Electrical circuit design help

[SJP0tato]

Hey guys,
I recently purchased a few of these:
http://www.amazon.com/Solar-Car-Smal...ords=solar+car

It's a simple circuit where the solar panel is connected directly to the motor. In the default setup, the solar panel only provides enough power to run the motor under direct sunlight.

My question is:
Is there a way of adding components (maybe a capacitor) to allow the solar panel to store energy in the component for a time, until it reaches an arbitrary power level and then supplies the motor with enough power to run for a short time? The behavior of the car would be to sit and charge for 10-15 seconds, and run for maybe 3-4 seconds at a time.

My reason for doing so would be to allow the car to work under much weaker lighting conditions.

Is something like this even possible? I have a decent amount of experience building already designed circuits, but for direct applications like this I'm in a little over my head.

Thanks for any help/suggestions!

 

[wirednuts]

adding a capacitor is over your head? i think not. they make some really small high capacity ones now too. just hook it up directly, no need for a cut-in/cut-out circuit

 

[SJP0tato]

Will that give the behavior I'm looking for? From some examples I've seen online it sounded like the power source would need to be removed before the capacitor would discharge.

Based on this: http://farside.ph.utexas.edu/teaching/316/lectures/node60.html
It sounds like the solar panel would charge the capacitor until it reaches saturation, then nothing would happen?

It would be nice if it's as simple as just adding the capacitor.

adding a capacitor is over your head? i think not. they make some really small high capacity ones now too. just hook it up directly, no need for a cut-in/cut-out circuit

 

[wirednuts]

its is that simple. no different then hand crank flashlights.

the sun will charge the capacitor and run the motor at the same time. cover up the power source (sun) and the capacitor will just feed whatever power it has.

there is no extra circuitry needed. maybe a tiny power switch on the motor so you can let the car charge without rolling around. im not sure about the capacitor only discharging once its charged, they dont work that way. they work just like batteries, the only difference being they can dump all of their load instantly.

 

[SJP0tato]

My trouble is the solar panel doesn't provide enough power to run the motor indoors though. The ideal design would store the trickle charge, then discharge the capacitor to run the motor once the capacitor was fully charged.
The switch is an interesting idea, one way would charge the capacitor, the other would connect the capacitor to the motor maybe.

It doesn't seem like there's any automated way to have the capacitor "know" to discharge once it's fully charged.

 

[wirednuts]

you could make a voltage sensor circuit, i just dont think it will be small enough.

 

[Jeff7]

The amount of light available indoors is actually substantially less than what's available from sunlight. Your eyes just mess with your head by dilating your pupils when you're indoors.
Solar cells aren't terribly efficient, and their output is usually rated for use under normal sunlight conditions. That kind of solar cell won't put out enough voltage under indoor lighting to run the motor, as you've discovered - unless you put it directly under a very bright light.

You can get some pretty beefy supercapacitors for a few dollars.
- They need to be able to handle whatever the maximum voltage of the solar cell is.
- The uF value to get will depend on the motor's current consumption. If you get a big cap, that'll allow more runtime, but it'll also take longer to charge up.

Also know, if you let a supercap charge up fully, and you then accidentally short out the terminals, they can dump a lot of power in a big hurry, which could be a problem for anything flammable nearby, or for the capacitor's guts.

Oh, and a group of 3 82F capacitors is fun to play with.

My trouble is the solar panel doesn't provide enough power to run the motor indoors though. The ideal design would store the trickle charge, then discharge the capacitor to run the motor once the capacitor was fully charged.
The switch is an interesting idea, one way would charge the capacitor, the other would connect the capacitor to the motor maybe.

It doesn't seem like there's any automated way to have the capacitor "know" to discharge once it's fully charged.

I do everything at work using PIC chips ("cheating"), but you might be able to do something with a voltage reference, a comparator, and a MOSFET. Maybe.
That would add a small amount of extra current draw to the circuit, and those chips would need to be able to run at whatever voltage is present in the car's guts.
- Allow the solar panel to power the comparator circuitry and charge the capacitor.
- Once the capacitor reaches the proper voltage level to indicate "charged," the FET is turned on, connecting the capacitor (and solar cell) to the motor.

I guess you'd want it to latch then, too, otherwise it will turn itself back off once the capacitor discharges a bit. :hmm:

 

[Mark R]

There are power status control chips that can be bought off-the-shelf. They monitor power voltage, and switch a load on-and-off when the voltage goes in and out of spec.

These often have programmable ranges, and might work. Have a look at the catalogs for various chip manufacturers.

 

[wirednuts]

you guys realize these cars are the size of a dime?

 

[Mark R]

That's alright, an ADM6316 power supervisor is the size of a grain of rice, as would be the MOSFET needed to get it to control the motor.

The Capacitor might be more of an issue. Soldering too.

 

[SJP0tato]

Thanks guys, this gives me a lot of info to chew on. I'll report back with how it turns out once I get a solution worked up.

 

[wirednuts]

Soldering too.

just wanted to repost that

 

[ussfletcher]

I doubt that the solar panel could supply sufficient voltage to build enough charge on the capacitor, making a single capacitor system useless.
Total charge is given by Q=CV where C is the capacitance and V is the voltage. Thus, we see that total charge doesn't build over time to a value greater than Q. This translates into current via I=Q/t, thus you could probably run the motor for a very, very brief time (fractions of a second). (Or you would need a giant capacitor..)

 

[mindless1]

Put a larger solar panel on the car, and if it isn't going to be played with by children, get some panel weight reduction by it not being encased in epoxy resin.

For that matter you could just build your own but do it with a realistic looking model car, put flexible solar panel skins all over it, and Bob's yer uncle.

 

[harrkev]

First, a disclaimer: I am not an analog guy, I make my living doing digital stuff. But, I do have a MSEE, so I know a little more than the average person.

A capacitor by itself will NOT help you in the least. Adding a capacitor MAY be useful in full daylight to help your car cross a small shadow (but I doubt that would even help you do that).

The problem is that the motor is still connected in the circuit... The motor sucks the energy away as fast as the solar cell makes it -- there is no energy left for the capacitor to hold. Imagine that you have a 50W laptop, and you are trying to run it AND charge the battery with a 30W charger -- it just won't work.



What you REALLY want to do is to disconnect the motor from the circuit (using a transistor), and then allow the solar cell to charge up. Once the voltage on the cap reaches a certain level, then you attach the motor. It will move a little, and the stop and charge up the cap again. This is like turning your 50W laptop off overnight, and letting the 30W charger just charge the battery overnight.

Like I said, this is an analog-ish problem, and I am a digital guy. I could imagine trying to crack this nut with a couple of transistors, a zener diode, a handfull of resistors, and (of course) a gigantic capacitor and a solar cell. Another approach could involve a zener diode and a comparitor IC (with some positive feedback to cause a little hysteresis). I would have to do a lot of trial-and-error to get it right. Maybe a more focused search on Google or Bing would help.

 

[MrDudeMan]

A capacitor by itself will NOT help you in the least. Adding a capacitor MAY be useful in full daylight to help your car cross a small shadow (but I doubt that would even help you do that).

The problem is that the motor is still connected in the circuit... The motor sucks the energy away as fast as the solar cell makes it -- there is no energy left for the capacitor to hold. Imagine that you have a 50W laptop, and you are trying to run it AND charge the battery with a 30W charger -- it just won't work.

This is only true if the motor is drawing more current than the solar cell can provide. There are many conditions where that won't be true. The analogy also breaks down for laptops because a 30W charger is sufficient to charge and power the computer in many conditions. My laptop is capable of drawing 120W, but it's idling at 10W.

With that said, the next part is mostly correct.

What you REALLY want to do is to disconnect the motor from the circuit (using a transistor), and then allow the solar cell to charge up. Once the voltage on the cap reaches a certain level, then you attach the motor.

 

[harrkev]

This is only true if the motor is drawing more current than the solar cell can provide.

Yeah. That was kind of the assumption. If the cell can provide enough current to run the motor, then all you need is two wires...

 

[Abwx]

This can be solved using a diode in serial to charge a few AAA Nickel Metal cells , the number of wich is to be determined by measuring the solar panel output voltage, and using a number of accumulators such that their nominal serial voltage is a little less than the one of the solar panel - schottky diode losses.

 

[harrkev]

This can be solved using a diode in serial to charge a few AAA Nickel Metal cells , the number of wich is to be determined by measuring the solar panel output voltage, and using a number of accumulators such that their nominal serial voltage is a little less than the one of the solar panel - schottky diode losses.

Nope. Sorry.
The problem is that you have to understand what a motor actually is -- a long wire (really, an inductor). If the solar cells are in insufficient lighting (say, an overcast day), then they will output a small amount of current. That current will go straight through the motor. The motor WILL consume a certain amount of current -- generating a magnetic field. Now, if the strength of the field is not enough to begin to turn the motor, then the motor will still suck all the energy that it can -- even without turning.

The real trick here is to design a circuit that will disconnect the motor if there is not enough power to make it turn. Once the storage unit (whether a battery or capacitor) has enough energy, then you connect the motor and allow it to run until the energy is low enough that the motor cannot turn again.

 

[mindless1]

Another vote for larger solar panel... nothing else makes any sense. No circuit will be worth it, that drains power itself and then the drop across the switching subcircuit.

Edit: Text removed because I was being an ass. Sorry.

 

[stormkroe]

First, a disclaimer: I am not an analog guy, I make my living doing digital stuff. But, I do have a MSEE, so I know a little more than the average person.

A capacitor by itself will NOT help you in the least. Adding a capacitor MAY be useful in full daylight to help your car cross a small shadow (but I doubt that would even help you do that).

The problem is that the motor is still connected in the circuit... The motor sucks the energy away as fast as the solar cell makes it -- there is no energy left for the capacitor to hold. Imagine that you have a 50W laptop, and you are trying to run it AND charge the battery with a 30W charger -- it just won't work.



What you REALLY want to do is to disconnect the motor from the circuit (using a transistor), and then allow the solar cell to charge up. Once the voltage on the cap reaches a certain level, then you attach the motor. It will move a little, and the stop and charge up the cap again. This is like turning your 50W laptop off overnight, and letting the 30W charger just charge the battery overnight.

Like I said, this is an analog-ish problem, and I am a digital guy. I could imagine trying to crack this nut with a couple of transistors, a zener diode, a handfull of resistors, and (of course) a gigantic capacitor and a solar cell. Another approach could involve a zener diode and a comparitor IC (with some positive feedback to cause a little hysteresis). I would have to do a lot of trial-and-error to get it right. Maybe a more focused search on Google or Bing would help.

QFT
I was going to suggest a transistor or scr. Same idea as a battery charger, except instead of disconnecting current to the battery cell, it's shunted to the motor. The motor runs, the panel keeps charging, the car stops when it's virtually depleted, and the power is switched back to the capacitor/battery until it's full again.

 

[Colt45]

You will need a comparator, with fat hysteresis on it, which switches the motor on when voltage at the cap exceeds a threshold. It then turns on the motor, and doesn't turn it off until the voltage drops a fair bit, as the cap is discharging (this is where the hysteresis is essential).

It then repeats ad nauseam.

But, like another poster said - eyes have a log response. There is /way/ less light indoors, even though it only feels somewhat darker - it's orders of magnitude less light... so the thing might have to 'charge' for quite some time, if the open circuit voltage is even high enough under indoor lighting (as this is what the cap will eventually charge to).

You'll have to account for the current the comparator uses also, which can be pretty miniscule, but exists.

That's the simplest way I can think of doing it, at least.


I've got a cheap panel here - it claims 5.5V 100mA. With the panel 3" away from a 60W bulb, it barely lights a standard red led - so maybe 2V at 10mA - 20mW, and the nameplate power, 550mW, should be attainable in direct sunlight. big difference.

 

[sm625]

You would need to add more components than just a capacitor. If all you had was a capacitor then the cap would immediately begin discharging as soon as you took it out of the sunlight, assuming you can even get the cap to take a charge. I'm thinking it wont even begin to charge since it is in parallel with a low impedence motor. But even if it did charge, by the time you got to the dining room floor it would already be dead, thanks again to the low impedence load of the motor. So you'd need a resistor to slow the discharge. And you'd need some sort of solid state relay (SSR) and a comparator to drive it.

I would say you'd need a diode in parallel with the resistor to allow it to charge fast, but that is irrelevent here because you're probably dealing in millivolts. That tiny solar cell probably never even goes over .7V. Does anyone know?

The added weight of the capacitor and resistor alone will be enough to slow the car down by a noticeable amount.

The most practical thing to do is simply install a small battery. lol. Otherwise you'd need a whole host of additional components to power the aforementioned SSR and comparator.

As a side note, this is an amazingly useful little device. It is essentially a timer (daylight detector) and actuator all in one. Such a thing, in this size, would have costed thousands of dollars 20-30 years ago. It sort of reminds me of those throwaway motion sensing toys they put in kids meals from burger king. Good stuff.

 

[Abwx]

The real trick here is to design a circuit that will disconnect the motor if there is not enough power to make it turn. Once the storage unit (whether a battery or capacitor) has enough energy, then you connect the motor and allow it to run until the energy is low enough that the motor cannot turn again.

A simple off/on switch in serial with the motor will allow
the cells to charge the batteries when the thing is not used...

 

[Modelworks]

Take apart some of the solar powered lights that people use around sidewalks. They use a NiCd cell and a 4 lead part, same size as a to-92 transistor that controls the charging and the on time. Basically if it detects current flowing on the solar cell it goes into charge mode, remove the light and the current stops signaling the chip to switch to battery power.
It is far from a perfect design but it works pretty well for what is there..