0
\$\begingroup\$

Long-haul freight trains and other systems that required an alert operator for safety would often implement a "dead-man" switch that would shut the system down if the operator couldn't actuate the switch in a timely fashion. Some time ago the military spent money on IFF (Identify Friend-or-Foe) systems to reduce friendly fire incidents. Wood workers sick of losing fingers to table saws got a clever SawStop system a few years ago: a circuit that drops the blade below the table if it senses the capacitance of a human body.

I began wondering why electricians don't have safety systems like that: that clamp and/or divert current away from a worker, or at least his heart. We do have GFCI breakers that can protect us from grounding a hot supply. But when we sit down to work on high voltage gear could we construct a circuit that breaks a power supply if it senses current across our chest? (The trick would be to avoid creating a circuit: i.e., do something better than just connecting a conductive patch over your heart to the neutral/ground and relying on GFCI/AFCI breakers, because then you'd trip the supply during contacts that, absent that excellent ground, wouldn't produce a dangerous shock.)

(This is mostly a "thought exercise" question, because I realize such a system wouldn't mitigate all electrical hazards. I guess if we wanted to avoid getting killed by all electrical risks, including things like discharging capacitors, we could wear a conductive shirt, so in the worst case we'd only get burned from points of contact up to the first good contact with the shirt?)

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Sawstop stops the blade by thrusting a sacrificial aluminum block into the path of the blade. The blade does lower, but the main safety feature is that the blade stops extremely fast. I think the descent is just due to momentum. \$\endgroup\$
    – mkeith
    Feb 29 '16 at 2:57
1
\$\begingroup\$

Not exactly. The way this is handled in every high voltage laboratory I have worked in is that the machine area is cleared of people and safety gates deployed before charging can commence, and opening any of the safety gates will abort the machine (sometimes messily - but damaging the machine is considered preferable to killing people; in practice, I don't think anyone ever was clueless enough to open a gate with a shot in progress while I worked at the various places.)

\$\endgroup\$
4
  • \$\begingroup\$ I'm thinking of less extreme machines and power. E.g., the sort of stuff you could conceivably be doing in a lab by yourself, where power may be just on the edge of what can kill you (but which almost never does), and where you may not have an attentive work partner who can cut the mains and resuscitate you. \$\endgroup\$
    – feetwet
    Feb 29 '16 at 3:56
  • \$\begingroup\$ @feetwet - you asked about 'high-voltage'. Regular mains power will kill you just fine (which is why GFCIs were invented). If playing with real high voltages, implement the safety interlocks properly. Don't try to half-ass a system and hope it will work. \$\endgroup\$
    – Jon Custer
    Mar 3 '16 at 15:11
  • \$\begingroup\$ @JonCuster - Please illuminate me. You mean any EE who's just working on, say, a flyback transformer for a fluorescent lamp, will have discharge circuits for all capacitors in place and "clear the area" before applying power to the circuit? \$\endgroup\$
    – feetwet
    Mar 3 '16 at 20:00
  • \$\begingroup\$ Well, I'd be wearing the right equipment, including rated gloves. On the other hand, the moving from here to there (i.e. replacing fixtures) was not clear in your OP. If that is the case, there is a further problem - you have to break in to the circuit in the first place to put your 'safety device' in line. An interesting trade off of what is less safe. I had read the OP as working on high voltage stuff on a bench - there it might be easier, but actually building an interlocked box would be as easy and more flexible. \$\endgroup\$
    – Jon Custer
    Mar 3 '16 at 20:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.