I am trying to build a gain-of-1 current controlled current source (current mirror) that is less sensitive to resistor value variation. The designs i have been making are highly resister dependent, as I am using shunt resisters.


  • On the input and output sides, I have 24vdc source
  • The input side control current is well behaved.
  • Input and output safe-operating-range must include 4mA-20mA
  • The load is 500 ohms

Current Approach

This is the circuit I have been playing with. The circuit is very dependent on the 50 ohm resisters.

I have looked at a current mirror, but the matched mosfets I can get only have a VDS of 10V, when given a 24vdc supply, bad things can happen.

enter image description here

  • \$\begingroup\$ a bipolar transistor is a current controlled current source. You need to be clearer with specs. \$\endgroup\$
    – markrages
    Mar 26 '13 at 20:13
  • \$\begingroup\$ What is the range of the input current? What voltage range must the output current be correct over? Does this device need to source current, sink it, or both? \$\endgroup\$ Mar 26 '13 at 20:28
  • \$\begingroup\$ input current is 4ma to 20ma. The output has a 24vdc supply. This device needs to source 4ma to 20ma based on the input. \$\endgroup\$
    – David
    Mar 26 '13 at 20:36
  • \$\begingroup\$ Please answer the questions. Your concept of the circuit to meet the specs is of little use at best since it doesn't work. Tell us what you want to accomplish, not how you think it should be accomplished. \$\endgroup\$ Mar 26 '13 at 20:42
  • 1
    \$\begingroup\$ A piece of wire will supply 4-20 mA out with 4-20 mA input... what are your input and output voltage compliance requirements? \$\endgroup\$
    – markrages
    Mar 26 '13 at 20:44

Traditionally you would want to use an instrumentation amp of some kind.

Here's a circuit that uses a ZXCT1009 instead. Accuracy is up to the 4.7 / 47 ohm resistor matching and the ZXCT's transconductance accuracy, about 1%.

enter image description here

  • 2
    \$\begingroup\$ Geesh, can you use a napkin and crayon next time? We can almost still read this one. \$\endgroup\$ Mar 26 '13 at 21:22
  • 4
    \$\begingroup\$ It's hard to erase on napkins; I've given up drawing circuits on them. \$\endgroup\$
    – markrages
    Mar 26 '13 at 21:30
  • 2
    \$\begingroup\$ @Olin - It happens to the best of us. I suggest that you bite the bullet, accept the unavoidable, and take the sensible action needed to address your age-related disability. Here the local $2 shops sell spectacles from 1 diopter to 4+ diopter in 0.5 diopter steps for only $NZ2/pair. Presumably your local shops offer similar. I keep a quantity of these scattered around the house and work areas plus have a screen-suitable pair on a cord around my neck. Makes reading of circuits, manuals, napkins and Global Tour pads a doddle. (May not be so good on crayon though). \$\endgroup\$
    – Russell McMahon
    Mar 27 '13 at 5:20
  • \$\begingroup\$ @markrages - I seem to recall that somebody or other did a long and detailed and 'not too bad considering' writeup on diagram presentation standards as a stack exchange answer. May be worth looking up :-). I have a penchant for squared paper myself. \$\endgroup\$
    – Russell McMahon
    Mar 27 '13 at 5:23
  • \$\begingroup\$ Have you checked that the circuit is stable ? what cures would you recommend if it is not ? \$\endgroup\$
    – Autistic
    Mar 21 '16 at 10:14

Not the answer you're looking for? Browse other questions tagged or ask your own question.