stupid current mirror - confusion

Thread Starter

pgo1

Joined Nov 7, 2012
48
I have never put a current mirror circuit together before. Today I did, and it doesn't seem to work. I'm using a SSM2220 matched dual PNP, current mirror as seen here -



http://sub.allaboutcircuits.com/images/03282.png

when Vcc is around 12V and Rbias is 10K, I get 1.153 mA bias current. Then if the load resistor is 5.6K, I get 1.297 mA load current ? Now as I understand it, these currents are meant to be the same, but it seems whatever the values of the resistors they never are. What might be the reasons for the difference?

....also does pretty much the same thing when i simulate it, although there are loads of bugs in this simulation program...

have i misunderstood something?
 
Last edited by a moderator:

crutschow

Joined Mar 14, 2008
30,137
No.
The simple 2-transistor current mirror load current does vary some with load resistance due to the collector impedance of the output transistor being finite, typically a few tens of k ohms (it would need to be infinite to give no variation).
For much better regulation you can go to a 3-transistor Wilson mirror.
You should find its current varies much less with a change in load resistance due to the negative feedback from the added transistor.

What simulation program are you using that has "loads of bugs"?
 
Last edited:

Thread Starter

pgo1

Joined Nov 7, 2012
48
Most of the "bugs" with Multisim reside with the user of the simulator, not the simulator itself. :rolleyes:
I have verified numerous bugs with National Instruments, so stop the condescending attitude

1.153mA flowing in a 10k resistor would drop 11.53V and that would only leave 0.47V across the transistor and therefore the collector base is forward biased. It won't work well like that.
https://wiki.analog.com/university/...chapter-11#imperfections_of_the_simple_mirror
huh, yes I suppose so. Thanks!
 

Papabravo

Joined Feb 24, 2006
18,839
I have verified numerous bugs with National Instruments, so stop the condescending attitude
I have to ask why you would continue to use a program that has "loads of bugs" when there are no cost alternatives that lack this feature?
Yes, I suppose some might consider that question condescending. Mea culpa!
 

crutschow

Joined Mar 14, 2008
30,137
I have verified numerous bugs with National Instruments, so stop the condescending attitude
I'll get right on that, sir.
But that could take awhile so, for your next problem with some "stupid" circuit, I'll let those with a more proper attitude determine where the stupidity lies. :p
 
Top