It varies significantly from project to project. What we depend on is the accuracy of existing records (30-60 year old microfilm scans typically, schematics and physical drawings), similarity of the failed and replacement units, ease of access to the site, support needed by other disciplines (e.g. concrete or steel work), and cost even if it is do-or-die.

A perfect like in kind replacement could possibly be done entirely by construction using proper procedure of documenting their disconnects if there were no structural modifications required to move the units in/out. This is rarely the case however and so we need to check everything from phase clearance on making the high voltage connections to whether disconnecting a failed transformer will trip out another perfectly healthy one, possibly making the situation much worse.

All that being said, we make every attempt to triage the work required and get information to construction as soon as possible so that no time is wasted waiting for engineering. We can have a conceptual layout circulating within engineering in a few hours and out to the field by the next day. Field technicians can tag and disconnect terminals right away while we figure out how to wire up the new unit. Perhaps in a perfect situation you might get all this done in a few days with all hands on deck but that's more luck than aptitude.