I'd like to toss out a question for general comments. I'm a PE in a practice where I'm the only PE, so I really don't have anyone in the shock hazard/Arc Flash space to bounce things off of and I'd really like some contrarian viewpoints on this. After the engineering hubris associated with the "Oceans Gate" submersible debacle, I've been REALLY worried over someone falling victim to my own hubris. So here's the situation:

I'm working with a client who assembles battery assemblies (Lithium). I don't want to go into who the client is or what the application is, but simply that it's a battery assembly using multiple individual battery cells. I've done the arc flash assessment using the "R.F. Ammerman, T. Gammon, J.P. Nelson, and P.K. Sen" methodology. Now I'm trying to 1) assess the actual risk of such an event and 2) what is the shock hazard risk? A little background.

The batteries that go into the assembly have covers over the positive and negative terminals, that prevent contact during handling. When put in position in the assembly, they are connected with insulated bus sections, one connection at a time. The worker exposes the positive terminal on one cell, and the negative terminal on the adjacent cell, inserts the bus, tightens the nuts, and closes the covers. Since there is no completed circuit, the voltage across those terminals is always zero, or nearly so. We've measured the voltage between two partial strings, and we've gotten 3 to 4 volts for for a time of <100 milliseconds (using a peak hold function of Fluke Multi Meter) because of the capacitive charging current that flows during connection between cells. Given the input impedance of a Fluke meter (~10M ohms), we aren't talking very much current (~0.4 micro amps).

My conclusion is to NOT require shock hazard protection during assembly of these strings. In order for there to be a shock hazard, there would have to be a short circuit across the design open for the assembled string. That has a cover over it. I can't conceive of a way that the circuit could be completed during this assembly (to result in an actual fault between string sections). Even if two cells had an internal fault, the case is non-conducting, so that fault would not be to ground. I'm thinking of recommending an 8 Cal/cm2 shirt and gloves simply because 1) I can't imagine NOT recommending flame resistant clothing when working around live electrical components and 2) those items don't impose a significant efficiency burden on the assembly process.

I'm looking for some push back here. Is there any way I'm wrong? Am I over thinking this?

Thanks in advance!! - Julie VanDyne