[SOLVED] Problem while implementing a Widlar Current Mirror

[adi0803199]

I have designed a Widlar Current Mirror for a current of 10µA. The design specification is as follows ,
Vcc = 3.0V
R1 = 2.2 kΩ
Re2 = 12kΩ
Transistors : matched BC547 with hfe = 261
The problem that i have is when I connect a load resistance of 2.2kΩ , I get a emitter current of Ie2 = 6.166 µA and collecter current Ic2 = 75.318 µA. (transistor Q2)
I don't know how to debug this problem.Any help to solve this problem.......


I referred this website to design my circuit :
https://en.wikipedia.org/wiki/Widlar_current_source

 

[FvM]

I get a emitter current of Ie2 = 6.166 µA and collecter current Ic2 = 75.318 µA

Sounds simply impossible. Either the circuit or the measurement is incorrect.

 

[nitishn5]

Is the current in the primary side (Q1, R1) correct without any load?
How does it change with load?

Your current measurements sound very weird. Please mention the node voltages.

 

[godfreyl]

Possibilities:

• Something connected the wrong way around.
• Transistor overheated and damaged while soldering.
• Really messy soldering, causing leakage current through burnt flux residue or whatever.

 

[adi0803199]

I found the nodal voltages at the emitter and collecter of transistor Q2 which was
Ve2 = 74 mV ,
Ie2=Ve2/Re= 274mV/12k =6.166 uA
Vbe1= Vc1 =Vb2 = 0.62 V
Vc2 = 2.834 V,
Vload =Vcc -Vc2 =0.1656 V
Iload= Vload/Rload = 0.1656 V/2.2k
=75.318 uA.

 

[godfreyl]

Transistors : matched BC547 with hfe = 261

Matching Vbe is much more important than matching hfe.
That doesn't explain your measurements, though - Ic should never be greater than Ie.

 

[adi0803199]

Sorry I didn't get your point...In a Widlar current source,
Vbe1 = Vbe2+Iload*Re2.
So Vbe of Q1 and Q2 can never be the same unless Iload =0 A.

 

[FvM]

Quite obviously, Vbe matching means equal Vbe for equal Ic. It's also a prerequisite for predictable behaviour of the Widlar current source.