On 7/10/2017 8:56 PM, trader_4 wrote:
On Monday, July 10, 2017 at 11:30:42 PM UTC-4, Bob F wrote:
On 7/10/2017 8:08 PM, trader_4 wrote:
On Monday, July 10, 2017 at 9:16:55 PM UTC-4, Lenny Jacobs wrote:
On 10/07/2017 23:01, trader_4 wrote:
On Monday, July 10, 2017 at 12:14:36 PM UTC-4, Lenny Jacobs wrote:

Theoretically possible but that's not the case here. Every year, when
the raining season comes and the main breaker is tripped, I can always
trace to the same sub-breakers that cause the problem and turn them
down. They are not tripped all at the same time but they are always the
same four breakers that have the problem.


How do you identify which one caused it? Does it immediately trip the
main again when you put the offending breaker on, ie the fault is
still there? If so, if it's not intermittent, I would think you'd
immediately further track down what the source of the fault is.


I flipped all sub breakers down and flipped them one by one to locate
the offending breaker. Once I found the offending breaker, I left it
down and everything would be fine. Well, except that particular circuit
that was turned off.



I studied the breaker box and found that it's not like what I usually
saw in the US. It has a 75 A main breaker labelled Ground-First FL Main
(this is a two-story building). There are two 63 A breakers labelled
Ground FL Main and another two 63 A breakers labelled First FL main.
Looks like this is a three-tier system. It is the same 63A breaker that
is tripped during raining season.

Do you really mean tripping or that it's something on that 63A breaker that
causes the main to trip? You said the main was tripping, not the
other breaker. Also this picture is different than what I think most
of us were envisioning, which was a typical main panel with individual
branch circuits, not a main that feeds what appears to be two subpanels,
That shouldn't really matter though.


Sorry for the confusion. I had not known the system was this complex
before I checked it. There are two panes. One contains a main 75A
breaker with four 63 A sub-main breakers. Before, I had been saying
"main breaker". It turned out to be a 63A sub-main breaker.

There is another panel containing about 50 breakers (actually, this
panel has two chambers, one containing 50 or so breakers, the other
empty). There are no more panels.

What do you believe the total load normal load on it is, just before
it trips? What's going on at the subpanels when this happens,
those 63A breakers must feed subpanels of some kind.
Anything tripped there? As to why the 75A trips first, it makes
more sense now. Breakers have curves as to how fast they trip
versus the size of the fault. Different breakers will have different
curves and they are not exact. What you have is a 63A breaker and
a 75A both exposed to an overload. Additionally, you don't tell us
what else is on that main panel besides the two 63A breakers, but
if there are other circuits, then the 75A is seeing whatever the 63A
one is, plus those other loads. With other loads, no surprise it
trips first. Even without other loads, if you present a major over
load to both, the 75A might trip first if it's curve is slightly
different.

I would think main breakers would have a slightly longer response time
so to give time to sub breakers to trip.


Where are you? We were assuming this is US.

Thailand.

Gfre has brought up a few times now that you may have a main breaker that doesn't just trip on overload, but one that also trips if there is even a tiny ground fault. A ground fault is current flowing to ground instead of confined to the conductors. In the USA they are called GFCI breakers. Gfre says in Europe they are called RCD. If that's the case, then a current of just 30ma can trip it and you can get that from wet connections, outlets, etc. It's very common. Does that main breaker have anything that makes it look different? Markings? A test button?


But that wouldn't trip the main.

It would if the main is a gfci. I haven't seen that here, but he's in Thailand, who knows what equipment they have there. Even here, it would be an easy way to gfci protect a whole house. The downside would be harder to isolate the fault and losing all power. And also I think his main may actually be a sub panel.


That seems a little far-fetched.