I designed a very basic Gilbert cell using BJT's. The function is a DSB modulator where one input is a 2.5mvpp sine wave, and the other a 5vpp square wave. The current at the bottom is a simple current source with a biasing emitter resistor. I tried changing the current from 10 to 200 mA and also the two resistors at the top to no avail of making the modulated signal higher than 25uv. What is the problem? The gain should be merely 10.
You "blindly" connected square wave sources (V3, V4) to the inputs.
The problem is that you do not define a proper commonmode level. I would try to make the commonmode voltage of V3 something around +10 V and for V3 use +5 V.
ALWAYS first make the DC solution in order (check the biasing of each transistor) before applying signals to it. Applying proper biasing is essential and a step that you missed!
Here's how I make a differential signal with a properly defined commonmode level:
Vcmm sets the commonmode voltage
Vin is the input voltage, note that if I want a 1 Vpeak-peak differential signal that I need to give Vin an amplitude of 0.5 V (not 1 V)