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

[rs14smith]

Hey guys/ladies,

Below is a short clip I made of the issue I'm having, and what I'm trying to accomplish.

Short clip: 2010-02-07_2356

After watching the clip, this is what I ended up changing to the circuit but ran into issues with the MOSFETs getting extremly hot, but note, the circuit did exactly what it was suppose to, moved to the left when it's suppose to, moved to the right when it's suppose to. Some have told me that circuit I did is wrong, which it probably is, but again, it works exactly like you would expect it to.

My version: https://i45.tinypic.com/fyj0nc.jpg

Thanks

 

[birdman0_o]

Mosfets need about 10-12V to be fully on/open, it seems your only feeding them 7, which might be a reason as to why they are getting hot

 

[rs14smith]

Mosfets need about 10-12V to be fully on/open, it seems your only feeding them 7, which might be a reason as to why they are getting hot

Well if I feed them more than 7v my FETs start to fry (they output like 5A), and are no longer useable. Again, that's in the circuit I built, but the one he gave me just simply doesn't work the way it should. At least mine does work right, but my FETs end up pushing out so much current they almost melt my breadboard lol no joke. My little hobby motor doesn't need that much current to run, but the voltage is also decreased from the FETs so I'm not sure what to do, other than try buying bigger FETs.

My FETs are N-channel by the way if that helps.

Update using 12V, I get the same results using his circuit: https://screencast.com/t/MjkxY2Nk

 

[rs14smith]

I'm thinking about buying a larger MOSFET for my second diagram in the picture, hopefully that works.

 

[ProFPGA]

I have successfully made a solar tracker using PIC16F84A and H-bridge driver ( which has a range of driving motors from 12V to 52V ).
For sensors i have used LDR and they point away at 45deg a part so whenever Sun moves away the tracker moves the motor to make the panels face the Sun.
You can have a look at the schematics in my Guru acct : Leaving Facebook... | Facebook

or this link : **broken link removed**
look into work samples .

 

[rs14smith]

I have successfully made a solar tracker using PIC16F84A and H-bridge driver ( which has a range of driving motors from 12V to 52V ).
For sensors i have used LDR and they point away at 45deg a part so whenever Sun moves away the tracker moves the motor to make the panels face the Sun.
You can have a look at the schematics in my Guru acct : Leaving Facebook... | Facebook

Great, I knew someone here had to have already accomplished this. The link you gave, I'm having trouble finding the schematic, can you tell me what I have to click to get to it, and is it free to look at or..?

Thanks

 

[mneary]

Your LEDs from the op amp to ground are limiting the MOSFET gates to about 2V.

The top MOSFETs Q1 and Q2 need at least 7V above the desired source potential. Since you want them at 12V (source follower) you need to supply 19V. However Q3 and Q4 should not be driven with more than 10V. (You need separate drive circuits).

Your circuit has no "off" state. This will cause the motor to continually oscillate and waste power. Worse, the 2 sides of the H-bridge has no protection from being driven momentarily at the same time while they switch direction.

 

[Boncuk]

Your circuit has no "off" state. This will cause the motor to continually oscillate and waste power. Worse, the 2 sides of the H-bridge has no protection from being driven momentarily at the same time while they switch direction.

... which I think is the main reason for the transistors getting hot. (current flow through two opposite transistors)

This circuit doesn't use an H-bridge, but relays.

It has an advantage over an H-bridge though. Even two pulled in relays won't do any harm to the circuit. The only thing which can happen is the motor receives positive voltage on both terminals if they are activated.

Boncuk

 

[ProFPGA]

hi

here is the link again
or this link : **broken link removed**
click to tab "Work Samples" .
then documents ,
then downlaod

 

[rs14smith]

I think what I'm going to do instead is just start from the ground up like many here have recommended until I have a fully working full bridge (H-Bridge) using MOSFETs and try to learn that way.

I would like to make a solar tracker for one of my school projects so I'd need to know exactly what is going on when I explain it to others, and it's hard to do that following other people's schematics. Thanks anyway ProFPGA. Some of the components you used in yours I have never seen in my life lol, so I rather try to diagnose the current one I have which I somewhat understand and try to get it to work properly.

I'll be posting my progress pretty soon to see if I'm on the right track creating the H-Bridge.

 

[ProFPGA]

hi , expect for the H-bridge driver i am fimilar with the rest of the components.
I will also post on the same acct how i am going to install solar panels on a moving frame .
i am having problems with fabricating the gears assembly that will move panels .

 

[rs14smith]

hi , expect for the H-bridge driver i am fimilar with the rest of the components.
I will also post on the same acct how i am going to install solar panels on a moving frame .
i am having problems with fabricating the gears assembly that will move panels .

I plan on using an actuators for my motor as it seems easier to work with rather than using the hobby motor type motors. The only thing is, actuators tend to fairly expensive even on ebay.

I just need to try to get this circuit down and I'll be good to go. I'm thinking I'm just going to post my progress here step by step and just get feedback on what to do next, as I really want to understand everything that is going on in the circuit.

 

[rs14smith]

Okay here is my updated circuit, which is a little cleaner to look at that my previous one.

I'm still trying to learn the proper way to build a H-Bridge, but I think for the most part, the way I have it setup in the picture attached is correct?

If so, I still think I need to add something to boost the voltage at the gate so the mosfets fully open when the OP Amp outputs a voltage at either 1 or 7? I know I saw somewhere people saying it can cause an issue if you don't fully open the gates.

 

[Ubergeek63]

Okay here is my updated circuit, which is a little cleaner to look at that my previous one.

I'm still trying to learn the proper way to build a H-Bridge, but I think for the most part, the way I have it setup in the picture attached is correct?

If so, I still think I need to add something to boost the voltage at the gate so the mosfets fully open when the OP Amp outputs a voltage at either 1 or 7? I know I saw somewhere people saying it can cause an issue if you don't fully open the gates.

that should work, though you should not be using crap opamps to drive fet gates.

the first circuit was broken on many levels... the opamps were only guaranteed to put out 5V and the 520s needed 7V, and the 520's 0.3 ohm resistance made for 7W dissipation at 5A under the best of conditions.

note: I have only read the first and last post here...

dan

 

[mneary]

The top MOSFETs Q1 and Q2 need at least 7V above the desired source potential. Since you want them at 12V (source follower) you need to supply 19V. However Q3 and Q4 should not be driven with more than 10V. (You need separate drive circuits).

 

[rs14smith]

The top MOSFETs Q1 and Q2 need at least 7V above the desired source potential. Since you want them at 12V (source follower) you need to supply 19V. However Q3 and Q4 should not be driven with more than 10V. (You need separate drive circuits).

Okay, so I basically need two different power sources, or be clever and find a way to increase the voltage to Q1 and Q2 and limit the voltage on Q3 and Q4. I'm not that clever yet lol, but I will probably take it to my school lab and manually supply them with two different voltages?

Can you recommend a way I can achieve this without doing it that way, like I believe a voltage regulator may be needed to limit the voltage to 10v, but not sure how to boost the voltage (capacitor maybe?)? Other than that mneary, should the circuit work properly once I figure out how to supply 7+ volts , and limit the other FETs to 10v?

Ubergeek, regarding your suggestion, can you recommend a better opamp as some other guy used the one I have now in his circuit which worked, so I figured I'd do the same, but wasn't sure how good or bad it was.

 

[rs14smith]

Here is my updated schematic with the 19 voltage where I think it should go, but I'm not sure how to limit the voltage to 10 volts yet, still trying to research that part, as there has to be a simple electric component I can use to accomplish that?

 

[mneary]

I'm sorry that I was unclear. The Q1 and Q2 gates need to be driven at least 7V (preferably 10) above the desired source potential.

 

[rs14smith]

I'm sorry that I was unclear. The Q1 and Q2 gates need to be driven at least 7V (preferably 10) above the desired source potential.

Nah you was clear, as for example if my main source of power is 5 volts, I would need to supply 12-15volts to Q1 and Q2. The thing is, I'm not sure how to hook up or create that needed voltage, nor do I fully understand yet how to limit the voltage for Q3 and Q4.

When you have time, could you possibly draw up a basic schematic of what that would look like , or how it should be hooked up? If not I'll try to keep playing around with it.

Thanks

 

[mneary]

There are a number of ways to make a high side driver that's required for Q1 and Q3, but generally they are either an IC or a number of discretes.
Here's an example:
I found it at: **broken link removed**. They don't tell you how to get VGH, which should be 7-12 volts above your top power rail, or in your case it would be 19-24 volts. It needs to have a generous peak current capability (0.1uF or more) but only needs to provide a few mA continuous. If you change R7, R10, and R11 to be 33k, it should still be fast enough for your circuit since you aren't changing directions frequently or rapidly.

[edit]The gate voltage should be independently, you can't hook two gates together.[/]

I wish I could offer a simpler solution, but this is why for one-offs I use the slightly more expensive P-channel MOSFETs instead.

[edit]Here's a thread with some simpler solutions. Post #10 from audioguru looks good. Note that it will only drive the MOSFET as long as the charge remains on the capacitor. This is a factor of the leakage in the diode, transistors, and of course the capacitor.[\]

In your application, I would seriously consider small relays.