I'm very familiar with that relay from a second source, Potter and Brumfield's T-90 series. Read up on the LM358 (dual version of the quad LM324), 2N4401, and TIP29.

ak

 

OK, so I brought one of these little panels outside today and got a reading of 1.4volts and 5ma... it's a bit overcast but that's what it produced... it's pretty small amperage but I guess the LM324 just has to see that there's a differential, correct?

 

Correct. One of the two panel inputs needs an adjustment to compensate for unequal efficiency. 1.4 V is excellent, lots of noise margin.

There is one schematic where the two sensors are connected in inverse parallel (with current limiting resistors in your case), so when the system is pointed dead on the net sensor voltage is 0 V, and the voltage goes positive to move one way and negative for the other. That is either an industry standard way of doing things, or a bit too cute to try in your situation. I'm puttering away at a schematic.

ak

 

Just got full sun temporarily, ran outside and read 2.5volts and 6.5ma... not sure if you need this info but I thought I would throw it out there...

 

That's good info. It means that the voltage range of the cells does not exceed the input voltage range of an opamp running on 9 or 12 V.

ak

 

Well I found an LM324 I can harvest, I guess this would be good if I decide to do a dual axis tracker a bit further down the road

 

Correct. One of the two panel inputs needs an adjustment to compensate for unequal efficiency. 1.4 V is excellent, lots of noise margin.

There is one schematic where the two sensors are connected in inverse parallel (with current limiting resistors in your case), so when the system is pointed dead on the net sensor voltage is 0 V, and the voltage goes positive to move one way and negative for the other. That is either an industry standard way of doing things, or a bit too cute to try in your situation. I'm puttering away at a schematic.

ak

Where did you find that schematic? I've seen lots that require LDRs, but haven't found any that sense with pv cells.... I'm understanding more every day, but can't quite get it down on paper

 

I remember seeing this a few years ago and thought it was clever.

This might be similar to what KeepItSimpleStupid was talking about.

 

I would like to trigger a normal 5 pin spdt 12volt relay coil tested at 85ohms. Would a transistor be able to do this? And if so, how would I go about figuring which transistor would have the optimum parameters? I realize these could be loaded questions, but any help I could get would be greatly appreciated! Thanks in advance

Here is circuit using relays driven by bipolar transistors, in the way you describe.
http://piments.com/solar/tracker_MkII.svg

This is the cheapest and simplest detection technique and uses two green LEDs as light captors. ( Yes they act like photocells as well as light emitters ). A couple of micro switches stop the motion in the dawn and dusk limits.

It uses fairly robust relays and 24V motor powered at 12V so as to provide slower movement. This will turn a standard , roottop 250W panel with the right motor. You could use a motor out of toy car or an old audio cassette player etc.

Another nice trick is to use a garden LED path light to detect end of day and trigger the return motion automatically. Just unsolder the white LED of the lamp and add a red one ( lower voltage drop) then connect the other leg to the base of one of BJTs.

here is less tidy version as a gif for those whose crippled browser can not display scalable SVG
http://piments.com/solar/tracker.gif


The cct with the relays will fit on a 2" square of matrix board and can go inside a jam jar and screwed on to the same support as the solar panel. Leave enough leg on the LEDs to tweak them so that the panel stops perpendicular to the sun. It will track the sun position to within about +/-2 degrees.

Set the axis on an equatorial mount ( axis inclinded at local latitude angle ) and it will be spot of from dawn to dusk. This solution is fully automatic 24/7 .

 

Since this is just a binary on/off use of the transistor, the only thing that needs choosing is the collector current and enough voltage to allow for a bit of resistance to switching spikes ( there is protection around relays.) . General purpose BJT.

 

FYI: https://contest.techbriefs.com/2010/entries/sustainable-technologies/326

FWIW: DiseqC is a free antenna positioning protocol. (think satellite dishes)

I don't think he is using the DiseqC protocol. He is just using the motor drive.
Pretty much based on my invention of using 2 LEDs connected back to back and using a micro to drive a relay H-Bridge.
Here are a couple other interesting tracker designs:
The LED5 is designed for low power motors.
The LED555 is a new invention where light intensity controls time.
The LEDProp is a sensor similar to the LED555 but digital inputs to a micro.
The LEDAC is a tracker that is powered from a low voltage AC source yet drives a PM DC motor.
I have many other original designs on this page.
redrok AD0TJ
redrok@redrok.com

 

I don't think he is using the DiseqC protocol. He is just using the motor drive.
/QUOTE]

OK, you can think what you want. Do me a favor and goto the page https://contest.techbriefs.com/2010/entries/sustainable-technologies/326 and search for "diseqc" on the page and we'll talk.

You might be right, he could have used the salvaged motor.

 

OK, you can think what you want. Do me a favor and goto the page https://contest.techbriefs.com/2010/entries/sustainable-technologies/326 and search for "diseqc" on the page and we'll talk.
You might be right, he could have used the salvaged motor.

He, Bryant Baker, said:
10. Motor - Modified DiSEqC motor is controlled by the Solar Tracker Sensor and capable of bearing loads up to 32lbs.
This sounds just like what I have done here:
Sadoun DG-240 H-H Mount which uses a DiSEqC drive in which I discarded the electronics.
Granted, there are tracker systems that do use DiSEqC protocol. But, his has a pair of LED sensors just like what I do.
redrok AD0TJ