Cad Cell to track the sun

[mikey ny]

Hello,

   I finnaly finished most of the improvements on my homebrew hot water solar panel. Another week or two and it will be completly finished. Now I am experimenting with a cad cell from an oil burner and a damper motor from a commercial heating unit. The cad cell works well if the eye opening is constricted to about 1 eight of an inch. It will only sense the bright sun when it is directed straight at it. I could have the sensor trigger a relay to run the damper motor and move the solar panel which will be mounted on an axis until it is in the direct sun.

 One of the problems I think i will encounter is in no sun conditions the panel will continuously search for the sun. I could put it on a timer for nite and day, but that won't help on a cloudy day. I hope to post some pics sooner or later. Also, with the panel on a single axis the sun's angle will change through the season and the sensor may not pick it up. Any suggestions??? Will it work, or is there an easier way. All parts are free so far.

                                               Mike

[Nando]

First define what a cad cell is in detail.

I may know it by another name.

Can it be a Cadmium light detector cell ?.

Single cell detection will not work well, one needs to have a differential set up for solar positioning, even if over cast cloudy day.

The actual one is one like to be used for a Solar Panel Photovoltaic collector.

Some are sold in Internet and there is one that sells one with 4 photo detectors ( large LEDs as detectors) quite good for your idea.

For yours to work you may need at least 2 detectors with differential setting and phase angle separation to equalize the centering of the panel.

Nando

[alcul8r]

If the problem were just changes with the seasons, you could use a slit instead of a hole to restrict your light input.  

All the circuits I have dreamed up or seen, however, need two sensors and amplify the difference between them to keep the unit pointed at the sun.  I fail to see how a single sensor can track in bright sun, hazy days, etc.

Is the increased efficiency for a hot water unit worth the complexity of a tracker?  Wouldn't it be easier to add a little storage or surface area and not worry about a few extra hours?  

I suppose if I was going to track I would use mirrors hinged on the sides of the units and move them with the sun.  I have thought about using this method to increase the heat in a stationary solar cookstove lined with bricks.

Rex

[dinges]

Fully agreed here.

I'm assuming you're speaking of a CdS-cell (Cadmium-selenide), but specify.

If you're looking for a schematic/building kit for a tracker, check out this site:

http://www.redrok.com/electron.htm

'Somewhere' on that page is the design of the LED3 tracker. Should satisfy your need. It's not using CdS-cells as light detector but LEDs.

Search this board for 'redrok' and you should get more info on the subject.

Peter.

[dinges]

I forgot to add: I agree with ALCUL8R. I wouldn't bother tracking a solar hot water heater. PV Cells yes, perhaps. I'm planning on building a tracker for a few small PV panels (old parabolic antenna polarmount), but I'm really doubting whether even there the extra cost, effort & complications are worth it. For solar hot water heaters, I wouldn't track.

Peter.

[drdongle]

 Actually Mother Earth News published ( and I built) a single detector design that worked quite well. It also had an "auto return to start position" function. It used one photo transistor, an SCR, three relays and two microswitches for "end of travel" detection. My version eliminated the SCR. The secret was the shadow box that the detector was mounted in.

 A two detector design could be set up to eliminate the "hunting" problem.

[mikey ny]

Thanks for the info,

   The cad cell i have is from an oil burner gun, I have not yet actually identified which type it is, but it sense's the bright flame of the burner and, in the abscence of the bright flame it shuts down the oil pump to prevent oil from entering the combustion chamber without an ignition source. I like the idea of the slit instead of a restricted hole, i will definitly experiment with that. I have an old universal joint with a flange on it to work as a dual axis mount. I am sure I could adapt 2 or more cad cells to work in conjunction. Experimenting is what it is all about for me, for now anyway. I realize I probably won't get much more out of a hot water solar panel by tracking the sun but the tracker can be fitted to anything in the future. My goal is to make anything (out of spare parts and spare time), work to it's best ability. I am pretty sure I could buy a computor program to track the sun for a few bucks also. Also using and moving mirrors is another possibility, and maybee easier.

                                      Mike

[mikey ny]

Well,

 after looking at that site I guess my tracker idea is pretty primitive since it is mostly mechanical.

[Volvo farmer]

We've had some cloudy days here recently, not completely overcast all day long, but signifiacantly cloudy at times. Every time I've been able to observe my Redrok tracker, it is always pointing at where the sun is or should be at that time of day. It also parks in the east every night. When I first got interested in tracking, I bought some relays and CDS cells and was going to build my own. But this thing from Duane is a work of art, it only costs $35, and it seems to work perfectly. If you seriously want to track the sun, I think these Redrok trackers are the way to go. Duane has improved the design over the years and seems to have worked most of the bugs out. I think I'd pay over $100 for a ready-built circuit that works as well as this one does but they cost a  THIRD of that. It's a no-brainer IMO.

[dinges]

Good to hear you are satisfied with it, Volvo.

Am planning on building it myself (from scratch, not SMD). It was esp. the parking function that I was having doubts about.

In your board, does it include the 22Mohm (I think) resistors over the 2 'outer' LEDs? (I hope this makes sense...) He said somewhere he had removed them because of difficulty finding them in SMD, I think. Is yours with or without these 2 high-R resistors?

Peter.

[adelaide]

Hi  there Mikey.

 I use 1 sender in some poly tube on a metal strap to adjust the summer winter inclination. I use a cordless drill on a long thread and with a nut welded to the tube as a mover .I use the solar panel output as the trigger to return the panel to the start position for the next days operation. It only can go forward when the sensor detects light (If you tap into the output line of the panel before the blocking diode this will swing negative with low or no light), very occasionally if it is a cloudy day I use a manual switch to catch up. I direct trigger a car central locking control box with the bigger light sensor (cad sulphide cells from ratshack) You need a cell with the highest output in milli amps that you can find.

I am using the tracker on hot water, solar and air heater it is useful in winter as well it is definitely worth tracking in winter. It is simple and can be improved but works well. Tip make it big so you can add other projects  I put post out the back and attach wood to adjacent collectors so 3  x  4 m square collectors move together. And if the panel is under tank level you may not need a pump. Air bubbles tend to easily collect so it is important to have all your warm pipes rising evenly with no chance to trap air.

I hope that has given you some ideas.

Regards.

Aaron (Adelaide).

P.S

Thank you Whirlybird for the translation from Aaronees to a more readable form of English

[(unknown)]

Another method is to use a reed switch activated by a series of magnets which travel                                           with the panel. A simple timer can control the delay between each magnet.

[Volvo farmer]

Hi Peter,

My tracker is in a peanut butter jar, 10' off the ground, about five miles from here. I don't think I will be able to answer your question very quickly. Duane is generally good at responding to Emails and he may be able to answer this for you. I just bought my tracker a month or so ago so I believe it is the current version of the LED3

P.S. What is "SMD"?

[jimjjnn]

Surface Mount Device ? Most Itegrated circuits are now surface mount instead of pins going through the circuit boards.

[dinges]

I mailed Duane once with this question. Let's just say I wouldn't be asking here if I had been given an answer.

As far as the SMD goes; you're right. A few days ago I recieved a brochure from Analog Devices with their newest products. Pages full of new ICs, not one in DIL. Fortunately, I have lots of stock. But I really wonder whether I will still be building in 10 years time. Sigh.

Peter.

[jimjjnn]

I believe that through-hole ICs may still be availabe even in the future since prototyping with SMDs is difficult to do on breadboards.

I could be wrong tho.

[SmoggyTurnip]

You can always buy sockets that convert SMD to through hole.

[ghurd]

Never heard of one.  Part numbers or suppliers?

G-