Could you explain a bit more in detail what you have and what you need to turn it into, please?
Above suggestion is good. Greater stability perhaps with op amp + reference voltage + transistor + feedback resistor as another option but perhaps not simplicity/size required.
Thanks for all your suggestions. Sorry I have not given full details earlier.
I have an instrument which is giving a 4-20mA current sink, where I have to connect to another instrument which will accept current source. I can't modify either of the device internal circuitry. So I have to insert a circuit in between these two instruments.
I have tried earlier as per tyassinsuggestion, but it gives 0.3mA added current output. May be I have to try with a matched pair of transistor or will try with MOSFET instead of BJT. I can happily accept If you have any other suggestions.
You neither did now.
* voltage range of the source
* voltage range of the load
* (or available power supply voltage)
* load resistance (range)
* expected precision (you complained about it, so it's urgent to give at least a clue of what you expect)
....
I'm simulating Easy's second schematic (post #5). Classic current mirror.. Various transistor gains. Various emitter (topmost) resistors.
Notice at bottom-left is a resistor where you substitute your detector which sinks current.
The current mirror sources current to the load at bottom-right, where you substitute your receiving instrument.
By adjusting the emitter resistor values you can get the right side pretty much equal to the left. To a certain degree it makes up for transistor mismatch.
You don't want to use MOSFETs in a current-mirror as the Vgs mismatch between two devices is much higher than BJTs.
Use a matched pair of BJT transistor for the current mirror.
The normal Vbe mismatch between two discrete transistors can cause the current offset you noted.
Also, you might use a 3-transistor Wilson current mirror (below) which is much less sensitive to the output voltage.
(Change all the transistors to PNP for your purposes).