Constant current sources

[Lucifre]

317 constant current

Hi... I need some ideas for constant current source....I know about lm 317 configuration but the resistances needed to generate currents i need are too low (almost impractical) for my purposes....
is there a adjustable voltage regulator with higher refrence voltage than 1.25V?

 

[IanP]

What about current sources build with opamps?
Here is one example:
https://members.tripod.com/michaelgellis/howland.html

Regards,
IanP

 

[ptmsl]

Any fixed regulator with any output voltage can also be used as an adjustable regulator with higher reference voltage. Use the ground pin as the adjust pin and the rest is the same only problem is that the shunt resistor used for current sensing will consume more power and for very low currents the regulator biasing current may produce a large error and reduced current regulation. Interested in trying with a shunt regulator (like 431) and a transistor or even an opamp? What is the current that you want to regulate and the max power (or voltage) that you will need at the output?

 

[Kral]

Lucifre,
You didn't mention your accuracy or stability requirements. A depletion mode FET (almost all JFETS are depletion mode) with gate connected to source makes a crude current source. It's output resistance will be very high, but you'd have to select devices to get the current you need.
Regards,
Kral

 

[VVV]

This is really puzzling me:

I know about lm 317 configuration but the resistances needed to generate currents i need are too low (almost impractical) for my purposes....
is there a adjustable voltage regulator with higher refrence voltage than 1.25V?

I always found the reference voltage too high for any practical current source using LM317. What kind of current are you considering?
I mean, even at the maximum allowable current of an LM317, which is 1.5A, the resistor needed is 1.25V/1.5A=0.83Ω (that is 0.825Ω standard 1%). This seems like a very reasonable value to me, not very small in any case.

Here is why I say the 1.25V reference always seemed too high to me: if you consider the power dissipation in this resistor, it becomes:
Pd=0.825Ω*1.5^2=1.85W. That is a lot of power to waste, on the one hand, on the other, it adds to the minimum voltage overhead, which becomes about 3.25V~4.25V. Finally, you need a 5W bulky resistor.
Why would you need an even higher reference?