variable constant current source?

[Zero Cool]

I would like to build a variable constant current source for experimenting with various stage topologies. I would like something that is variable from say 1ma to more than 10ma maybe as high as 100ma?

any suggestions?

 

[EUVL]

See TL431 datasheet. Fig.32.
http://www.ti.com/lit/ds/symlink/tl431a.pdf


Patrick

 

[Zero Cool]

I will take a read. I think I may have some of those? I will have to look

 

[CharlieLaub]

This might be of interest to ya, mate!

Constant Current Source/Load (LM317) | The Paranoid Troll

In general, the LM317 IC is used like this:

An externally hosted image should be here but it was not working when we last tested it.

I did a quick Google search, based on my memory of seeing this in a data sheet once upon a time. I am sure that there are other resources about this out there.

-Charlie

 

[EUVL]

A more modern alternative to LM317 for low current:
LT3092 - 200mA 2-Terminal Programmable Current Source - Linear Technology

I also use LM317 or LT1083, etc., but usually for current above 500mA.


Patrick

 

[jcx]

while regulators can be hacked to give CCS they often have limited bandwidth

purpose designed chip:

LT3092 - 200mA 2-Terminal Programmable Current Source - Linear Technology

can also make "Howland Current Pump" with op amps or power chip amps depending on requirements

http://www.ti.com/lit/an/snoa474/snoa474.pdf

 

[abraxalito]

I would like to build a variable constant current source for experimenting with various stage topologies. I would like something that is variable from say 1ma to more than 10ma maybe as high as 100ma?

I would have suggested LM317 like the others but I think down at 1mA it won't regulate properly. Also as jcx says the bandwidth is somewhat limited - watch out for the parasitic capacitances.

A solution that will encompass 1mA would be an opamp driving the gate of a MOSFET with a resistor to a power rail from its source. -ve input of opamp to source, +ve input of opamp you feed a voltage which defines the current you want. Choose the resistor to suit the current range you require. You need an opamp whose CM input voltage goes down to the -ve supply with an N-channel MOSFET - this makes a current sink. For sourcing, a P-chan MOSFET is called for and the opamp needs to allow CM voltages up to the +ve supply.

 

[PaulBysouth]

A depletion mode mosfet, like the DN2530 or DN2540, will make very good current sources up to about 100mA. See attached drawing.
pb

 

[Zero Cool]

A depletion mode mosfet, like the DN2530 or DN2540, will make very good current sources up to about 100mA. See attached drawing.
pb

I have plenty of DN2530's on hand. I like that solution. a little limited at only 25ma but that will work for most things. I will review the others, thank you everyone!

 

[iko]

Why not something simple like this. Almost any general purpose npn and power mosfet will do. Play with R1 and R2 to get the current you want.

 

[CharlieLaub]

I would have suggested LM317 like the others but I think down at 1mA it won't regulate properly. Also as jcx says the bandwidth is somewhat limited - watch out for the parasitic capacitances.

A solution that will encompass 1mA would be an opamp driving the gate of a MOSFET with a resistor to a power rail from its source. -ve input of opamp to source, +ve input of opamp you feed a voltage which defines the current you want. Choose the resistor to suit the current range you require. You need an opamp whose CM input voltage goes down to the -ve supply with an N-channel MOSFET - this makes a current sink. For sourcing, a P-chan MOSFET is called for and the opamp needs to allow CM voltages up to the +ve supply.

Oops, you are right about 1mA - I misread the post. But it seems that the 317 can get down to a couple of mA depending on the voltage drop. See below:

-Charlie

 

[GILLES GOSSELIN]

Useful 0-10A CURRENT SOURCE

I'm (new here ),and wish you a New Year! and bring new things like a 0-10A variable precision current source implemented with a XTR110 from Burr-Brown.

I'll come up with a class A later done with-it

 

[DUG]

I would like to build a variable constant current source for experimenting with various stage topologies. I would like something that is variable from say 1ma to more than 10ma maybe as high as 100ma?

any suggestions?

What voltage swing do you need to accommodate?

Kind of important.

Some more:

Isolated? (floating)
Stable?
Accurate? (or you could measure what you dial)
Equivalent source resistance? (Mohms)

Been there, done that.

 

[FoMoCo]

You can also do a Howland Current Pump with an op-amp and external power device. An advantage of this is that the input can be derived from a reference. (Even a DAC if you so please.) The Howland does have its disadvantages, such as needing twice the load voltage as a supply voltage and sometimes being hard to compensate.