The actual discovery of the wiring problem is pretty interesting. It starts on page 17 of the Engineering Factual Report[1], which is one of many released[2].
The wire in question was carrying unknown current (likely 20-50 mA), but 110 volts DC signaling.
Comparing the photos on page 22 (Dali) and 25 (Cezanne), you can clearly see the blue ferrules in the correctly wired panel on the Cezanne.
If it happened, it wasn't preventable. If you say it was preventable, you mean that in theory, it could have been prevented by people who don't exist (front-line and management, with consciences), working processes that don't exist, using resources that don't exist, within cultures that don't exist.
They say thermal imaging could have identified the issue, does that mean that it was actually a low voltage power line, not a signal line? Do they have cameras that can see the heat from sub-mA connections?
I wonder what kind of diagnostics and logging they have for their power distribution?
> Do they have cameras that can see the heat from sub-mA connections?
If it was hooked to a circuit breaker, it wasn't a sub milliamp connection. A thermal camera might have been able to see the heat from arcing at the connection point. It could have caused a hot spot visible on thermal imaging.
According to the report, on page 7[1], it was connected to a WAGO terminal block 280-681 connector. Looking that up on Digikey[2,3] finds that it's a connector suitable for 28 through 12 AWG, up to 800 volts and 24 amps.(UL 20 Amps)
According to "3 - ENGINEERING - DALI - ENGINEERING FACTUAL REPORT"[5], page 20, the circuit was DC, 110 volts, going to a circuit breaker's "under voltage release" terminal. It was 17 AWG, according to page 23.
It apparently goes to a solenoid, and if [6] is in any way typical, it's only 2-6 watts of power in normal operation. This yields a current of 20-50 milliAmps.
Here's a nice animation of the connection failure on that wire.[4]
"That sticker, identifying the line, kept the wire from getting a good connection in a circuit breaker – which in turn ultimately caused the first blackout. "
The wire in question was carrying unknown current (likely 20-50 mA), but 110 volts DC signaling.
Comparing the photos on page 22 (Dali) and 25 (Cezanne), you can clearly see the blue ferrules in the correctly wired panel on the Cezanne.
[1] https://data.ntsb.gov/Docket/Document/docBLOB?ID=19228257&Fi...
[2] https://data.ntsb.gov/Docket/?NTSBNumber=DCA24MM031