I recently acquired the IDEMAT 2025 database (RevA7f) and noticed I was getting some anomalous results when calculating the impact from input electricity flows: The impacts seem to be off by 1000x.
When digging deeper and looking at a providing process for an electricity flow i.e. "B.046.04.101.250101 Electricity Beijing production" the input flows are in GJ, and the output GHG flows appear to be consistent with the input flows (i.e. 2.72 GJ coal would be 257 kg CO2 using IPCC multipliers).
The issue appears to be with the reference output flow which appears to be dimensionless in the output panel. Inspecting this flow further ("B.046.04.101.250101 Electricity Beijing production"), in the flow properties it appears to be pegged to 1MJ (although it also has a mass property oddly enough).
So it seems the problem is that all electricity processes are dimensionless and do not specify that the unit is actually in GJ, and as the default for the flow is MJ, when downstream processes "consume" the flow, openLCA maps the impacts as if the environmental outputs were per MJ, however in reality they are per GJ, thus leading to everything being off by 1000x.
Wondering if the behaviour is replicated on your end, and if so is there a quick fix or would it be possible to get the database updated? Found the DB quite useful for textiles in it's earlier pre openLCA iterations and would love to be able to merge it into ecoinvent etc. within openLCA!
