News for hackers. Stuff that matters.
Re:Severely injured worker sues Tesla by sokoloff on Thursday May 24, 2018 @06:27AM (#17142660)
>>> Tesla should have cut electricity to the equipment he was working on, but refused to because the company didn’t want to temporarily stop production.

>Umm, WTF? That's insane at mains-level voltages.

At mains-level (240V and below), working on live systems is fairly commonly done. I've taken 120VAC hits dozens of times and 240VAC a few times. It's not comfortable (and has an outside chance to be fatal, but extremely^2 rarely is) and I try to disconnect power prior to working on something, but sometimes that's not possible (such as troubleshooting power supply components or other PCB work that needs to be done hot).

The most uncomfortable I've been at work was supervising (as a customer) a UPS commissioning and testing when the contracted electricians were making the 480V/3 connections on live circuits to cutover to the new feeds. To us, we didn't need the load carried live; the servers and other equipment was on only to provide a representative load for commissioning testing and phase balancing.

Electricians didn't want to wait for us to shut everything down to take the last step (even though they were no doubt making time-and-a-half on a Saturday afternoon), so they decided to cutover live-live. They acted like they did it all the time. Seemed insane to me, but that's still a modest voltage and from subsequent inquiry, this is apparently very common.

Agreed that actual high-voltage systems are different from mains-level, of course.

Re:Severely injured worker sues Tesla by eigenvector on Thursday May 24, 2018 @06:09PM (#17148922)
We need to distinguish between shock hazards, which are proportional to voltage level in terms of the proximity required to cause a shock, and arc flash hazards which are proportional to the available current and the speed of the nearest upstream protection device.

Low voltage systems are often much more dangerous for arc flash than high voltage ones: the current is often higher (thousands of amps instead of tens of amps) and the protective devices (circuit breakers or even fuses) are slower. That means the amount of hazardous energy discharged into a worker before the circuit is automatically isolated can easily be enough to maim or kill even at relatively low voltages such as 120V. There's a 208V transfer switch in one substation I work at that has potential arc flash incident energy of 38 cal/cm^2. That's more than enough to kill even from 2 metres away.

At 480 or 690V it is sufficient to simply not touch energized equipment to avoid being shocked, and PPE is available that allows hand-contact (class 0 gloves). However, available arc-flash energy in the event of a fault while the energized equipment is exposed can easily be above the level that any PPE can protect from (usually considered to be 40 cal/cm^2). If the arc flash hazard is at a level that cannot be mitigated with PPE, it will have to be reduced in some fashion either by re-configuring the circuit or de-energizing it before work can be done.

Employers are responsible for performing engineering studies to determine the arc flash hazard level at every relevant location in their facility, labeling it appropriately, training employees on these hazards and providing adequate PPE. Unlike voltage, you can't just look at something and tell how much arc flash potential it has which is why engineering studies need to be done in advance. We have widely-accepted standards such as NFPA 70E and IEEE 1584 that codify the appropriate methods for engineers to use in estimating these hazards, and there's no excuse for failing to do that.