Requirement of emitter resistors in current mirror circuit

Join the discussion
Registration is free. Start your own thread to ask a follow-up.
9 replies · 4K views
brainbaby
Messages
232
Reaction score
5
In a current mirror circuit ..presence of an emitter resistor (R1 and R2) increases the output resistance which minimises harms of early effect...how?
why its advantageous to have a high output resistance in current sources/current mirror circuits?
 
Attachments
  • current_mirror_re.gif
    current_mirror_re.gif
    1 KB · Views: 647
Engineering news on Phys.org
Think about how the early effect manifests itself. As the output voltage changes, the output current changes. That indicates a finite output impedance. Since adding the emitter resistors increases the effective output impedance, it reduces the change in I caused by a change in V.

delta I = deltaE/deltaR. If R is infinite, no amount of E change will change I.
 
  • Like
Likes   Reactions: brainbaby
brainbaby said:
why its advantageous to have a high output resistance in current sources/current mirror circuits?

Because - by definition - an ideal current source is identical to a voltage source with an infinite source resistance.
 
  • Like
Likes   Reactions: brainbaby
meBigGuy said:
Think about how the early effect manifests itself. As the output voltage changes, the output current changes. That indicates a finite output impedance. Since adding the emitter resistors increases the effective output impedance, it reduces the change in I caused by a change in V.

delta I = deltaE/deltaR. If R is infinite, no amount of E change will change I.
nice explanation ...thnks
 
R1 and R2 help linearise the mirror for higher currents. They reduce the effect of Vbe negative temp coefficient and reduce the thermal imbalance.

By placing another transistor above the mirror you can equalise the collector voltages and so make a mirror with a better thermal power balance.
 
  • Like
Likes   Reactions: brainbaby
Baluncore said:
R1 and R2 help linearise the mirror for higher currents. They reduce the effect of Vbe negative temp coefficient and reduce the thermal imbalance.

By placing another transistor above the mirror you can equalise the collector voltages and so make a mirror with a better thermal power balance.
Are you talking about this circuit(refer attachment)..
The transistor Q3 is involved just to cancel the effects of base current ...i mean base current is quite small in magnitude (that should be neglected ) then what is the requirement to minimise base current...and involve Q3?
 
Attachments
  • current_mirror_buffer.gif
    current_mirror_buffer.gif
    976 bytes · Views: 607
See attached. The accurate current balance here is done by Q1 and Q2. The rest of the circuit is there to help them.
For high currents R1 and R2 swamp the Vbe of Q1 and Q2 so as to prevent thermal offsets or runaway.
Think cascode, Q2 and Q3 make the collector voltages of Q1 and Q2 very similar in magnitude, so heating is the same in both Q1 and Q2.
D1 keeps Q1 and Q2 on the edge of saturation, a small collector voltage makes less heat.
Q5 produces the base current for all other transistors. If Q5 is high beta then most base drive comes from Vcc, not Iin.
Only Q1 and Q2 need a common thermal environment. Temperature of other components is unimportant.
 
Attachments
  • CurrentMirror.jpg
    CurrentMirror.jpg
    21.2 KB · Views: 561
Baluncore said:
See attached. The accurate current balance here is done by Q1 and Q2. The rest of the circuit is there to help them.
For high currents R1 and R2 swamp the Vbe of Q1 and Q2 so as to prevent thermal offsets or runaway.
Think cascode, Q3 and Q4 make the collector voltages of Q1 and Q2 very similar in magnitude, so heating is the same in both Q1 and Q2.
D1 keeps Q1 and Q2 on the edge of saturation, a small collector voltage makes less heat.
Q5 produces the base current for all other transistors. If Q5 is high beta then most base drive comes from Vcc, not Iin.
Only Q1 and Q2 need a common thermal environment. Temperature of other components is unimportant.

Shouldn't the transistor designators above in red be as I've shown?

There's more on the topic at: http://en.wikipedia.org/wiki/Wilson_current_mirror
 
The Electrician is quite correct, my typo.

Note that the diode in my diagram can be replaced with a transistor to give base current cancellation. See attached.
This is probably a buffered, improved, Wilson Current Mirror.

The problem with the Wilson mirror is that the Vce voltages are different for the critical pair of transistors. Power dissipation, and therefore heating differs by a factor of two between the two matched transistors.
 
Attachments
  • mirror2.jpg
    mirror2.jpg
    24.5 KB · Views: 525