Understanding Amps as through a current source

I am having a hard time understanding the role of a current source. Although I know this isn't true, to me, it seems like the only thing you really need in electronics is voltage to motivate or apply pressure to the electrons that already exist in the copper of the traces and the components of a given circuit or project.

Despite the material I have read, it seems like an Ampere is just an idea - a product of the electrons in a system moving past a point. They wouldn't move without voltage. So when you have a current source from some sort of power supply are you actually introducing more electrons to the system to increase the Amps at a given voltage?

 

Fullnelson,

Both sources, voltage and current accomplish the same thing, but in different ways. In both cases there is a voltage across the load, and a current through the load.

Lets assume the load is a resistor...

Voltage source:
A voltage source holds a constant voltage across the resistor.
The load current is then whatever is required to maintain this voltage across the resistor, I=V/R

Current Source:
A current source forces a constant current through the load.
The voltage across the load is whatever is required to maintain this current through the resistor, V=IR

 

Current is just as real as voltage. Sure, if only voltage sources existed most everything we make in electronics would still be there, albeit there would be some extra complications in those cases where driving the current is better then driving the voltage.

A physical current source, assuming it's a constant current, is a circuit designed to work very hard to produce a given current within it's design limits. The voltage at the output will change to whatever level is required to keep that current flowing.

Mister Kirchhoff would be very disappointed if you don't believe in current.

 

Maybe an example helps?

A 5V voltage source applied to a 10ohm resistor results in a half amp current, 5V/10ohm=0.5A

A half amp current source applied to a 10ohm resistor results in 5V across the resistor, V=0.5A*10ohms=5V

 

So you ask, why would I ever need a current source, right?

Lets say I want to drive an LED at 20mA to get a certain brightness. And I'm driving it from a 2-cell Lithium Ion battery.

Using a voltage source approach ... I put a resistor in series with the LED and I size the resistor to give me 20mA through the LED when the battery is fully charged to 8.4V. But, when the battery is discharged to 6V the voltage has decreased and the current decreases, and the LED brightness decreases.

If I instead use a current source, I can design it to always source 20mA through my LED when the battery voltage is at 8.4V or 6V.

 

Constant current sources also are used for driving non-linear loads like heater filaments in high power tubes or plasma drivers where the controlled parameter is an electron emission, ions or heat. The resistance slope can be positive or negative making long-term precision stable operation with voltage regulation feedback tricky. The on-rush current with a cold filament is also eliminated increasing tube or source life.

Most good battery chargers also use a constant current mode when in the absorption phase.

A message from the master.
http://www.youtube.com/watch?v=411f0DvXu18

 

So are you implying that you have two inputs into your circuit, one is a battery and one is an independent current source so that when the voltage drops the current is sustained?

Can you give me an example of a constant current source in this case?

 

I think he means the LED brightness will remain the same at different voltages if we regulate the current to a fixed value.

If we think about why constant current regulation works in LEDs and other devices in the same class we see that actual electron/hole (charge carriers) movement is important to the production of output. The movement of the charge carriers gives us the desired response (light) to the energy flowing in the circuit via a EM field from a battery or power supply. The PN junction in the LED creates a barrier that makes the charge carriers lose the gained energy needed to jump that barrier when they recombine on the other side. That loss of energy is what we see as light (and heat) in a LED. So if current is the movement of charge carriers in a conductor and the amount of carrier movement determines the LED brightness, it's logical to regulate current.

 

Here ya go. Here's an 2V LED, biased by a 20mA current source vs a resistor as the battery voltage goes from 6V to 8.4V.
There are several ways to make current sources (and sinks).

 

An inductor is a constant current source for a limited time until its energy is expended. A capacitor is the counterpart being constant voltage until its energy is spent.

The generators at the power plant can be operated as either, constant voltage source (CVS), or constant current source (CCS). To obtain a CVS, you spin the generator at constant speed. To obtain a CCS, you spin at constant torque. The CVS is used exclusively in power distribution because losses are lower. Conductors lose more power than insulators, so they generate at 100% voltage all the time, & the current varies with load, & it is usually well below 100% capacity.

A car alternator is similar. The engine speed is not constant so the alternator voltage is not either. The torque is not constant so neither is the current. A feedback system is used to maintain a CVS by varying field current, and/or duty cycling the output so that the battery & 12V bus are maintained at a steady voltage.

Batteries can be produced for CCS mode, but they perform better & last longer with CVS mode. A nuclear battery is best run at CCS mode. But it is not likely that nuclear cells will be commercially available anytime soon. They do work better as constant current sources, but conventional cells are more suited for constant voltage.

If magnetic levitation for trains is utilized, a CCS is ideal for such an application. Current is needed to maintain the magnetic field, but very little voltage is needed, zero for super-conducting rails. Also, should LED lighting replace conventional street lighting entirely, a CCS would be a great source for such lamps.

Have I helped?

Claude

 

No, it is used eclusively because the distribution network is AC and you need all the generators to be synchronized. If you had a constant torque the generator speed would change and that directly changes output frequency, which is not a good thing when you have more generators in parallel.

 

The way I describe it is as follows:

A Voltage source provides a constant voltage across a load, with varying current.

A Current source provides a constant current through a load, with varying voltage.

 

No???!!! CVS was used even prior to generators being tied together. Insulator losses are V^2*G, while conductor losses are I^2*R. Ask any EE & they will affirm that conductor losses are much higher than insulator losses. Another point worth mentioning is the stepping up of voltage via xfmrs. Insulator losses are so low compared with conductor losses, it pays to maximize V while minimizing I.

As a side benefit, constant frequency is a great feature. Not only does it facilitate tying generators together, but it allows the use of synchronous motors. With CCS, the steady speed of synchronous motors would not be realized. CCS would provide varying freq.

Batteries, OTOH, are not synchronized with other batteries, yet they are designed & built for CVS. As I said, nuclear cells are a long way off for commercial use. But nukes are better as CCS, conventional cells are better as CVS.

But your point about needing constant frequency to maintain synchronization is well taken. I just mentioned losses to point out the historical reasons for CVS.

Claude

 

I would be very interested in seeing examples of these.

 

OK try this; connect 9 AA cells in series and try to start your car.

 

My car started right up.

 

Don't forget to take out the regular battery first

 

Any switching power supply uses inductors to store energy for part of the cycle. Then, the inductor delivers its energy to the output & it functions as a CCS. Take a buck converter. When the power switch (FET) is on, current builds up in the inductor as well as energy. The output is powered since the load is in series with the input supply, FET, & inductor. The output cap is also being charged.

The FET is switched off. The inductor de-energizes through the catch diode, output cap, & load. At the transition moment, the inductor maintains its present value of current. It functions as a CCS. It loses energy as it transfers energy to the output, per conservation law.

A similar scenario exists for a switching LED constant current driver. LED lamps which are driven with constant current use a switching topology where the inductor drives the LED along with catch diode, FET, & a cap if needed. A search for LED drivers should turn up good info.

Claude

 

I knew all that. I was taking issue with the phrases (I've highlighted them in bold) in your original statement: