The recoil-β tagging technique was used to identify transitions associated with the decay of the 2 + and, tentatively, the 4 + excited states in 74 Sr. Combining these results with published data for the A = 74 isobars, triplet energy differences (TEDs) have been extracted, the heaviest case for which these values have been evaluated. State-of-the-art shell-model calculations using the JUN45 interaction and incorporating a J = 0 isospin nonconserving (INC) interaction with an isotensor strength of 100 keV can reproduce the trend in the TED data, with particularly good agreement for the 2 + state. This agreement for the TED data taken together with the fact that agreement has also been shown between shell-model calculations with the same strength of INC interaction in the f 7/2 shell and recently for A = 66 strongly suggests that such an interaction exists throughout the nuclear chart and cannot have a strong dependence on details of nuclear structure such as which nuclear orbitals are occupied. It also supports the hypothesis that only a J = 0 component of the INC interaction need be included to explain the observed TEDs. Isospin symmetry is a central concept in nuclear physics, based on the idea that the proton and neutron are two isospin states of a single object, the nucleon. This would imply an identical pattern of excited states across a T = 1 isospin triplet. These symmetries are broken due to a number of nuclear structure effects, most importantly the fact that the proton carries an electric charge while the neutron does not. When exploring such breakdown of symmetries, it is instructive for the study of isospin triplets, to evaluate triplet energy differences (TEDs) as a function of spin J defined bySuch isotensor energy differences are essentially independent of single-particle effects, which makes them particularly simple. Since contributions involving Coulomb effects are readily calculable, TEDs are particularly sensitive to additional terms such as isospin nonconserving (INC) components. It has been shown that such a component is necessary to reproduce experimental mirror-energy differences (MEDs) in the f 7/2 shell [1]. Indeed, recent work identifying the (4 + 1 ) and (6 + 1 ) excited states in the N = Z − 2 nucleus 66 Se [2] has allowed the evaluation of the TEDs for A = 66, the first case above the 56 Ni closed shell; and it appears that an INC term is also mandatory in reproducing the TED. Nucleon-nucleon scattering data have previously been used to infer the INC component to the nuclear interaction, the isotensor component of which is consistent with that previously observed in the f 7/2 orbital [3]. The measurement of such a component in-medium is of considerable interest, and it is not yet clear whether the INC component arising from energy-difference measurements is indeed the same as that determined from scattering data. It is of high interest to extend such studies to higher masses, where the structure of the low-spin states involved is very different, evolving from dominance by fp orbitals...