The Brazilian gravitational-wave detector Mario Schenberg was conceived in the early 2000s and operated until 2016 when it was dismantled. A straight path to evaluate the viability of the reassembly of the Schenberg antenna is to verify the possibility of detecting gravitational wave (GW) signals within its design sensitivity features. The eventual identification of significant signals would operate as motivation for the Schenberg rebuild. As the antenna was dismantled, we can get some indication from the third observing run (O3) data of the LIGO detectors. It is based on the similarity between Schenberg ultimate sensitivity and the sensitivity of the interferometers in the O3 [3150-3260] Hz band. We search for signals with milliseconds to a few seconds without making assumptions about their morphology, polarization, and arrival sky direction. The data were analyzed with the coherent WaveBurst pipeline (cWB) with frequencies from 512 Hz to 4096 Hz and the search targets only signals with bandwidth overlapping the Schenberg frequency band. No statistically significant evidence of GW bursts during O3 was found. The null result was used to feature the search efficiency in identifying different simulated signal morphologies and establish upper limits on the GW burst event rate as a function of its strain amplitude. The present search, and consequently Schenberg, is sensitive to sources emitting isotropically 5×10 −6 M c 2 in GWs from a distance of 10 kpc with 50% detection efficiency and with a false alarm rate of 1/100 years. The feasibility of detecting f-modes of neutron stars excited by glitches was also investigated. The Schenberg antenna would need at least 5.3 years of observation run to get a single detection of the f-mode signal, given E glitch ≈ 10 −10 M c 2 .