simulation of a gilbert mixer

[Ashtaroth]

hello

i want to simulate a double balanced gilbert mixer with pspice, my questions are:

1) how can i make a DC analysis, i mean what kind of signals to you apply in the RF and LO input?

2) In the Ac analysis what signal do we apply in the LO input???

thanks

 

[flatulent]

The circuit has to be biased properly to operate. The simulation should have the same values as bias you are going to use in the real circuit.

Your AC analysis will not give you results in the mixing action. You will have to use time domain analysis. You can unbalance the LO port with a DC offset to get the gain from RF to output.

As a matter of history, this circuit goes back to about 1950 when Gilbert was wearing diapers.

 

[Ashtaroth]

thanks
now i understand why i got strange results in the Ac analysis. but how much DC Offset. is there a way to calculate it or just use a typical value for example 1Volt.

in the DC analysis the circuit is biased properly. i am just asking about the RF and LO signals. should I apply them, short circuit the RL and LO input????????

 

[flatulent]

Try using the DC value equal to the peak of the AC value you plan to use for the LO. This will give you an idea of the "gain" of the mixer. This way you can try different passive component values and bias currents and compare them on an equal basis.

Use 0 V on a port to get an idea of the feed through of the signal on the other port.