Hi,

I've been working on a circuit which has a differential amplifier stage. At first, i setup my circuit in the following way:

http://img526.imageshack.us/img526/9225/diffamp1.jpg

However, I'm now trying to modify the circuit so that the differential amplifier has an active load, which will make it single ended output. Here is the modification i made:

http://img402.imageshack.us/img402/1256/diffamp2.jpg

I was wondering if i have done this correctly, as i'm getting very contrasting outputs for active load circuit.

Thanks.

 

Hi,
Basically, your circuit topology is correct, but the circuit values, especially resistor values, matter. Your 100 ohm input resistors are extremely low and normally are in the 10's to 100's of KOhms. There must be some design involved, not just getting the architecture correct. Your current sink is set to 2.5mA, with the zener biased with 3.8mA, leaving only 1.3mA into the 6.2V zener, which may not be optimized, especially over temperature. You don't need anywhere near 1 second of transient data; it's a waste of time and computing power, try 50mSec. Also, at least initially, do not use 2 separate input voltages, just one Vi1, and set Vin2 to gnd. Now, also realize that with active loads, your open loop gain will be extremely high, which could be part of your problem. Try some feedback initially at least, like setting your amp to a non-inverting amp (gain=+1), by setting Vin1 to the output directly and using Vin2 as the input with a 1 volt input to start. You should see the output quickly settle to the same +1V level as Vin2. Then, knowing the amp is basically working, you can go back to open-loop tests, but remember, your gain may be 100,000 or higher, so the input must be in the microvolt region to avoid saturating the diffamp.
I hope this helps. Here is a good link for some basic analysis and design of active load diffamps. http://www.seas.upenn.edu/~ese319/Lecture_Notes/Lec_16_DiffAmpActLoad_08.pdf
Good Luck,
Kamran Kazem

 

Thanks for your help. Forgot to mention that i have changed the coding the diode so it is a 5.0V diode, as when building the circuit i used a 5.0V zenner.