be we could make a little microcontroller, and program it to power up this monitoring circuit at regular intervals, say 15 min, and then read the LDR voltages and make a small plant adjustment.

I would like to know if we can collaborate. I am not sure of the viability of these techniques.

geoblr at yahoo dot com

Mridul Kashatria said...

Hello Anonymous!
Thanks for the appreciation!

Regarding your thoughts;
1) LDR calibration is simple, as the circuit has a small variable resistor which is used for 1 time calibration. However, I understand that its not that reliable.
2) Yes you are right current leakage is there, but it is very less, in the order of milliwatts from the ICs. Motors are coupled to the IC with zener diodes which allow the current to pass only when the voltage reaches certain level, so current leakage from the motors is avoided. Also, the circuit is active only when sufficient solar power is there so, no leakage from the battery.
3) As you can see that rotation speed required (of the panel) is way too low (<1 rpm), torque (rotating force) output from the gear box is too high as compared to the motor (in my project its 3600 times) so the frictional torque has almost negligible effect.

About the microcontroller, you are right, I'm thinking about this.
Some of the benefits of using microcontroller;
1) Reliability
2) Programmability, flexibilty
3) Ability to handle some additional tasks like auto position at night, immunity to varying Sunlight, battery charge control, auto. clean panel etc etc

But it could be costly, I'd like to work out the cost of the project with a microcontroller.

Actually, my aim is to make a kind of fully maintainance free and cheap Solar Powered lighting system to be used at remote locations. The design should be such that it should utilise full potential of the Solar Cell and should last long.

Best Regards,
Mridul Kashatria

Anonymous said...

On thi site there's a good and affordable solution for sun tracking !!

Eromrab said...

Hey man, i don't really know what i'm talking about, as i haven't spent much time researching this stuff... but what i was thinking about with the "sensors" that react to the shade: what if you used solar panels from a calculator... and instead of just using 4 in the 4 corners, you could surround the thing with them in a circular pattern so that it was more accurately aimed at the sun (i also assume that the closer to the middle they are, the more accuracy can be attained?).

So basically, a digital type circuit would need to be setup to read the calculator panels... when they were receiving light they would send their power to the motor for the arm and when they were not receiving light, they would use the power from the other panels to move the setup in the desired direction... sorry if this is confusing sounding but i'm having a difficult time trying to put this image into words.

So yeah, the motor would ideally work from the power from the calculator cells, thus freeing up all the power from the unit itself to goto energy production for use or storage.

Mridul Kashatria said...

Hi Eromrab, felt glad that you commented!

Yeah you are right, this system would be more accurate. But its application areas might be at some Solar Power plant where high accuracy trackers are required.
Most ideally, it'll be a CCD sensor (that used to capture image in digicams) that gives almost 99.9% accuracy, as it contains millions of point sized sensors.

imagine building a super accurate tracker using a low cost digital camera. If you some how take digital electronic output from the camera. Hook it up to a computer with a software that analyzes images and powers motors.
Wild! isn't it?

manjusha said...

thats really great!

congratulations for your idea

coming to applications, there can be many applications only if we can make it a little more feasible..

we can have it over huge buildings,apart from just street lights.we can get more applications as we go ahead..

i dont even think its difficult to install it..

Anonymous said...

A microcontroller is an ideal choice for a controlling device. As an example the Freescale MC9S08QG8 is $2.34 from Digikey, has 8 Analog to Digital input pins (4 would be needed, one per LDR), leaving 8 I/O pins to drive the motors. This part will operate on 3mw (1ma @ 3Volts), and has ample resources to drive a solar tracker. There are many other suitable micros as well. I would consider using small photovoltaic cells (perhaps 2" by 2"), instead of the LDRs. They would then act as both sensors to determine solar orientation and supply the energy to power the solar tracker itself, independant of the main PIV array. Using a small Lithium Ion battery to accumulate this energy, there would be power on demand (even at night), to power the controller and "motors". An area of research might be using a solonoid ratcheting mechanism to incrimentally rotate the panel say 1 degree per click, driving the panel as needed by the micro. A micro pin would drive a left/right mechanism for returning the panel to morning position. The Micro has sufficient programming capability and an internal clock, so it would also be possible to store in its program all of the data required for the micro to "know" where the sun is at any date/time and so operate as an open loop system (no sensors required).
Regards for an excellent Blog.

krishna said...

Hi Mridul!

I'm an E&E student from Malaysia

Great to know abt u n ur ideas..
Do you have any ideas to extend ur circuit to have some more features like..tracing night and returning automatically to the initial position where the panel must stand in morning? do u have any ideas to implement these without using Micro controllers ,as u said mic is so exp and i don't have enough knowledge to program a mic(just started learning PIC mic only)

I'm learning using C to program PIC mic but its too early!!

Mridul Kashatria said...

Hi Krishna,
The tracker circuit already has a system to prepare itself at night.

The panel has a trip switch at the end, when it rotates fully towards one end the switch is depressed and the circuit gets broken. Now, this happens after evening. You can use this switch to trigger a circuit that prepares the panel for next morning.

William Hart said...

Solar tracking is definately a great idea, but I was wondering about the energy involved in moving the solar cells. Do you this data?

steven said...

I want to install an off grid system in mallorca 1.2 kw I understand that trackers are not so useful with batteries especially in winter better to buy more panels and use a fixed system, is this true?

Mridul Kashatria said...

Hi Steven,
It may be true, but largely depends on the weather conditions. Sun Tracker analysis shows that it performs best in clear direct sunlight and worst in diffuse light conditions.
So if you have sunlight partially blocked by clouds, fog etc then a tracker is not useful.

ahmed said...

dear kashatria,
ur idea is great but u know we wanna use a bigger type to produce electric to a small town. u must to know how we could do that with low cost,if u do that i think u must take noble prize

andrew-cocke said...

Is there any advantage of large solar panels over an equal area of small solar panels?

Purvil said...

hi..its a wonderful idea..can u tell me if u know of any solar tracker which does not use any electrical energy!

Mridul Kashatria said...

purvil, look for passive sun trackers.

Anonymous said...

we want more about components and especially about motor

sunil said...

hi mridul.....

excellent activity done by u at this level and age....

actually i visited ur blog abt to 6 time and i m really impressed....

this is sunil patel frm gujarat...and i m in B.TECH final year electrical engg student....
and i m right now doing project on solar tracking system...

abt this toic i wnt ur help..

becos i m going to make tracking system like ur model but ma project partner is think to make project on micro controller based so we r little bit plz tell me which one is better.....?

and in ur ckt, i know its fundamental ckt but can i know more and detail in this ckt.....?

and i wnt know abt the price of solar panel rating abt 12volt and 40 watts ....i asked one dealer he said me tha price of multicrystline panel on same rating is 8000.....

and can u give the info abt from where we get the panel.....

Mridul Kashatria said...

Things not shown in the circuit diagram are the resistor values and the transistors. Well I can tell you the transistors, the SL100(T1) and SK100(T2) . As you'll see the total circuit consists of two identical parts, one with S1 S2 and other with S3 S4(thats for 2-axis tracking). The Sensors are easily available LDRs and the IC is "op-amp 741" (op-amp means operational amplifier) . The resistors you can configure yourself (even I dont remember the values now, just experiment with them). Its very easy once you understand the funda behind it.

We are just checking the voltage drop at both the sensors and finding the difference and then amplifying this difference voltage and provide it to the motors. When the two sensors are equally exposed then the voltage drops will be the same (as resistance of LDRs would be the same), thus their difference would be zero. But when one sensor is in dark its voltage drops more than the one in light (cause resistance of LDR depends on light), so the difference say comes out as positive, means the npn (SL100) transistor will work and rotate the motor, but in the other case when the difference is negative then the pnp(SK100) transistor activates and rotates the motor in opposite sense.

But, this design is very basic, just to illustrate the thing, it works though. I suggest you to go for a programmable microcontroller if you really want to control a outdoor solar panel, otherwise this circuit is fun.
A similar electronic circuit diagram (but single-axis) with full parts list is given in a Book titled "Electronic Projects For the 21st Century" by John Iovine.


sunil said...

thanks dear.....
but i wnt know abt the price of solar panel ,,,,

supreeth said...

hi mridul,

iam very impressed by ur blog.we have decided to build a solar tracking system using a MC for our final year project..........can u or anyone else who wants 2 help plz send me a detailed ckt diagram of ur would b loads of

thank you


Anonymous said...

I'm researching solar trackers. Could you please recommend some good resources?

thecrazymick said...

Do you have (or know where I can get) plans to build a small sun tracker for a solar cooker?

Anonymous said...

Hi Mridul, thats an intelligent design, was this your original idea or you just developed upon something already available

loveena lionel said...

hey, Mridul!
Can you please tell me how LDRs can be mounted on the solar panel?