You might be familiar with passive crossover network installed inside your speaker box, consist of inductors and capacitors. The problem with passive crossover network is that they dissipate the audio power, so it’s not environmentally friendly, contributing a little disaster of global warming. Moreover, the capacitor and the inductor in the passive crossover network contribute the distortion of the signal since it must drive a low impedance load (the loud speaker). The best way to get high quality audio is to separate the high and low frequencies before feeding the signal to the amplifier. The drawback is that you need two amplifiers, to amplify the low and high audio signal separately. Here is the schematic diagram of the active crossover circuit:
The circuit use a constant voltage method, means that the output of high and low frequency is summed up, and then fed back to be compared with the input to make sure this sum is equal to the input signal. This method ensure that the total response is flat, if we summed back the separated high and low frequency output.
You have to use 1% tolerance for the resistors to give a precise response. According to the formula, the capacitor C should be 6.6nF to give 1kHz crossover frequency, but a 6.8nF can be used because it waidely available, and the crossover frequency will be shifted to about 975Hz. [Circuit’s schematic diagram source: National Semiconductor Application Notes]