• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Wind power charge controller problem

Thread starter #1
OK, so I decided on a simple charge controller circuit for my wind generator. I went with the one on this site; wind charge controller . It is the top most post for the charge controller by Thomas Miller. Anyways, I decided to recreate the circuit board shown in Eagle editor since there was no board layout image for download. I put the board together and fired it up with my bench variable supply. After noticing that I had missed a solder bridge and cooked at least the 4001 logic chip I decided to install IC sockets for better testing. I decided to test each part of the circuit separately.

The first part of the circuit uses a common op amp voltage window detector circuit giving you a high and low trip point with an in-between voltage window. So to start the test, I set the supply at 12.5 volts and set the voltages at the test points to 7.25 for A and 5.9 for B. This should make the window between 11.8 and 14.5 volts. So I then set the supply for 10 volts and installed the LM339 leaving the 4001 out to check the comparator outputs. at 10 volts, the output at pin 1 was pulled low since it was below the low trip point, and pin 2 was high which it should have been. As I slowly ramped up the voltage and got past the 11.8 volt mark, both pins 1 and 2 read high which was good. I continued to ramp the voltage past the 14.5 volt mark (~16 volts) but pin 2 remained high and was never pulled low by the comparator. I then double and triple checked the voltages at the touch points and made sure that the voltages were also seen at the correct pins on the IC socket. I also verified the input voltage at the IC at ~8.06 volts which seemed to me to be in tolerance. Once all my voltages were verified, I had thought that I had also cooked the LM339. I had a couple new ones in the parts box and installed a new one. Low and behold, I got the same results, pin 1 was low below 11.8V and pin 2 never pulled low.

I figured that I would do some further research and come back to the comparator side of the circuit later and see if the RS flip flop part of the circuit was working. I did not have any more 4001 2 input NOR gate chips, but had a 4011 2 input NAND gate. Knowing that these are pinned the same I figured that they would work in a similar fashion when configured as an RS flip flop. I removed the LM339 and installed the 4011. When powered on with 12.5 volts, I had the red light on. I then used a jumper wire to manually pull pins 1 and 6 low alternately. As I did this, I was flipping between red and green lights telling me that the flip flop was working.

Now I am here. I am still left with the problem with the window detector circuit that is not working. Does anyone have any ideas as to what would cause pin 2 of the op amp chip to never get pulled low? I have triple checked the traces on the circuit board along with all the voltages at each pin. I have swapped in 3 other LM339 chips with the exact same results from each. I am at a loss....


Well-Known Member
Most Helpful Member
Its difficult to suggest what the cause of the problem may be without a circuit diagram.
Can you post a diagram or a link to the original.??


Well-Known Member
Most Helpful Member
In the first example: Vin10 volts, Output should be plus. This comparator inverts. It almost sounds like you have A and B reversed either in your mind or in the circuit. I hate window comparators- they play with my head. :)
An aside -there is a board image just download it and use a perspective changer say with Gimp to get it flat
Compare it with yours .Sounds like you may have a track wrong

I did it for you


Last edited:
Thread starter #7
Here is a side by side comparison of the original VS my copy created in Eagle layout editor. When I created it, I first drew out the schematic in Eagle and then placed the components on the board view as close to exactly the same as the original. I did a side by side comparison before I etched the board.

I had originally tried to use the image that you (tytower) have shown, but to get the pin spacing exact was a pain. Not to mention the original image was quite blurry when blown up, so to try and get traces that didn't overlap or other issues was going to be a pain. This is why I decided to use Eagle to make as close to an exact copy of the original as possible. I manually routed the traces in Eagle instead of using the auto-router so I could trace them as seen in the original. The only thing I missed in mine that the original has was pads for the touch points A and B, but I was able to add those to my board without too much trouble using some 12 gauge wire.

This gets me back to the original post of my problem. Ronv, you mention that with Vin at 10 volts, the output should be plus. Which output are you referring to? The output from U2-A or U2-B? Each comparator acts differently which gives you your set and reset logic going into the flip flop.

U2-A should go low or negative when the input voltage goes above14.5 volts (or double the voltage at TP-A). This should trip the relay to then dump the incoming power form the generator to the dump load. Conversely, U2-B should go low when the input voltage drops below 11.8 volts (or double the voltage at TP-B). This should then trip the relay back to dumping the voltage into the batteries. When in-between 11.8V and 14.5V, the flip flop should be at it's last switched state. That is the reason for the two buttons. This allows you to manually switch between charging and dumping when the circuit is in it's window state.

I feel that I have a pretty good grasp of how the circuit should function and the idea behind the use of the window comparator circuit used which is why I am baffled as to why it is not working. If my explanation of how the circuit is supposed to work is in any way wrong, please correct my understanding.

Thanks for the posts given already and in advance of posts that may help me figure this out.


Thread starter #9
Your simulation looks correct. In your sim, your 2A would be the U2-A output and your 1B would be the U2-B output. The graph is showing 2 windows of the circuit. The windows are when both outputs are low at the same time with the first being just below 1.0ms as the voltage (Vbat) ramps up, and the second being just above 2.0ms as the voltage (Vbat) ramps down. The problem I have with my set up is that U2-A is high (8V) at all times, where according to your simulation it should be low below 14.5 volts. My U2-B is fine and will stay high at anything below 11.8V and low at anything above that.
Your lines under the 4001 are very close
Thread starter #11
I scaled down the image a bit before posting it, but I did do a continuity check on all traces on the board and made sure that I do not have any cross connects or solder bridges. I inspected the board also under my magnifier and all looks good.

When looking at the two board layouts side by side they look correct to me, but when you stare at something for too long you can miss things, so having another set of eyes check it out might help. Do the board traces look correct?


Well-Known Member
Most Helpful Member
If that's the case then U2-2 must be shorted to +8 volts. You said you measured pin 4 to set the threshold and we know pin 5 is okay because the other comparator uses it and is OK. Check for shorts around that trace or under the pull up resistor.
Might be a good idea to check the input voltages right at the IC just to make sure.
Last edited:
Thread starter #13
I FINALLY got it working. I did a bit of testing with my breadboard and figured out that if I reversed the trip voltages, then it worked. By this I mean that the circuit diagram and article tells you to set the voltage at TP-A to 7.4V and TP-B to 5.7V. If I set TP-A to 5.7V and TP-B to 7.4V and ramp my voltage from 10 to 16, the comparators go low at there set levels. I then installed the 4011 to activate the flip flop and connected my LEDs and ran a test. It is working like a charm.

I want to extend my thanks to all who gave input on this. It had been baffling me for some time.

Latest threads

EE World Online Articles