| Actually, quite often an in-line resistor/shunt is *not*
safer. If the
| primary circuit is high-voltage, the leads used to sense the voltage drop
| across the shunt will also be at line voltage. One of the advantages of
| CT's is the secondary circuit is *low*
voltage (<0) and is easier to run.
Which is why my plan is to build the unit to run entirely isolated, with an
optical output of some kind, possibly as simple as a pulse modulated LED, or
at best a fiber optic cable.
| True, it is 'practice', not 'theory'. The 'practice' is that most
| commercial metering use 5 amp current circuits. A typical kW or MW meter
| will be designed for 120V and 5A at the meter terminals. So most PT's are
| ratio of <line-voltage> : 120 and CT's are <line-current> : 5
Which makes perfect sense now that the practice is known.
|> My question was one of practices ... a politics category. And apparently
|> one that some engineers want to keep others from learning about ...
|> to protect their jobs.
| Occam's razor. Don't assume nasty conspiracies when the simpler answer is
| you just haven't made your question clear enough for people to understand.
That could be. Too often people are put off be highly detailed, precisely
expressed questions (they get tire of reading it after 100 words or so).
So I more often try to be brief ... too brief in some situations.
| You don't. You have to make the initial installation with the system dead.
| As part of the initial checkout, you verify the shorting links work
| correctly. If you have to work on the metering, you either kill the
| circuits, or trust that the shorting links still work correctly. If you
| have any doubt about the shorting links, you don't trust them and have to
| reschedule the work when the system is dead.
My trust in technology is limited when it can kill. Things break. Things
come loose. Things don't have expected capacity, etc.
| If the shorting links are separate from the meter and normally open, you
| should see the meter reading drop to near zero when the links are closed.
| If the reading doesn't change, then obviously the link isn't shunting
| current away from the meter. Some meter cases have 'finger blades' that you
| pull to disconnect the meter and short leads at the same time. So when you
| pull the 'blade', the meter goes to zero, even if the shorting fingers fail.
| These types of cases have been around since Westinghouse, shorting-link
| failures are very rare in this type of unit (after all, it isn't rocket
| science, they designs are simple and robust).
Still, a failure can kill. I think we can accept that technology is not 100%
perfect. Goals are in terms of MTBF and percentages, etc. Life is just once.
| Some CT's have a thin (mica?) member between the secondary terminals. If
| inadvertantly open-circuited while energized, the voltage spike will
| 'puncture' the insulation member and allow an arc to jump across the
| terminals right at the CT. Have to replace the CT to repair it, but the
| high voltage won't be sent through metering lead damaging other components
| (or people). Not *all*
have this feature.
Sounds like a good idea.
| Phil Howard KA9WGN | http://linuxhomepage.com/ http://ham.org/ |
Click to see the full signature.