I have questions about tripping breakers (or blowing fuses). For the purpose of this post, we'll consider mainly breakers.

We'll consider a given load, a 400W M59 ballast and lamp. The circuit is 15A, 120V, 60Hz, fed from a panel that is, for reference, 200A. There may be more than the one 15A breaker in said circuit (such as a sub panel, etc).

Here it goes : in my mind, the following statements are true :

1.  Load is fed directly from the 15A breaker in the main panel. Load fault causes 15A circuit breaker to open, 200A main breaker stays closed.

2. Load is fed from 15A circuit in sub panel, which is fed from a 15A breaker in main panel. In essence, there are two 15A breakers, one downstream from the other. Load fault causes 15A breaker from sub panel to open, 15A breaker in main panel stays closed, as well as the 200A main breaker.

3. Load is fed directly from the 15A breaker in the main panel. An inline fuse rated for less than 15A is in luminaire. Load fault blows only the inline fuse in the luminaire, leaves the 15A breaker closed and the 200A main breaker closed as well.

I am quite sure statements 1 and 3 are true, however I doubt my understanding of statement 2. What do you think ?

I've wondered this same thing too, my house is wired in such a way that both branch breakers and the main are 20A.  Yeah, small, but the panel is fed with #12 so I can't safely upgrade it anyway without replacing the feeder wire, and it's an off-grid home run off a generator without an electric dryer, water heater, electric baseboard heat, etc. just lights and a few appliances.  Never tripped it once.

Breakers are made to trip within some time frame for a current value. For example, here is what a European standard "Type C" breaker behaves like :



Lets say the breaker is 10A



With somewhat exceeded current (typically overload or "weak short circuit" with bad contact and not very high current), it trips by the heating of a bimetal strip. The strip takes time to heat up, the higher the current, the faster it does. So the breaker won't trip immediately

For 10A breaker :

At up to 11.3A, it will never trip

At between 11.3 and 14.5A, it might or might not eventually trip, but anyway it will stay on for long time

Above 14.5A, it will definitely trip sometime..

At 15A it will trip in 1 min to 9 min

At 30A it will trip in 3 sec to 15 sec

At a little under 50A it will trip in 1 sec to 5 sec

At up to 100A, if the electromagnetic trip does not activate, it will trip in no more than 2 sec



With greatly exceeded current (typically "good contact" short circuit), the electromagnetic trip is activates and trips the breaker within about 1 AC cycle (though in marginally low currents, where the electromagnet only barely activates, it might buzz for a moment before tripping)

For type C breakers, this is between 5x..10x rated current. So, the 10A breaker will never "magnetically" trip at 50A or less, and will allways "magnetically" trip at 100A or more. What happens inbetween is up to the individual breaker, in some the electromagnet reacts to lower current threshold than in others. So, for a 80A one breaker might allready trip magnetically and another might still wait a couple seconds until the bimetal heats up



When breakers are added in series, there is some current in all of them

supply ------------/200A/------/15A1/------------/15A2/------------ light

If the only load present is the light, then its the same current in all breakers. If there are more loads connected, then there is some more current going in the 200A one and if something else plugged on the 1st 15A circuit, then some more current going through the 15A1 one too

If currents are within the "thermal" range of breakers, they heat up untill something trips. If for example the current is 50A through all breakers, then both 15A are heating up. One of them will trip sooner (differences between breakers, difference in ambient temperature in the rooms where the panels are). When it trips the current is stopped, so the other one stops heating and won't trip. For most purposes, it can be assumed that which breaker will trip is random. However, it probably will be allways (or mostly) the same one if you try again with the same breakers

If currents are within the "definitely magnetic" range of the 2 15A breakers (short circuit on a long branch circuit), electromagnets pull at the same time, and the mechanics of breakers start to move at the same time. I dont know about USA breakers, but with European ones usually both will be tripped, less commonly only one, and if one then it is random which one. But if the current is not too high, it is still in the "thermal" range for the 200A one so that one will stay on

If currents are within the "definitely magnetic" range of all 3 breakers (good connection short circuit between thick conductors), all 3 electromagnets pull and the mechanics start to move. It is possible to find all 3 breakers tripped

If currents are within the "maybe magnetic" range of the 2 15A breakers, it is possible that one will trip magnetically and one not



US breakers are more like European "Type D", where magnetic activation is in the 10x..20x range instead of 5x..10x. But otherwise the logic is the same..



A fuse have its blow time chart too, which more or less resembles the "thermal" part of the breaker's time chart, so the same logic applies. However, i think it is possible for a fuse to start to melt before the breaker trips and "saves it", in which case the fuse is going to blow faster or at lower current the next time



Arranging the system so the main breakers dont trip when the situation can be handled by the small ones is called selectivity. There are some considerations to it :

As a rule, it is more predictable what will happen in the "thermal" range of the breakers, as it takes time to heat. In the "magnetic" range everything happens too fast, and mechanics start to move at the same time before any contacts open

Estimating the max short circuit current available in a sub circuit. It is limited by resistance of the wires of the circuit (up to the point of short circuit). If the current is not very high (within 100's A), it is quite possible that it will be in the "thermal" range for the main. So you got what you wanted

Using breakers of different types. If you use Type D (10x..20x) for the main, it means that currents will have to be higher to trip the "magnetic" of the main. In the sub circuits you want faster breakers of Type C (5x..10x) or Type B (3x..5x). However, you should be carefull with the "high" types (D) since you extend the range of currents at which the breaker will not trip "magnetically" for a "not very strong" short circuit. Generally, Type D are not to be used in a home system in Europe

Using more advanced breakers, where there is better control over the timing in all current ranges. Those are bigger breakers, usually used in industry and for main in big industrial/commercial site panels

In my set up

I live in a RV. Got  30 amp -------30. Amp-----  20
                                                                        15
                                                                        15
                                                                        15

120 volt circuit.      When i hace my Aircon running
It pulls 16 amps, then refrigerator 3 amp running,
5 amp in lighting,  about 3 amps in voltage conversion
I was running at the limit,  then I plug in my 100 watt MV
In . About 5 mins later i started noticing lights were flickering a bit,  like breaker go should I trip or dont trip.
So i knew i getting into tripping point. I hurry unplug MV
Before i send every thing into compleat darkness.. I will
Get the voltage converter off soon as get my solar panel instelled and my battery bank.  That way i can free up a few amps on 120 volt side.

Having a 15A breaker behind another 15A breaker (of the same speed category) is nonsense - you never know which one will trip and which won't, so you have to always check both.
And when the load has a high surge inrush current (like tank capacitors in computers or so), you loose the possibility to restart the electricity by gradually charging the capacitors by repeating the turn ON (it is quite dirty procedure, but many times it is working).

Although it may seem as still safe from solely electrical perspective, it does cause frustration and stress - and that is by far not safe situation  at all (you will way more likely make some real severe mistake, yielding to really dangerous harm; an example would be rushing up and down a fly of stairways between those two breakers, falling on them, yielding severe injury).

This is a problem not limited at all to breakers of same rating, but to pretty much all MCBs and RCDs that go on a DIN rail

This is a problem i see in many installtions where there is a panel and a sub panel :

 - Standard requires every panel to have main breaker

 - Power company decides what is the max current value you can use. Lets say 40A

 - Allthough not expressly prohibited, it is frowned upon to use D breakers in domestc circuits

So, your main breaker is a C40, and thereis no selectivity - EOL of an Incandescent lamp on a circuit in the sub panel will trip the B10 there, C25 main there, C32/C40 on the line feeding the sub panel in the main panel, and C0 main....



The case with 2 identical breakers (and for me it would be 16A) would be expected in a different story :

I use some power tools in my work. Drill, Grinder, you name it

In some sites where i want to plug the tools in, i dont trust the electrics. So i made a device like plug-----box-----receptacle, where in the box are MY breaker for 16A, RCD for 30mA, and some lamp indicators to check whether the upstream receptacle wiring is not screwed up completely

If i make a short or earth fault on this device, it might trip, but also the upstream might trip...

Interesting to read the time it takes for them to trip!  I know some combos that will make some American breakers trip:

20 amp...a 1500w space heater and a 20 amp electric pressure washer.  Start pressure washer, instant breaker trip it seemed. 

Electric skillet and other heating appliances= slow trip on 15A. 

At my worksite have Powell Electric MV remote trip circuit breakers.These are for 4160V 3Ph,one shot.One of them tripped last we and meant it!The breaker controlled the 4160V primary to one of our building 4160V-208/120V transformers.One transformer in the bank has a shorted primary.So now  generator temporarily replaces the transformer bank that failed.Now the plant is taking bids on replacements.Will see how this goes.Figure won't be anything new until after the holidays.In the meantime the Sunbelt "Whisper-Watt" generator does its job!

The term you are looking for is coordination. When you are working on coordination of overcurrent devices you need to consider fault current and the curves of the breaker. In your examples I will examine a few fault scenarios these all assume the main breaker is not near rated load.

Example 1, Short Circuit fault:
The 15 amp branch circuit breaker will open due to a trip of the magnetic (instantaneous overcurrent) mechanism breaking the fault. In addition the fault current will start warming the thermal mechanism of the main breaker (time overcurrent), but not activate the instantaneous overcurrent part of the mechanism. Since the 15 amp breaker opened before the time overcurrent trip characteristics of the main were reached, it remains closed.

Example 1, Overload:
Overloading a circuit usually only activates the thermal part of the breaker which is the time overcurrent mechanism. For a 30 amp (200% overload) it may take the 15 amp breaker a few minutes to warm up and trip. Since no overcurrent condition exists at the main breaker it has no effect.

Example 2, Short Circuit:
In this example the 15 amp breaker to the subpanel and the 15 amp branch breaker both trip due to the instantaneous overcurrent condition. Since the trip time is delayed by the speed of the trip mechanism the differences in the breakers trip curves should not have any effect. The effects on the main breaker are the same as example 1, short circuit fault.

Example 2, Overload:
In this example we have to consider the load on both breakers. Assuming a 200% overload and equal loading on both breakers (no other branches in the sub panel are loaded), either breaker or both can trip. Now if the sub panel breaker has a higher load (say 5 amps plus the 30 amps overload) the sub panel breaker has a higher probability of tripping first. It is also possible that both breakers will trip out. The depends on the manufacturing variations in the breakers. In either case the main breaker does not come into effect.

Example 3, Short circuit:
This is VERY Dependent on the fuse curve. Assuming a typical fast acting fuse it is possible that the device fuse will blow prior to the upstream circuit breakers (main or branch) tripping. It is still possible that the fuse will actually open at the same time as the branch circuit breaker opening (breaker has to move it's mechanism to open, fuse does not). If a slower blowing fuse is used, the possibility of a simultaneous opening of fuse and breaker increases. The possibility of a slower blowing fuse staying intact while the upstream branch circuit breaker trips comes into play. The action of the main circuit breaker is the same as all short circuit examples.

Example 3, Overload:
This is also dependent on the fuse curve and becomes very tricky with inductive loads. In addition, the fuse only operates if the device it is protecting has an overload. A fast acting fuse will probably blow under normal conditions with an inductive load. A slower blowing fuse may not open prior to the circuit breaker unless the time overcurrent characteristics of the fuse show otherwise. In any case if the fuse opens due to an overload and the branch does not, that means the branch circuit breaker did not warm up enough to trip, as with all overload examples the main is not involved. If the circuit breaker opens instead, it is identical to example 1.

This does not go into all possible cases, obviously. Special industrial information was left out (protective relays, current limiting fuses, etc).

Thanks jrmcferren.

So this brings two other questions :

1. In a very large room where it is desirable to have individually fused luminaires, how does one coordinate the operation of a fuse should a fault occur in a luminaire ? In other words, how do you augment the probability of the fuse operating before the branch circuit breaker ? (so a large area won't be dark if a fault occurs).

2. It is common for European members to indicate that an EOL incandescent lamp sometimes trips the circuit breaker and any other breaker upstream up to and including the main breaker. Is this due to breaker curves or is it due to some houses having a low amperage service (~40A) and the over current caused by the arcing is enough to operate the main breaker ? Most houses here are 200A with a few older installations being 100A. Never heard of an EOL lamp tripping the main breaker.

European C10 breaker (standard) vs Glass fuse for couple A or less : The fuse will blow virtually allways, the breaker trip sometimes. US breakers are for what i know equivalent of European D breakers, so it would have atleast the same tendency to allways blow the fuse

This means, that if there was a short circuit and the breaker is tripped, the fuse is allready blown, and you are not resetting the breaker into a still existing short circuit

But this still does not mean its good to connect all lighting on the same circuit :

 - Some short circuits happen where the cable enters luminaires, where it is cut against a sharp edge. That is, before the fuse thats in the luminaire

 - When repairing the luminaire that shorted, you would like to have light from some of the other luminaires



The tripping main from Incandescent EOL is from the arc in the lamp. The arc is pretty good conductor, i think the resulting short circuit is on par with what would happen if you firmly connect the filament supports together directly

There is a fuse in the lamp base - basically a section of bare thinner wire in one or both wires going into the seal, probably fast acting for 1A or so. Sometimes when this fuse blows, the arc moves from being between the fuse wire ends to being across the 230V inout wires inside the cap, so making a 2nd short circuit immediately (milliseconds ?) after the 1st, this time without a fuse in the way. The tripping of breakers happens regardless of whether there was or wasnt second arcing in the cap

When the lamp shorts, the short circuit current is determined by the impedance of the wiring up to the lamp. This means easily 100's A. Thick cables in the supply to the house are exactly the reason why the current can be that high - the only significant resistor remaining is the length of #14 circuit between the breaker and lamp, and #18 flex if the lamp is plugged into the wall

The most common main breakers are between 25 to 40A, with magnetic trip ranges between 100..125A (L25, L = old type, equivalent of modern type B) to 200..400A (C40)

To activate magnetic trip on a 100A+ main it'd take currents in the few kA range, the length of circuit from the panel to the lamp would probably limit the current below that

The main breakers of US panels are big, is it possible that they move slower (bigger parts inside with more inertia) ?

And at 120V perhaps Incandescents dont arc as severely

Here (Czech Rep.) the only allowed breaker type homes is "B" type, so pretty fast.

The only way to ensure the required selectivity is to make the short circuit current below the electromagnetic trigger of the upstream breaker. But that is possible to ensure only with fusing inside of electronic ballasts or
other similar devices (using the fusible resistors in place of the fuses - the resistance limits the fault current to 10..20A rage), but these protect only against faults within those ballasts.

With shorts in the installation wiring, it is practically impossible to guarantee the selectivity.
But what helps to make the main breaker to trip less likely is using really the minimum rating breaker and wiring gauge for the branches (so e.g. B6A with 1.5mm^2 for lights, with the B25A main). The thin wiring ensures there is enough resistance to somewhat reduce the current (lower the current, smaller forces in the electromagnet in the main breaker, so it takes longer time to activate it, plus lower current means the arc in the branch breaker extinguishes sooner; so the branch breaker has chance to really open the current paths sooner). educing the short circuit current from ~3..6kA to something around 1kA or below A means there is very high chance just the "6A" branch one will trip and keep the "25A" main closed. But it is not guaranteed...

In an industrial environment are used special devices, which limit the fault currents. But these are expensive, bulky and mainly designed for way higher current levels, their main usage is when the wiring impedances are so low, the short circuit current exceeds the current breaking capability of common breakers (6..10kA for the types used at homes). I'm not aware about any product designed for lower currents...

How does fault current in the 100's A range ensure a B25 stays closed ?

Isc goes at the same time through B6 and B25 breakers

It is sufficient to pull in the magnetic core (that will release the mechanism) in both breakers

The breaker tripping mechanism : core pulled into solenoid. Core pushes a pin from the other side. Pin pushes a tab on the release mechanism. Hook holding the mechanism is released. Spring starts contracting. Spring tension on the contacts ended and they start moving apart. Arc is formed. Arc is pushed into extinguishing chamber. Arc is extinguished

By the time the "faster" breaker breaks the current, it is likely that the other breaker allready had enough time to release the spring as well, so it is on the way to open the contacts too even if they are still closed in the 1st moment

All breakers from the same range have practically the same mechanic (springs tension, contact mechanic,...), they differ just in number of turns of the fast acting and the heating resistance of the slow acting trigger.
The 25A breaker has less turns on that solenoid than the 6A one. So the same current causes less force.
And less force means less acceleration to the mechanism mass, so it takes longer till it reaches the "no return" point.
As the 6A has 4x more turns, the tripping gets way faster (about 16x higher acceleration), so it releases the mechanism 16x sooner. Then the mechanism act on its own (spring driven; that part does not depend on what caused the trigger).
Because the current is lower, the arc gets quenched very fast, way faster than these devices are rated for (they are rated for 6 or 10kA currents, 100'A is nearly 100x lower).
Because of the inertia causes the 25A trigger to accelerate so much slower, there is high chance the current disappears (by the 6A one) before it trips.

But as I wrote, it is just high chance, it is by far not guaranteed...

Back when we used Incandescents, an EOL one would take out both L10 and L32 breakers (old panel with "L" characteristic breakers)



And is B really the only type allowed for homes ?

We have a 2.5hp air conditioning unit which working current is 10A and locked rotor current ~80A. It was installed in the 90s when i was little kid, with a L25 breaker (as i understand L being the old equivalent of B ?). With 2.5mm^2 cable

Eventually i found this out and replaced the L25 with a C16 - as i thought 25 is too high for 2.5mm^2, and 16 is sufficient for the unit. Besides, we had a locked rotor condition on there couple times as parents abused the thing (switched it on when the outdoors coil is frozen), and the C16 tripped within couple seconds every time. The L25 probably would not

What breaker would be used in your country for a unit like this ?