What does it mean that a device or IC (transistor, DAC, etc.) is "current out" versus "voltage out"? I and V are present all the time -- if there is no voltage, there is no current -- so this topic confuses me. When does one use the current versus the voltage? When should I prefer to use a current source, rather than a voltage source?

 

No Voltage no current.
Current out could mean max value that a Typical device can source or sink. Voltage out is also limited to the supply.
Values can be found in the device datasheet.

 

The distinction comes about by looking at what is controlled and how that control is effected.
In the case of a Bipolar Junction Transistor the device behaves as a current controlled current source. It can be placed in a circuit in three different configurations. The controlled current output creates different voltage behavior based on the configuration of the external circuit.
In the case of an enhancement mode MOSFET the device behaves as voltage controlled current source where the gate to source voltage controls the drain current. It can also be placed in a circuit in three different configurations. The controlled current output creates different voltage behavior based on the configuration of the external circuit.

The answer to your other question depends on what you are trying to do. If you have a load that requires a constant voltage at whatever current it requires then a voltage source is appropriate. If on the other hand the load requires a constant current at whatever voltage is required a current source is the thing.

A resistor is all you need to convert a voltage source into a current source, to drive an LED for example.

 

Current output DACs are designed to output a digitally-controlled current into the "virtual ground" presented by the inverting input of an external opamp. Usually, current output DACs also provide a feedback resistance for the opamp which compensates for any resistance tempco of the DAC's R/2R output ladder. Most current output DACs are also what we call multiplying DACs, because they are designed so that they can be used with either positive or negative voltage reference inputs or even an AC reference voltage input, in which case they can be used as "digital volume controls."

Voltage output DACs have the amplifier built-in, and do exactly as their name implies: they output a digitally-controlled voltage. Usually they are restricted to a positive-only reference input.

Which one to use? Depends on whether you need the flexibility and versatility of the current-output DAC, or need the low overall component count of the voltage-output DAC.

If you want to understand current-output DACs a bit more, here's a link to the page for the LTC8043, which I use frequently:

http://www.linear.com/product/LTC8043

 

Many lab supplies have a constant-current or constant-voltage mode.
In the CC mode the output current is constant, independent of the load (up to the maximum voltage output of the supply).
In the CV mode the output voltage is constant, independent of the load (up to the maximum output current of the supply).

 

The distinction comes about by looking at what is controlled and how that control is effected.
In the case of a Bipolar Junction Transistor the device behaves as a current controlled current source. It can be placed in a circuit in three different configurations. The controlled current output creates different voltage behavior based on the configuration of the external circuit.
In the case of an enhancement mode MOSFET the device behaves as voltage controlled current source where the gate to source voltage controls the drain current. It can also be placed in a circuit in three different configurations. The controlled current output creates different voltage behavior based on the configuration of the external circuit.

The answer to your other question depends on what you are trying to do. If you have a load that requires a constant voltage at whatever current it requires then a voltage source is appropriate. If on the other hand the load requires a constant current at whatever voltage is required a current source is the thing.

A resistor is all you need to convert a voltage source into a current source, to drive an LED for example.

Let's take the BJT for example. This is a CCCS -- yet to turn it on, I need to ensure that the base voltage (NPN) is 0.7V higher than the emitter voltage. So it seems to me that it being controlled by a voltage. How is it current controlled? And by current source, do you mean that if there is a load on the collector, that the current will be constant even if the load changes (within the practical limits of the device)?

 

It should be has the current flows through the Vbe diode, when it reach up to the turn on voltage about 0.65V, and then the C will be going to low in saturation situation about 0.2V when the Ib continuing to increasing, you just care about that you input the voltage and the Vbe will be turn on, and then the c will be turn on too, but when you forgot the Rb which is a current limiting resistor, if you didn't use it and just adding the voltage cross on the Vbe, you can see what will it happen?

 

Here's another way to think about this.
1) Start with voltage. If there is no voltage, there is no current, and nothing happens.
2) Every voltage source has an impedance. If it didn't, you could weld the Eiffel tower with a 9 volt battery.
3) Every load has an impedance. If it didn't, the power source would be shorted out instantly.

Immediately you realize that voltage must be applied, but it can only withstand the load within the limits of its own source impedance. Then you look at the circuit. I think you're talking about a common emitter amplifier. What impedance will it present to this voltage?

Well, a base-emitter junction won't present much impedance, so you simply must have a resistor on one end or the other. If you place the resistor on the base side, that resistance will limit the current flow and the gain of the transistor will allow some greater current through the collector if it is available. Too much resistance in the collector circuit and the collector voltage collapses and the transistor is in saturation. Very good for digital switching purposes.

If you place the resistance in the emitter side, you can apply most any reasonable voltage on the base and the emitter resistor will limit the current through the base-emitter junction. If there is current available at the collector, it will flow through the emitter resistor and fulfill most of the required current to keep the emitter voltage a few tenths of a volt below the base voltage.

If that kind of description rings your bell, reply in that vein for more information.

As a last word, I would like to point out that 0.6 volts or 0.7 volts is not a magic number. A small signal transistor will conduct collector current according to the base current across a range of millions to one. How do I know? I measured it.

 

Let's take the BJT for example. This is a CCCS -- yet to turn it on, I need to ensure that the base voltage (NPN) is 0.7V higher than the emitter voltage. So it seems to me that it being controlled by a voltage. How is it current controlled? And by current source, do you mean that if there is a load on the collector, that the current will be constant even if the load changes (within the practical limits of the device)?

You can always look at a source from both viewpoints.
You do need a minimum voltage to start the base current flowing, but at that point you need some way to control or limit the base current (generally performed by a resistor in series with the base).

Yes, the transistor collector current is only slight changed by a change in collector voltage for a given base current.

 

) Every voltage source has an impedance. If it didn't, you could weld the Eiffel tower with a 9 volt battery.

I didn't know the Eiffel tower needed welding, but I was looking for an interesting project.

Let's take the BJT for example. This is a CCCS -- yet to turn it on, I need to ensure that the base voltage (NPN) is 0.7V higher than the emitter voltage. So it seems to me that it being controlled by a voltage. How is it current controlled? And by current source, do you mean that if there is a load on the collector, that the current will be constant even if the load changes (within the practical limits of the device)?

Actually, you need to inject a defined amount of current into the base to turn on a transistor.
In order to do this you will need some voltage, as everyone keeps telling you.

 

Let's take the BJT for example. This is a CCCS -- yet to turn it on, I need to ensure that the base voltage (NPN) is 0.7V higher than the emitter voltage. So it seems to me that it being controlled by a voltage. How is it current controlled?

For an NPN BJT, the circuit driving the base sees a PN junction (a resistance) and base current is required to turn it on. Hence it's a current controlled device.

A circuit driving an MOS transistor sees the gate capacitance. Voltage applied to the gate controls channel formation, so it's a voltage controlled device.