The results of the 9P/Tempel 1 CARA (Cometary Archive for Amateur Astronomers) observing campaign is presented. The main goal was to perform an extended survey of the comet as a support to the Deep Impact (DI) Mission. CCD R, I and narrowband aperture photometries were used to monitor the $Af\rho$ quantity. The observed behaviour showed a peak of 310 cm 83 days before perihelion, but we argue that it could be distorted by the phase effect, too. The phase effect is roughly estimated around 0.0275 mag/degree, but we had no chance for direct determination because of the very similar geometry of the observed apparitions. The log-slope of $Af\rho$ was around -0.5 between about 180--100 days before the impact but evolved near the steady-state like 0 value by the impact time. The DI module impact caused an about 60%{} increase in the value of $Af\rho$ and a cloud feature in the coma profile which was observed just after the event. The expansion of the ejecta cloud was consistent with a fountain model with initial projected velocity of 0.2 km/s and $\beta$=0.73. Referring to a 25~000 km radius area centered on the nucleus, the total cross section of the ejected dust was 8.2/$A$ km$^2$ 0.06 days after the impact, and 1.2/$A$ km$^2$ 1.93 days after the impact ($A$ is the dust albedo). 5 days after the event no signs of the impact were detected nor deviations from the expected activity referring both to the average pre-impact behaviour and to the previous apparitions ones.