RR Lyr is one of the most studied variable stars. Its light curve has been regularly monitored since the discovery of the periodic variability in 1899. Analysis of all observed maxima allows us to identify two primary pulsation states defined as puls
ation over a long (P0 longer than 0.56684 d) and a short (P0 shorter than 0.56682 d) primary pulsation period. These states alternate with intervals of 13-16 yr, and are well defined after 1943. The 40.8 d periodical modulations of the amplitude and the period (i.e. Blazhko effect) were noticed in 1916. We provide homogeneous determinations of the Blazhko period in the different primary pulsation states. The Blazhko period does not follow the variations of P0 and suddenly diminished from 40.8 d to around 39.0 d in 1975. The monitoring of these periodicities deserved and deserves a continuous and intensive observational effort. For this purpose we have built dedicated, transportable and autonomous small instruments, Very Tiny Telescopes (VTTs), to observe the times of maximum brightness of RR Lyr. As immediate results the VTTs recorded the last change of P0 state in mid-2009 and extended the time coverage of the Kepler observations, thus recording a maximum O-C amplitude of the Blazhko effect at the end of 2008, followed by the historically smallest O-C amplitude in late 2013. This decrease is still ongoing and VTT instruments are ready to monitor the expected increase in the next few years.
We used the GEOS database to study the Blazhko effect of galactic RRab stars. The database is continuously enriched by maxima supplied by amateur astronomers and by a dedicated survey by means of the two TAROT robotic telescopes. The same value of th
e Blazhko period is observed at different values of the pulsation periods and different values of the Blazhko periods are observed at the same value of the pulsation period. There are clues suggesting that the Blazhko effect is changing from one cycle to the next. The secular changes in the pulsation and Blazhko periods of Z CVn are anticorrelated. The diagrams of magnitudes against phases of the maxima clearly show that the light curves of Blazhko variables can be explained as modulated signals, both in amplitude and in frequency. The closed curves describing the Blazhko cycles in such diagrams have different shapes, reflecting the phase shifts between the epochs of the brightest maximum and the maximum O-C. Our sample shows that both clockwise and anticlockwise directions are possible for similar shapes. The improved observational knowledge of the Blazhko effect, in addition to some peculiarities of the light curves, have still to be explained by a satisfactory physical mechanism.
