我要设计一各single-balanced Mixer using CADENCE, RF port is 1.8GHZ (single input) LO port is 1.7GHZ (differential input) IF port is 100MHZ (differential output) I use a port cannect to the load as the IF port. (both port terminals connect to the differential load) I use one port in the LO level MOSFET(M1) and set the input 2V peak-to-peak. I use one port in the other LO level MOSFET(M2), still 2V peak-to-peak, but 180 phase shift to make LO input differential. Both LO ports connect to GND. I use one port in the RF level MOSFET(M3) with 1mV(-60dB) RF input signal. M3 is the tail current source for the Mixer. So, I use PSS to see the Gc(voltage gain from RF to IF) Original, I set the width of M1 and M2 380um (M2 is also 380um). I got the amplitude at IF~=-55dB(100MHZ), so Gc~= 5dB. However, it does not match my hand calculation. gm(M3)~=30mA, RL=1K, so Gc should be Gc=gm3*RL*2/pi~=25dB. So, I change the width of M1 and M2 to 100um. I got Gc~=17dB.(I got IF~=-43dB at 100MHZ). However, I think it makes no sense, because when I decrease width of M1 and M2, I am decreasing gm2 and gm3, so that decreasing the speed of turn on-off time of M1 and M2. It should result in lower Gc, but I got higher Gc! Is there anything wrong or I connect something wrong? Or, it makes sense? If yes,why?? Actually, I do not know how to choose the width of M1 and M2. I really need help~ Thank you so so much! --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 69.231.125.135