# Pulsing Current w/ PWM / Direction

#### kcarring

Joined Jan 22, 2011
38
I still get confused with conventional current flow theory and electron direction, etc.. but..

In a circuit like this one, where you use a 555 timer to then pulse a mosfet and ultimately limit power to a motor...

http://www.dprg.org/tutorials/2005-11a/index.html

The motor's - lead is connected to the output of the Mosfet.

If you were to put "any" circuit into that schematic, in other words, not a motor... rather limit by way of PWM/MOSFET - the ability for a circuit to have current/power...

Would you attach to the negative side, like that?

It threw me for a loop, I assumed you would be attached to the + side of the motor, or any circuit...

If instead of the motor for example I had a whole seperate circuit, maybe a blocking oscillator or something... could I limit the available RMS voltage (and thus available current, effectively) by supplying it with the MOSFETs output tied to it's - side?

Thanks

#### Potato Pudding

Joined Jun 11, 2010
688
Most people would add a snubber capacitor and snubber diode(reverse biased to source voltage) across that motor because a running motor that is switched off becomes a generator and that can create a high counter EMF voltage until the motor finally stops unless you provide a current path.

There is no direction control in that circuit. It is only useful for speed control. You don't have the ability to reverse the motor.

You are only controlling the average applied voltage so current will be very load specific.

For use with other loads than a motor.

Many circuits will want a regulated voltage, so you would need to filter the output. That filter would be load specific and only for a fixed resistive load unless you added the proper feedback control. A blocking oscillator of high enough frequency might appear as the resistive load, but one of low frequency will definitely load and unload the voltage.

What you are really suggesting without knowing it is using the circuit as a type of switched mode power supply. It can be done but there would need to be more questions and explaining and most likely changes made than I can make this morning before I have to head for work.

Last edited:

#### kcarring

Joined Jan 22, 2011
38
Thanks. What i was getting at though... is not all about whether the motor can be reversed, etc. I'm interested in knowing why the negative of the motor is connected to the output of the FET, and if a blocking oscillator took the place of the motor, would you connect to the negative side of that circuit.

Thanks

#### SgtWookie

Joined Jul 17, 2007
22,220
It is more convenient to use a low-side switch than a high-side switch.

Comparing N-channel and P-channel MOSFETS; if all other parameters are the same, the total gate charge will be ~2.5 times higher for a P-channel than an N-channel MOSFET. N-channel MOSFETs are most easily used on the low side of a load. They can be used to switch the high side as well, but that requires a high-side driver which I won't get into in this thread.

An N-channel has another advantage in that the grounds for the timer and the load can be the same, and the timer and load voltages can be different. You might have the timer running on a 12v supply, and the load running on a 40v supply - it would still work as-is. It would be much more difficult to use a P-ch MOSFET then.

Since P-ch MOSFETs have pretty much fallen out of favor except for special functions, there is not as much of a selection of them as for N-channel MOSFETs, and they tend to be more expensive.

The blocking oscillator is a different matter. With the motor, you only had mechanical rotational output. The blocking oscillator has an electrical output. If the ground is disconnected, the output will float to the positive supply value via the transformer secondary. If the positive is disconnected, the output will be open-collector with a transistor that is turned off; basically a fairly high impedance circuit. In this instance, the manner in which the circuit is supplied will affect the output, which will impact subsequent stages.

#### kcarring

Joined Jan 22, 2011
38
The blocking oscillator is a different matter. With the motor, you only had mechanical rotational output. The blocking oscillator has an electrical output. If the ground is disconnected, the output will float to the positive supply value via the transformer secondary. If the positive is disconnected, the output will be open-collector with a transistor that is turned off; basically a fairly high impedance circuit. In this instance, the manner in which the circuit is supplied will affect the output, which will impact subsequent stages.

Thanks very much SGT.

Maybe I can elaborate a bit more here with some goals, etc. What drew me into all of this, was:

The goal, is to create a simple, (if perhaps very feeble and not typically useful) inverter circuit. I have many inverters ranging from a 3000 watt Xantrex Freedom, to a WAGAN pure sign, to a cheapo Motormaster 75 watt inverter.

The problem is, all of these inverters while fantastic do not suit my particular needs on this project. I truly want a sh$%~y, feeble, low output inverter. Reason being i want to run about 200 feet of AC line out from this small dedicated inverter, from the off-grid power shed, and give a small amount of light to a few areas. I am totally willing to sacrifice lumens, but the ability to use off the shelf AC LED lamps is nice, and I already tried it with homemade lamps on DC, the power losses are too great. The issue is, that, when I attach the configuration of AC LED bulbs to a simple inverter [three (2) watt AC LED bulbs] - the typical draw is about 1 amp or slightly over. For this application this is unacceptable, I need to put up with less lumens, lower output, and benefit from low draw. All of these inverters draw at least 150 mA unloaded, even... I want to use 500 mA and run five lights. So anyways... I got laughed at a whole bunch (don't need anymore of that really) and told "you can't have that...) and being stubborn I set out to have some sort of inverter or blocking oscillator that was current limited, in a sense. By using a PWM. I did manage to get what I wanted, but my methodology seems/seemed very wonky. I hooked up a reasonably high output PWM with frequency and duty cycle controls, right to the power rail of a flip flop inverter. It worked. Well, sort of.. Lets just say at certain duty cycles and frequencies it gave me what I was after, current limiting without heat. So now I want to figure out how to do the same thing, but more sensibly. So in short, I want a sh%$#! inverter. Cause I can't "afford" the power.. not the inverter, I have many... I'm not trying to save money, I'm trying to limit power usage.

Here is the illumination I was able to get across (3) AC LED lamps at the energy cost of 385 mA:

Using an even more simple blocking oscillator arrangement with the same transformer I was also able to fire up a fluoro at the 5 watt level, and I'd like to be able to "dial down" that output slightly too:

This is the flip flop inverter of sorts (albeit very crude) that I experimented with to then control with the PWM:

Ran normally as shown, in the circuit diagram, it is fairly ridiculous. No control whatsoever, no voltage regulation or feedback... I realize that.

Please keep in mind I am not looking for a typical text-book inverter. I own 5000 dollars worth of those already.

Unless of course you can point me towards one that can do this, run an AC LED lamp, dimly @ 20mA or so:

I realize it may not make a whole lot of sense, but I'll tell you what. Living off grid is like that sometimes.

*I'm trying to fix the pictures here, sorry.

Last edited:

#### SgtWookie

Joined Jul 17, 2007
22,220
The last schematic is really bad. There's one just like it over on arroncake's site that has been giving hobbyists headaches for years.

It's not that you need a lousy converter. You need one that is not only voltage limited/regulated at the output, but also input current limited.

You also need it to be as efficient as possible.

Taking it back a step further, the higher the voltage you're sending across this 200 feet, the smaller gauge wire you'll need - or if using the same gauge wire, your power losses will simply decrease.

At the receiving end, you could use a step-down transformer to get back to the line voltage that you require.

But to keep it more simple, you might just modify an existing inverter that works. Add on an input current monitor, so that when the input current hits the limit, the output voltage is reduced.

I don't know what your battery bank voltage is. If it's 12v, then 0.5A out is only 6 Watts to deal with.

Last edited: