| How do you propose to ground the case and transformer? So there is no touch | potential? Most of the installations I have seen the grounds are tied to the | ground block in each side of cabinet. Yep that makes them common. There is | also ground rods driven sometimes in both sides always on the secondary side | for the neutral. | Same issue as lightning protection. Why in the world would you want to tie | an ground lead from lightning protection to your electrical service? equal | potential
I don't have a specific proposal right now. I want to find a proposal that does not involve having a metallic path from MV primary distribution neutral to LV service drop messenger or neutral.
|> Why connect the grounded conductors from each side at all? The behaviour |> of the neutral blocker would suggest the answer is to pass surges such as |> lightning. |>
|> Of course I'd rather not have a lightning surge on the utility primary |> passed on to my secondary service drop any more than I'd want primary |> voltage passed on. | | Primary lightning protectors are connected to their ground locally. | | I suppose the surge could arc across inside the |> transformer and damage it. But isn't the best lightning protection to |> send it to ground instead of a customer drop? Why can't the utility |> just put surge protectors on the primary lines and connect those to |> earth ground, and connect the secondary to a separate earth ground? | | I have installed lighning protection for 20 years. I have seen that | sometimes lightning does exactly what it wants to do and not exactly what we | and or NFPA 92A planned for. | (sure hope that is the right numeber). Fortunately lightning behaves most of | the time and follows the path given.
What I know about lightning is that it substantially behaves as RF energy. It has a substantial leading impulse rise with lots of energy well into the microwave range. We know that "electricity takes the path of least resistance" is not true and that it really takes all paths, with current on each relative to the inverse of the impedance (after any breakdown where that applies). RF is the same, except that its "interpretation" of impedance varies from what would expected at 50 or 60 Hz. It can see that 90 degree bend in the wire as substantial impedance, and since it is already flowing current, produce a huge point of voltage right there, breaking down the air and arcing across to the next metal that was not connected. Look at that pole transformer as a circuit for X Hz, where you repeat that for every X from 50 Hz to 10 GHz. This is why lightning does what we don't expect, because we really can't figure out what the right expectation is.
What I am exploring right now is not better lightning protection, but rather, better MV to LV protection, while trying not to compromise the lightning protection. But I may have to get involved with exploring lightning protection, anyway, where it might be wrong but in the way of solving the MV to LV problem. I don't believing shunting lightning from the MV lines to the LV lines is proper. I believe it should go only to ground ... and that MV should go to neither (it should return on the MV distribution lines or not flow at all).