2015ApJ...809..183X

''Star G,'' near the center of the supernova remnant of Tycho's SN 1572, has been claimed to be the ex-companion star of the exploding white dwarf, thus pointing to the progenitor being like a recurrent nova. This claim has been controversial, but there have been no confident proofs or disproofs. Previously, no one has seriously addressed the question as to the exact explosion site in 1572. We now provide accurate measures of the supernova position by two radically different methods. Our first method is to use the 42 measured angular distances between the supernova in 1572 and bright nearby stars, with individual measures being as good as 84 arcsec, and all resulting in a position with a 1σ error radius of 39 arcsec (including systematic uncertainties). Our second method is to use a detailed and state-of-the-art one-dimensional expansion model for 19 positions around the edge of the remnant, where the swept-up material has measured densities, and we determine the center of expansion with a chi-square fit to the 19 measured radii and velocities. This method has a 1σ error radius of 7.5 arcsec. Both measures are substantially offset from the geometric center, and both agree closely, proving that neither has any significant systematic errors. Our final combined position for the site of the 1572 explosion is J2000 α = 0^h25^m15.^s36,, with a 7.3 arcsec 1σ uncertainty. Star G is rejected at the 8.2σ confidence level. Our new position lies mostly outside the region previously searched for ex-companion stars.