My project I am creating a system that should be battery powered, at most 30-50 VAC.
I am generating square waves to stimulate human arm.
I need current to be constant between 0 - 10 mA across human skin resistance.
Therefore I need a constant current source.
I also need current to be going both direction.

I think constant current source could be either op amp based or transistor based?
What is difference?
Do they all use varying voltage to regulate current?
Do they all allow current to go both direction?

I am new to electric engineering so I apologize if anything is unclear.
Could anyone point me a direction?
Thank you.

Also I tried

With V+ = 5V, Vcc = +-10V, using 180 ohm instead of 250 ohm, AD822 op amp.
circuit works when Rload = 0, 28 mA.
but when I put Rload as 680 ohm, current exceeds 28 mA as Vcc increase.
Anyone know what might be the error?

 

My project I am creating a system that should be battery powered, at most 30-50 VAC.
I am generating square waves to stimulate human arm.
I need current to be constant between 0 - 10 mA across human skin resistance.
Therefore I need a constant current source.
I also need current to be going both direction.

I think constant current source could be either op amp based or transistor based?
What is difference?
Do they all use varying voltage to regulate current?
Do they all allow current to go both direction?

I am new to electric engineering so I apologize if anything is unclear.
Could anyone point me a direction?
Thank you.

Also I tried

With V+ = 5V, Vcc = +-10V, using 180 ohm instead of 250 ohm, AD822 op amp.
circuit works when Rload = 0, 28 mA.
but when I put Rload as 680 ohm, current exceeds 28 mA as Vcc increase.
Anyone know what might be the error?

By 'transistor' I take it you mean 'discrete' transistor?

A discrete transistor (as the sole active element) cannot perform the function of a constant current source... Perhaps you are inquiring as to the function of a (discrete) pass transistor? (in which case the 'short answer' is that such an arrangement is implemented as a means of increasing current handling capability...)

I think constant current source could be either op amp based or transistor based?
What is difference?

Um... The former is integrated whereas the latter is not...

Best regards
HP

 

The simplest generatoris built on JFET-transistor and resistor.

 

The simplest generatoris built on JFET-transistor and resistor.

If you have an example circuit for a single semiconductor CCS I'd love to see it! --- Sincerely!

Best regards
HP

 

Also generator is based on MOSFET-transistor with built-in channel (depletion mode) and resistor.
These transistors can be high voltage.



Moderators note : Please cut away the white space.

 

Also generator is based on MOSFET-transistor with built-in channel (depletion mode) and resistor.
These transistors can be high voltage.View attachment 86924

Interesting! 'Tho I must confess to being at a loss as to how these circuits maintain thermal stability sans junctions operating in their Zener region...?

Moreover, It seems the left-hand circuit would be perpetually 'cut off'?!
Is it the case that the device is being operated in breakdown so as to produce a Zener effect?

EDIT: I get it! -- It's apparently a depletion mode device? -- Still, I don't see any measure of thermal stability, though...?

Thanks for the intriguing post!
HP

 

A BJT transistor makes a dandy contant current source if you have a stable voltage source and a resistor. I am a little late in this thread, but if you want to discuss it let me know.

 

A BJT transistor makes a dandy contant current source if you have a stable voltage source and a resistor. I am a little late in this thread, but if you want to discuss it let me know.

An example (circuit) would be interesting

 

Current Source

Current Sink

The diodes are quick and dirty voltage sources.

 

Added note, with minimum parts they also make voltage programable constant current sources.

 

Current Source

Current Sink

The diodes are quick and dirty voltage sources.

Thanks! --- 'Tis always nice to learn new tricks, and, especially, perspectives! -- Keeps me in my youth, as it were!

Best regards and many thanks
HP

 

If you have an example circuit for a single semiconductor CCS I'd love to see it! --- Sincerely!

Best regards
HP

http://forum.allaboutcircuits.com/blog/low-temperature-drift-jfet-constant-current-circuit.579/

 

http://forum.allaboutcircuits.com/blog/low-temperature-drift-jfet-constant-current-circuit.579/

Many thanks!

 

Wow! You're quick this morning!
Anyway, that should intrigue you. I think it's exactly what you asked for.

 

Wow! You're quick this morning!
Anyway, that should intrigue you. I think it's exactly what you asked for.

Indeed it is! To come to a thread as an 'instructor' but come away as a student is a most welcome turn of events indeed!

Again, many thanks!
HP

 

To come to a thread as an 'instructor' but come away as a student is a most welcome turn of events indeed!

That's one of the best parts about this site. (Read my signature.) There is almost always somebody that knows more than I do about any particular subject. When half a dozen of us conspire, a thread becomes a resource.

 

Here is an example of a constant current source with a different polarity.Used ICs and transistors.



Moderators note : AGAIN , Please cut away the large white spaces

 

Hello,

@Bordodynov , Please show ONLY the essential parts of the picture.
Please cut away the large white spaces, as I have done now.

Bertus

 

Indeed it is! To come to a thread as an 'instructor' but come away as a student is a most welcome turn of events indeed!

I've always thought true learning entailed three simple steps:
1. See one.
2. Do one.
3. Teach one.

Step 3 is the most importaint one as you encounter questions you would never have thought of yourself.

And yep, I too learn many things here myself.

 

The one-(bjt)transistor CCS has been around for ages. It has moderate thermal issues without some form of stabilization. For the bjt, that can be as simple as adding 1 diode or a diode-connected transistor in the base bias network. I've never liked the single-fet version because both the thermal drift and absolute value have such bad tolerances. But I think all CCS "diodes" are in fact the single-fet version internally.

A few years ago I had to drive a 4 V pulse down a 50 ohm source and destination terminated coax cable with only 5 V power. Rather than boost the 5 V to 10 V to power a voltage-mode driver, I drove the cable with a one-transistor (PNP) CCS with a diode-connected identical transistor in the base and a small trimpot in parallel with the emitter resistor. Worked first time.

ak