Solar hot water heater circuit

[frazelle09]

We are trying to finish building a solar hot water heater. This one is based on the concave mirror design and is 10 ft. long and 40" wide. We have achieved a 100 F degree temp. rise in 20 minutes. Since this collector needs to track the sun, we have received a circuit from Wayne in Canada. His website is greenwatts dot info

We have uploaded some photos of our work at our **broken link removed** website.

Our problem is that our mecatronic student/helper is not going to be able to continue to help us. We have the circuit and at this point it works on two 6volt battery packs and is controlled by a sun sensor. This circuit then controls the motor which moves the collector East to West as the day progresses.

The problem is that the motor we are using is a small 3.6 volt screwdriver motor which, if we connect it to the circuit and then to an AC 110 volt outlet (actually two of them - one for each direction) our circuit burns out. It burns out when we try to rotate the collector towards either extreme (pointed either straight East or straight West). While the collector is moving through the middle of the arc the electro-mechanical system doesn't have any problem.

We think we need a larger motor but we like this one very much because of the planetary gear system that it has. Since we don't have anyone who can help us fix our circuit board again if it goes out, we think that the first thing we need is to figure out some way of protecting our circuit board with some sort of fuse between the wall (AC 110v) and our circuit board.

With this same design we have already burnt out a circuit board as well as a previous motor we were going to use (a car electric window motor) so if we are going to buy some sort of "new" motor, we think we may also need a fuse between the circuit and the motor.

Any ideas? i'm going to upload a copy of Wayne's circuit as well. Any help would be appreciated. Maybe we are going about this the wrong way.

Have a great evening!

 

[peufeu]

You ask about your motor, let's look at that.

When you look at a motor, there are several specifications :

- mechanical power output : This is interesting only if you want to move a load in a specific amount of time (for instance, an elevator). In your case, the sun doesn't move so fast, so don't look at this spec.

- maximum torque : This is more interesting for you. You need a motor with enough torque to be able to rotate your assembly in all expected conditions. A gearbox (like you have) will reduce the output RMP and increase the output torque proportionately (it is, in fact, just like a mechanical lever...)

- and of course, working voltage, rpm, efficiency, etc.

So, you need to test if your motor can provide enough torque to rotate your solar collector, with a decent safety margin to account for wind, weight of rainwater in the collector, etc.

You probably don't have the gear to do precise measurements, just plug the motor with a suitable power supply and an ammeter, and run it through several complete cycles, maybe loading the collector a bit. Check that the current draw corresponds to the motor specs, and than the motor doesn't stall, doesn't drop its rpm too much, and doesn't sound like it's gonna die.

Anyway, maybe your electric screwdriver motor will work perfect, maybe not, no way to tell but to try !

You can also increase the mechanical advantage of the motor, I don't know how it's coupled to the collector, but you could add a pair of pulleys, or some salvaged bicycle gears.

You could also balance the load by adding weights, to counter the weight of the collector, and make it easier to rotate.

Now, let's suppose the motor is able to move the collector without problems. You still have problems at both ends : if the tracker tries to move the collector over the end of its allowed movement span, something will get blocked, and the motor will stall. It will also stall in normal use if it is not strong enough, as you seem to have experienced.

When this happens :

- the motor is not rotating, so the insides get no cooling airflow
- the motor stops being a motor, it becomes simply a resistor equal to the resistance of the copper wiring inside.

In this case, the current becomes very high, and the motor heats rapidly and usually overheats and melts, if the driver doesn't blow first.

So, you will need a way to stop the tracker from trying to go over the allowed limit of mechanical movement. For this, you can use mechanical or optical switches (end-of-travel switch I think its called) or simply install the light sensors in a way that, when the end of mechanical travel is reached, something casts a shadow on a sensor, and the tracker thinks it has to move the collector in the other direction.

You can do both, including mechanical switches as a safety, cutting the power if the motor goes over the limit (but in this case, it won't be able to go back, someone will need to push it).

You can stick a thermostat on the motor, or the driver transistor heat sinks, cutting the power if it overheats.

-----------

Your mechanical build looks nice, unfortunately the electronics are, quite frankly, crap.

> It burns out when we try to rotate the collector towards either extreme

Given this **broken link removed**, that is to be expected.

1- There is no overcurrent protection, so the motor and drivers will blow if the motor stalls / gets stuck.

2- There are no protection diodes, so the output transistors should also blow randomly.

3- There is no time constant, so it will follow the clouds

4- There is no protection of any kind, so if something (a fly, a dead leaf, etc) obscures a sensor, it will act random and probably blow motor and sensor

5- The design needs 2 power supplies where it would be pretty easy to use one

6- No end of travel switches

- And someone dares to sell THAT ?????

You can fix 1 and 2 above easily.
Install some free-wheeling diodes on your PCB (google it) that will cost you 1$.

Get some current-limited wall warts. After having measured the maximum motor current during normal conditions, get a switching wall wart (not the usual heavy transformer) with 50% more current specification. Those usually include an internal current limit.

Fuses are bad. You can use a Polyswitch (it is a kind of electronic device which acts like a fuse, when too much current goes through it, it heats, and reduces the current) those are slow, but don't need replacing, once the error condition stops, they cool down and go back to normal. Put it in series with the motor.

 

[frazelle09]

Dear peufeu:

Thank you very much for such a complete post and answer. i really appreciate it. i think that we can follow/try out most of your suggestions w/o any (or very little) electrical/electronic training - which is the situation in which we find ourselves. Needless to say we are very sad that you find Wayne's circuit a piece of crap. Since we ourselves don't know much about electronics we must rely on others.

Motor> i don't know if the Multimeter we have will measure amps, i think it does, but am not sure how to do this -- the motor only has two leads from the circuit. Is this is where we would measure amps? If this is not possible, we can use your other "indicators" to see if the motor is up to the task.

How is the motor connected/coupled to the collector? > We have a long, galvanized, threaded steel rod which runs through a screw connected to the collector. This makes the collector turn even slower (which is what we want).

The two power supplies question!!!??? i was also stumped by the fact that the design called for two power supplies. This doubles the investment that must be made in this aspect of the design and i couldn't for the life of me figure out why it was set up this way. i'm glad that i'm not the only one that saw this. Even my partner mentioned that he thought that it was pretty poor design. What can be done here?

The only other question i have is: Where would the Polyswitch go?

Thank you again for all your suggestions. We've been at this project since December and hope to have it working one of these days! Have a great morning!

 

[peufeu]

> i think that we can follow/try out most of your suggestions w/o any (or very little) electrical/electronic training

Yeah, that was the idea !

> i don't know if the Multimeter we have will measure amps, i think it does, but am not sure how to do this

OK, if you got 2 multimeters, wire them like this (on my doodle : wall wart, voltmeter, ammeter, motor) so you can read volts and amps. Change the motor polarity manually to make it go in the right direction.

Online Sketch: Untitled Sketch by Anonymous Author

If you only have one meter, you'll have to make 2 measurements.

Remember, current goes through ammeter so it can measure it. The ammeter is like a piece of wire (very small resistance actually) so don't plug across on your wall wart to check how many amps it can supply

Use the 10A ammeter setting, that should do the job, no need for great precision.

You need to know :

- the max current the motor needs (look at the amps) to size the power supply, driver, and protection
- the current with motor stalled (only for a few seconds)
- if your powersupply can supply it (watch if the volts stay constant)

Counter-intuitively, a wimpy power supply will burn your motor : not enough power, the motor stalls, then it burns. A powerful enough supply will allow the motor to not stall, unless of course it hits the end of travel limits, or there is something blocking the rotation, in which case the polyswitch should protect it.

The screw thing is good, should give more torque.

> The two power supplies question!!!???

Well, that's simple, it saves a couple transistors and resistors (about $1), plus some headache, to the guy who makes the kit.

> Where would the Polyswitch go?

In series with the motor, like a fuse. You could put it in the power supply (but you got 2 supplies), however if it trips, the tracker will lose part of its power supply, I wonder what it would do...

Once you measure your motor, we'll see about the rest...

 

[4pyros]

Thare are many solar trackers out thare and all kinds of info on the web like this one
Electronic Projects

The problem is that the motor we are using is a small 3.6 volt screwdriver motor which, if we connect it to the circuit and then to an AC 110 volt outlet (actually two of them - one for each direction) our circuit burns out.

Did you hook 110VAC up directly to the board? Thats bad!!! You need to step the voltage down and make it DC at the right voltage for the motor.

 

[gary350]

I have built a lot of solar project over the past 35 years. The type of system you are building it a little more trouble than it is worth I tried that too many years ago. Tracking the sun needs to be very accurate other wise the focal point misses the hot water pipe and heat is lost. It helps if the entire reflector and pipe is inclosed with 2 sheet if glass over the top to hold in the heat. Lots of problems with this system.

Flat plate solar collectors are very easy to deal with. They do not have to track the sun and they make heat on extremely over cast days even with the solar collector covered with 2" of snow. I used snap disk thermostates in my flat plate solar collectors they are very cheap $3 to $4 each and are rated 20 amps at 240 volts. You can buy snap disc thermostates at any hardware store. You can salvage several from the inside of a junk home type clothes dryer. **broken link removed**

I had a 70 degree F snap disc inside of my flat plate solar collector. When it warmed up to 70 degrees it turned on the water pump. When the sun goes down the water pump goes off. You need 2 sheets of glass on your flat plate solar collector it will collect 200 BTUs of heat per square foot in a well insulated box. A sliding glass patio door measures 26"x76" that equals 13.7 sq ft of surface area = 2744 BTUs of heat per door. 4 of these will give you 10977 BTUs.

You need 6 rows of copper tubing inside each flat plate solar collector running length wise.