I am trying to do something like this. My load will be up to 3K ohms and I want to be able to drive up to 10mA current through it. Which would mean I would need a supply rail of +/- 35V to be comfortable. Looks like there are very few Op-Amps that can handle 70V R-R voltages. Is there a simpler way of doing this?

The load has be on the low side, it cannot be a high side / floating load.

 

How accurate and stable does the current need to be with varying load resistances?

 

Why do you need -35V? Do you need -10mA?

 

You could used matched voltage dividers to reduce the voltages at the inputs of the opamp, and use a level-shifter on the opamp output. That would allow use of a 'normal' opamp (albeit with reduced current resolution related to the division ratio).

 

nowwith3shldetect -- if such for any improvement
No credits of any kind ((the feedback here is not exactly what it should be -- i just tested if i manage to fetch a mosfet 3shld - which now i approx can imagine is to figure out minimum conductance OpAmp circuit . . . ))
+more fq. tolerant v.
((nothing to do))

 

Here's a simple circuit that should do what you want, using a dedicated constant-current IC.
The LT3092 requires only a minimum of 1.2V headroom.
The LTspice simulation is shown below for the pot U2 rotation of 0 to 100%.

 

Hi crutschow.
Iout = (R(U2) / R2+1) * 10uA

 

Hi crutschow.
Iout = (R(U2) / R2+1) * 10uA

By that you mean ((R(U2) / R2) + 1) * 10uA to be totally precise and add the 10uA reference current to the total current (which can typically be ignored unless you are trying to generate very low currents).

 

Why do you need -35V? Do you need -10mA?

I do not need -10mA, but I need to be be able to get down to 0uA. Dont I need -35V to be able to get close to 0uA.

 

Here's a simple circuit that should do what you want, using a dedicated constant-current IC.
The LT3092 requires only a minimum of 1.2V headroom.
The LTspice simulation is shown below for the pot U2 rotation of 0 to 100%.

View attachment 113488

I want to be able to go below 0.5mA, apart from that this was a great idea. Will probably use this sometime in the future.

 

I want to be able to go below 0.5mA, apart from that this was a great idea. Will probably use this sometime in the future.

That requirement would have been nice to know up front.

The simulation shows a minimum of about 0.3 mA.
The worst-case minimum current @ room ambient would be about 2/3 mA minimum due to the offset voltage.

 

go below 0.5mA

? and what's the difficulty in here
http://www.vishay.com/docs/70596/70596.pdf
http://www.radio-electronics.com/info/circuits/transistor/active-constant-current-source.php
https://en.wikipedia.org/wiki/Current_source

 

That requirement would have been nice to know up front.

The simulation shows a minimum of about 0.3 mA.
The worst-case minimum current @ room ambient would be about 2/3 mA minimum due to the offset voltage.

Ur right its too big a parameter to be missed out. Anyways thanks for the suggestion as I mentioned, I am sure this will come in handy someday.