Context. Double-lined, detached eclipsing binaries are our main source for accurate stellar masses and radii. This paper is the first in a series with focus on the upper half of the main-sequence band and tests of 1-2 M evolutionary models. Aims. We aim to determine absolute dimensions and abundances for the detached eclipsing binary V1130 Tau, and to perform a detailed comparison with results from recent stellar evolutionary models. Methods. uvby light curves and uvbyβ standard photometry have been obtained with the Strömgren Automatic Telescope, and high-resolution spectra have been acquired at the FEROS spectrograph; both are ESO, La Silla facilities. We have applied the Wilson-Devinney model for the photometric analysis, spectroscopic elements are based on radial velocities measured via broadening functions, and [Fe/H] abundances have been determined from synthetic spectra and uvby calibrations. Results. V1130 Tau is a bright (m V = 6.56), nearby (71 ± 2 pc) detached system with a circular orbit (P = 0. d 80). The components are deformed with filling factors above 0.9. Their masses and radii have been established to 0.6-0.7%. We derive a [Fe/H] abundance of −0.25 ± 0.10. The measured rotational velocities, 92.4 ± 1.1 (primary) and 104.7 ± 2.7 (secondary) km s −1 , are in fair agreement with synchronization. The larger 1.39 M secondary component has evolved to the middle of the main-sequence band and is slightly cooler than the 1.31 M primary. Yonsai-Yale, BaSTI, and Granada evolutionary models for the observed metal abundance and a "normal" He content of Y = 0.25−0.26, marginally reproduce the components at ages between 1.8 and 2.1 Gyr. All such models are, however, systematically about 200 K hotter than observed and predict ages for the more massive component, which are systematically higher than for the less massive component. These trends can not be removed by adjusting the amount of core overshoot or envelope convection level, or by including rotation in the model calculations. They may be due to proximity effects in V1130 Tau, but on the other hand, we find excellent agreement for 2.5-2.8 Gyr Granada models with a slightly lower Y of 0.23-0.24. Conclusions. V1130 Tau is a valuable addition to the very few well-studied 1-2 M binaries with component(s) in the upper half of the main-sequence band, or beyond. The stars are not evolved enough to provide new information on the dependence of core overshoot on mass (and abundance), but might -together with a larger sample of well-detached systems -be useful for further tuning of the helium enrichment law. Analyses of such systems are in progress.