Electronics question (PWM)

[Mutant Funk Drink]

Hey guys, I'm again trying to improve my bastard PWM circuit for the purpose of controlling a DC motor.

The circuit was originally an LED flasher, but I removed the LEDs and any unnecessary resistors. I think it does what I need, but my question is basically a stupid one. Though I need to ask this question just for reassurance.

When testing my circuit in the falstad circuit simulator, I'm getting square waves but only the mA completely gets close to zero. Voltage, while it has square waves, doesn't get anywhere near zero and in fact it is only reduced by about 0.4V.

Does this really matter? The point of the circuit, other than to reduce the speed of the motor, is to save power. Does it matter if the voltage isn't really effected by the circuit so long as I get decent alternation of milliamps? I would think it would be just fine since I think if there is no power at any given point then no energy is being lost?

[roasted]

Give us a little more info on the setup...what is your end goal/application?

Anyway, to answer your question: the voltage isnt that important. Current modulation (mA) is the main thing. If you are reducing the current to almost zero, then you will save power (of course power will still be lost in the process...).

Consider a taser....huge voltage (+1,000 volts), but quite low current (like 2-5 mA); compared to a power outlet....low voltage (~120), but high current (like +15 A).



TL;DR? Voltage isnt that important. In your setup, it probably wont matter that much

[Spatula Tzar]

It's hard to know what you're talking about without more details, but in general 0.4 V is well below the "low" threshold of most circuits. If it's a problem, you can easily adjust it by adding a pulldown resistor on the FET you're using to drive the motor.

Anything which reduces the average power to the motor will reduce the speed, and the, well, power. Whether or not the valleys go all the way to zero makes no difference. The only concern might be drawing some power when turned off, which is wasteful.

[Mutant Funk Drink]

It's meant to power an auger that will lift water for a hydroponics/aquaponics project. I wanted to go with a low-power application like that because an auger is more resistant to low-speeds, meaning I may be able to get away with using solar energy.

Yup, that's what I thought. It seems like the circuit pulses the current enough that my next phase will be to test it out in real life to see if it works better in comparison to no PWM.

EDIT: By 0.4V, I meant that as how much the voltage drops. The power source is 2.4V, so in the software, the negative part of the pulses only go as far down as 2.0V.

[Spatula Tzar]

A difference of 0.4 V probably won't do much. I suggest redesigning the circuit until it gives you something better. Posting a schematic will help.

Augers are rather inefficient pumps. You want one of the many rotary positive displacement pump designs. My favourite is the peristaltic pump, but others are probably better for plain water.

[vovka351]

Even at radioshit, the parts to build a circuit specialized for this purpose would cost you less than $10.

Use a 555 timer (it's pretty much the simplest source of a pwm square wave that goes down to 0V) and amplify the output through a MOSFET. The 555 timer consumes very little energy of it's own.

Ohm's law says a 20% reduction in voltage will cause a 20% reduction in current and thus a 36% reduction in power consumption. With solar, you probably want more than that.

[Mutant Funk Drink]

I wanted to try going without the 555 timer, both for the sake of simplicity and because I like trying to get away with the bare minimum.

While I think my circuit would probably work fine, it seems like the 555 would do a pretty good job because, unlike my circuit, the frequency doesn't seem to chance significantly with decreased voltage.

Here is the 555 timer circuit that I will probably go with:
http://tinyurl.com/7phoeax

And here is the circuit I initially made using two transistors:
http://tinyurl.com/6mh5m8h

Just so you know, the LED in both circuits is merely a stand-in for where the DC motor would be.

The problem I had with the 2-transistor circuit was that if the voltage is raised or drops a little bit, the frequency and the wavelength changes significantly in the simulation. If it goes below 1.5 volts, the waveform disappears entirely. This would be a big problem if I were to use a solar panel with rechargeable batteries.

However, I made a less efficient version of that circuit in real life, and it seems to still work even at a lower voltage. I don't know if the simulator has flaws or not.

But now looking at the 555, I have to admit that it seems pretty reliable and efficient.

EDIT: Neither circuit includes a pot. I may add one later on, especially since that would be easier to do with the 555.

By the way, the reason I am going with a small auger(3/4" diameter) is because my application is relatively small, and I want it to run off of some NiMH batteries. This is a really small-scale solar project, and I settled on an auger because the one I have put together has pretty decent flow for the low speed I give it. While it may not be the most efficient pump, it can operate with very little power and is resistant to jamming. I actually looked into using a peristaltic pump, and while I didn't experiment with one too much, it seemed like it wasn't feasible for my project. I want my project to consume as little power as possible, and while the peristaltic pump is cool, it seems to require more torque than the power can be provided on a regular basis.

[Spatula Tzar]

I can't get the schematics to load, but 555s are definitely the way to go. They're incredibly versatile.

If you're concerned about motor torque, there's a little trick to increase it. Drive the motor with a higher than rated voltage, and use PWM to reduce the average power. Small brushed DC motors can easily take more than their rated voltages in the short term. You could even water cool it if you're concerned.

Also, don't forget about gearing it down. They're usually too fast to use directly.

[Mutant Funk Drink]

The schamatics load through the falstad java applet, so if you don't have java, that might be why it's not running. Otherwise, my 555 circuit seems pretty similar to all the other ones on the net.

The motor that I'm using is rated at 0.5V - 6V(or something like that) and it's lower RPM but higher torque. If I were to try doing a peristaltic pump again, I'll have to try that trick and see if it makes a difference(if not that motor than with another one).