Solar Tracker Circuit Is Giving Me Issues - Appreciate Anyone's Help

[Dragon Tamer]

A limit switch does have a spring in it. If it didn't it wouldn't be much of a limit switch (it would just be a switch). Most limit switches nowadays have 2 positions, a normaly closed position and a normaly open position. What you want to use for your prodject is a normaly close switch (PBNC) so that your liniar actuator can run properly as long as the switch is pressed. Just going by a breif look at your schematic, I think that you may need a by-pass diode on the limit, so that the actuator can run in the oposite direction even if the limit swithc is pressed. By the way that link to the solar tracker does have limit switches in it, but the circuit does not work, we had a nice descussion before about that.

 

[marktinkerman]

By the way that link to the solar tracker does have limit switches in it, but the circuit does not work, we had a nice descussion before about that.

can you give directions to that "nice discussion"

 

[Dragon Tamer]

Here is the link to that thread; Solar Tracker Using 555 timer

 

[rs14smith]

U2 C/D shut the motors down when it is dark. You cannot depend on U2A/B to be "off" when both inputs are 0V.

So how about this idea mneary:

What if we placed a limit switch on both outputs of U2 C/D that becomes activated once the solar tracker has returned as far east as it can go and hits the limit switch.

Once it hits the limit switch, this turns back on U2 C/D which as you said, shut the motors down when it is dark.

When it's light again, U2 A/B will move the solar tracker off the limit switch again once the tracker starts following the sun again, which in turn, deactivates the limit switch (deactivates U2 C/D), and once it is dark again, returns back east until it hits the limit switch.

Refer to the image below I quickly sketched up:

 

[mneary]

The limit switches would do the job of stopping it before it got too far. I'm not understanding how the circuit would seek East in the morning.

 

[rs14smith]

The limit switches would do the job of stopping it before it got too far. I'm not understanding how the circuit would seek East in the morning.

Well correct me if I'm wrong, but doesn't U2 A/B determine rather or not to turn the solar tracker left and right? The only problem they have is, if it's night time, you cannot depend on them stopping the motor from searching for sunlight at night.

So as long as U2 A/B are still active this whole time, the tracker can still follow the sun east or west in the morning time. My understanding is, U2 C/D's job is just to prevent the dark issue of the motors running, thus my reason for picking around with it's outputs.

 

[rs14smith]

Okay, so here is my new implementation which may be wrong, but from the virtual test performed, in theory it seems it would work correctly, however, I remember mneary saying in order to see what is fully going on I need to hook up the Multimeters across Q1/Q2 to see if they are turning on and off which I haven't done yet.





Note that I couldn't find the limit switch symbols in multisim so I just used 2 regular switches. That's the only thing I changed in the circuit was placed two switches on the outputs of U2 C/D

 

[mneary]

You would place one limit switch from the output of B&C to ground and the other from A&D to ground. Remember, you won't be at both limits at the same time.

Might be easier conceptually if you labeled components east and west.

 

[rs14smith]

You would place one limit switch from the output of B&C to ground and the other from A&D to ground. Remember, you won't be at both limits at the same time.

Might be easier conceptually if you labeled components east and west.

Okay, I'll try messing with it some more later today and make those changes, as of now, I need sleep asap! lol

 

[mneary]

It's still yesterday for me. Later~~~

 

[rs14smith]

You would place one limit switch from the output of B&C to ground and the other from A&D to ground. Remember, you won't be at both limits at the same time.

Might be easier conceptually if you labeled components east and west.

Okay I didn't really try to comprehend your post the other night because I was tired, but after re-reading it, I'm not quite sure I understand why we would want more than 1 limit switch in the circuit?

As in the sketch I drew up as you can see there is only 1 limit switch that determines if the solar tracker has returned to the east postion. So it wasn't a mistake that I made when using the same limit switch in the schematic, but rather thinking about it from a constructional point, as when limit switch U2 B/C enable, so should U2 A/D.

But then again, maybe you know something I don't lol which seems to always be the case, but that's the reason I did that..

 

[rs14smith]

Also, after we implement these limit switches in correctly, would it be safe to say we created a rough version of a single axis tracker ?

I'm hoping I can find time before I go to bed to add in those ground connections and upload that circuit to see what needs to be done next.

 

[nickelflippr]

Here's your basic dark activated circuit for the return to East function. It would depend on the limit switch, with a resettable fuse as a backup, to shutdown the motor. I am looking at using signal transistors that will OR together the dark sensor and normal East motor activation circuit. I like the clicking of the spdt relays, and may try them since they are on hand.

 

[rs14smith]

Here's your basic dark activated circuit for the return to East function. It would depend on the limit switch, with a resettable fuse as a backup, to shutdown the motor. I am looking at using signal transistors that will OR together the dark sensor and normal East motor activation circuit. I like the clicking of the spdt relays, and may try them since they are on hand.
View attachment 41025

Yeah the only problem I has is how to implement it into the current design. We're so so so close to creating a single axis tracker, but I'm still not sure how to implement your design (dark sensor) nor mneary's at the moment.

 

[Dragon Tamer]

I was going to say that you could just use a 555 that is activated by the dark.

(p.s. the limit switches in multisim are under electro mechanical)

 

[nickelflippr]

Yeah the only problem I has is how to implement it into the current design. We're so so so close to creating a single axis tracker, but I'm still not sure how to implement your design (dark sensor) nor mneary's at the moment.

I need to look back for Mneary's dark sensor, missed that.

Get a book like the Forrest Mims mini notebook series, this one has basic ttl, cmos, transistor, opa/comparator, etc. logic explanations in it. Also there are some simple example circuits to build for each type.

If you have an extra comparator and ldr (or photodiode) laying around then breadboard the Dark Sensor circuit, less the transistors and relay stuff. Have the comparator output light an led for confirmation of the dark state. Now you are ready to OR the output of the Dark sensor to say the East (U2A/D or U2B/C?) output. Please refer to the LM339 data sheet for OR'ing outputs, it is very simple and clear.

I am thinking that the East output and the output of the Dark Sensor should have 1n4148 diodes so as to not interfere with each other, comments?. Also, how about putting NC limit switches "inline" with the circuits output to the Nfets, as opposed to grounding the outputs? Guessing it depends on best practice, or which type of switches you have (i.e. NC or NO).

Dragon Tamer

I was going to say that you could just use a 555 that is activated by the dark.

Whatever works, and defer to others as to the best path. For the moment, I am following a parallel but slightly different approach.

 

[rs14smith]

I need to look back for Mneary's dark sensor, missed that.

Get a book like the Forrest Mims mini notebook series, this one has basic ttl, cmos, transistor, opa/comparator, etc. logic explanations in it. Also there are some simple example circuits to build for each type.

If you have an extra comparator and ldr (or photodiode) laying around then breadboard the Dark Sensor circuit, less the transistors and relay stuff. Have the comparator output light an led for confirmation of the dark state. Now you are ready to OR the output of the Dark sensor to say the East (U2A/D or U2B/C?) output. Please refer to the LM339 data sheet for OR'ing outputs, it is very simple and clear.

I am thinking that the East output and the output of the Dark Sensor should have 1n4148 diodes so as to not interfere with each other, comments?. Also, how about putting NC limit switches "inline" with the circuits output to the Nfets, as opposed to grounding the outputs? Guessing it depends on best practice, or which type of switches you have (i.e. NC or NO).

Whatever works, and defer to others as to the best path. For the moment, I am following a parallel but slightly different approach.

There seems to be many ways for accomplish this task (detecting night), however, considering I'm about to learn micro-processors, I may leave this current circuit open to those who wish to edit it from here and try to accomplish this.

From my understanding, once you know how to code the micro-processor, we can basically program the solar tracker what to do in different conditions rather than how to use electric components to do that job (limit switches, etc..).

At the moment the tracker does at least turn back and forth following the sun, so it's a decent start for someone else I'd say

Using a Micro-processor seems like it will simplify the circuit a little, as from what I see, a lot of ideas suggested (to make it detect night) here are combining almost another full circuit to the current design, which will make this one complex circuit

 

[marktinkerman]

I been looking at arduino clones at about £17 and from chatting with a guy at Maplins (UK electronics shop) it seems one unit can control several arrays (which is great for me) and once youv'e learnt the language you just tell it what to do. He says the chip on the arduino is only about a fiver but for me having it already assembled and guaranteed seems worth the extra bit. Be nice if you could post your latest circuit diagram. mark

 

[nickelflippr]

There seems to be many ways for accomplish this task (detecting night), however, considering I'm about to learn micro-processors, I may leave this current circuit open to those who wish to edit it from here and try to accomplish this.

From my understanding, once you know how to code the micro-processor, we can basically program the solar tracker what to do in different conditions rather than how to use electric components to do that job (limit switches, etc..).

At the moment the tracker does at least turn back and forth following the sun, so it's a decent start for someone else I'd say

Using a Micro-processor seems like it will simplify the circuit a little, as from what I see, a lot of ideas suggested (to make it detect night) here are combining almost another full circuit to the current design, which will make this one complex circuit

Well at least you got your h-bridge squared away, and the light sensors mostly figured out. Mneary has had some great suggestions, deserving a gold star. Curious to know if the photo diodes get saturated in full sun?(i.e. requiring a filter). Have got some on order, so will find out soon enough.

As far as the limit switches go, won't be getting rid of them any time soon, unless binding linkages or jamming acme screw nuts sound like fun. Linear actuators may not be affected because AFAIK the limit switches come built in?

It's possible that the dark sensor circuit could be pulled from an existing photodiode. So, just another resistor divider and comparator, haven't tried it that way. For my relay version, servicing a two axis tracker will take 6 comparators ((1) LM239 and (1)LM293), six signal transistors, light sensors, and some passives, not so complex by itself.

Using a microcontroller will be a two edge sword. Passives are pretty cheap. If you ditch the light sensor say, then add an accelerometer, electronic compass, and solar declination tables to the mix. Not cheap, nor especially simple, still would be a fun project tho.

BTW - couldn't belieive the price on these spdt 12v coil 10A realays, so ordered some.

 

[pathos89]

Comparator

Hey this is a really cool project! i'm trying to recreate what you have here and i was hoping somebody could help me out with a couple things.
First, does the comparator need to be LM339D, I am currently using LM339AN and hope that works...
and second, is that a schottky diode? i don't see it in your circuit and it seems they only really sell surface mount Schottky/blocking diodes.
This forum is awesome i've been reading it and learning a lot. hopefully my questions aren't redundant.
thanks, pat