Context. Gamma-ray bursts (GRBs) can be detected at cosmological distances, and therefore can be used to study the contents and phases of the early Universe. The 4−150 keV wide-field trigger camera ECLAIRs on board the Space-based multi-band Variable Object Monitor (SVOM) mission, dedicated to studying the high-energy transient sky in synergy with multi-messenger follow-up instruments, has been adapted to detect high-z GRBs. Aims. Investigating the detection capabilities of ECLAIRs for high-redshift GRBs and estimating the impacts of instrumental biases in reconstructing some of the source measured properties, focusing on GRB duration biases as a function of redshift. Methods. We simulated realistic detection scenarios for a sample of 162 already observed GRBs with known redshift values as they would have been seen by ECLAIRs. We simulated them at redshift values equal to and higher than their measured value. Then we assessed whether they would be detected with a trigger algorithm resembling that on board ECLAIRs, and derived quantities, such as T90, for those that would have been detected. Results. We find that ECLAIRs would be capable of detecting GRBs up to very high redshift values (e.g. 20 GRBs in our sample are detectable within more than 0.4 of the ECLAIRs field of view for zsim > 12). The ECLAIRs low-energy threshold of 4 keV, contributes to this great detection capability, as it may enhance it at high redshift (z > 10) by over 10% compared with a 15 keV low-energy threshold. We also show that the detection of GRBs at high-z values may imprint tip-of-the-iceberg biases on the GRB duration measurements, which can affect the reconstruction of other source properties.