Buzzing from my circuit breakers???

[gearheads78]

I have been slowly been getting my new shop wired. I got the overhead lights done which consist of 2 15 amp breakers going to 8 each 105W CFL bulbs run with #12 wire. When I first turned them on I heard faint buzz and I thought it was from the CFLs but I got my ear close to the breaker box which is still uncovered and noticed the noise is coming from there. It does it on both circuits. I can turn one or the other off and it gets quieter but I still hear it. I can ever push on the breakers and the pitch of the sound changes slightly.

This the box I am using from Home Depot if it makes any difference. Do I have something to worry about?

http://www.homedepot.com/h_d1/N-5yc...Flow=3&catalogId=10053&langId=-1&ddkey=Search

 

[mrb]

its likely the CFLs. Without spending an hour discussing how they work, they dont draw power constantly like an incandescent lamp does. They draw their power in spikes. Cheaper CFLs draw their power in shorter, higher current spikes. I have seen 13w CFLs that say not to put more than 12 on a 20 amp circuit. Try unscrewing half of your bulbs on one circuit and see if the buzzing diminishes. If you have a clamp on ammeter with a peak hold function you will probably see 40 or 50 amps on that circuit. The noise is the magnetic trip coil rattling the contacts. It will eventually wear the breaker out and it will start tripping.

 

[mrb]

a little research shows these large CFLs typically have a power factor of 0.5-0.6 which means that the current draw is 14 amps @ 120v considering power factor of 0.5 (105w/0.5pf)/120v = 14 amps thus your circuit is overloaded. If its all 12ga wire then replace the breakers with 20 amp ones (you might consider using 20a switches as well)

note to all: when you want alot of light, use T8s with good ballasts with a pf of 0.9 or better.......

 

[gearheads78]

a little research shows these large CFLs typically have a power factor of 0.5-0.6 which means that the current draw is 14 amps @ 120v considering power factor of 0.5 (105w/0.5pf)/120v = 14 amps thus your circuit is overloaded. If its all 12ga wire then replace the breakers with 20 amp ones (you might consider using 20a switches as well)

note to all: when you want alot of light, use T8s with good ballasts with a pf of 0.9 or better.......

So 8 105W bulbs are drawing 14amps?

 

[mrb]

So 8 105W bulbs are drawing 14amps?

somewhere around that depending on the power factor of the particular model you have.

for explanation look here http://en.wikipedia.org/wiki/Power_factor and scroll down to non-linear loads. the picture of the oscilloscope screen shows you what is going on.

realize this extra current draw isnt costing you anything, the lamp is still using 105w (plus small ballast loss)

 

[theoldwizard1]

When I first turned them on I heard faint buzz and I thought it was from the CFLs but I got my ear close to the breaker box which is still uncovered and noticed the noise is coming from there.

The buzz is coming from inside each individual breaker.

From Wikipdia - Circuit Breaker

A breaker uses a solenoid (a type of electromagnet) to release a spring which opens the contacts. ANY electromagnet (connected to an AC circuit) has the potential to "buzz" (mechanical vibration), if some of the components are not firmly seated (next to impossible). All transformers buzz.

 

[theoldwizard1]

...Without spending an hour discussing how they work, they dont draw power constantly like an incandescent lamp does. They draw their power in spikes.

To the best of my knowledge (which is limited), any "electronic" fluorescent "ballast" is essentially a constant current power supply. (Fluorescent bulbs are odd devices; once started they will continue to draw more and more current until they burn out UNLESS there is a device (ballast) that limits the amount of current they can draw.)

All modern "high efficiency" power supplies are "switchers". Simple put, they switch on and off at high frequency (> 400 kHz) varying the ratio (percentage) of on to off time as required.

At > 400 kHz, it is highly unlikely that these "spikes" could cause mechanical vibration (buzz).

 

[mrb]

To the best of my knowledge (which is limited), any "electronic" fluorescent "ballast" is essentially a constant current power supply. (Fluorescent bulbs are odd devices; once started they will continue to draw more and more current until they burn out UNLESS there is a device (ballast) that limits the amount of current they can draw.)

All modern "high efficiency" power supplies are "switchers". Simple put, they switch on and off at high frequency (> 400 kHz) varying the ratio (percentage) of on to off time as required.

At > 400 kHz, it is highly unlikely that these "spikes" could cause mechanical vibration (buzz).

thats not correct. the switcher operates at high frequency but this is after the AC input is rectified. The rectifier presents a non-linear load to the AC supply. most power supplies such as laptop bricks, cell phone chargers, new wall warts, and good electronic fluorescent ballasts have power factor correction. CFLs generally do not. Think of power factor as a skew in the AC voltage waveform at which point current is drawn. A unity power factor of 1 will have current consumption exactly follow the voltage waveform. as the power factor decreases the current consumption falls out of sync with the waveform. Read the link i posted on power factor. Also I found this site which has a picture of an oscilloscope displaying the current consumption of a CFL.
http://cdonner.com/power-factor-and-compact-fluorescent-bulbs-cfl.htm

 

[mrb]

Also, the magnetic coil in a residential/commercial thermal magnetic breaker will never make any noise unless it is overloaded. The thermal element in the breaker trips on overloads and the magnetic coil provides the instant trip function for short circuits and extreme overloads ~100 amps for a 20 amp breaker. Look at the trip curve for say a square D QOM115. At normal operating current there is not enough current flowing through the coil to create much of a magnetic field.

 

[mrb]

in my math: current draw is 14 amps @ 120v considering power factor of 0.5 (105w/0.5pf)/120v = 14 amps

I forgot part of it. Should read:
current draw is 14 amps @ 120v considering power factor of 0.5 (105w/0.5pf)/120v = 1.75A x 8 bulbs = 14 amps

 

[gearheads78]

The buzz is coming from inside each individual breaker.

From Wikipdia - Circuit Breaker

A breaker uses a solenoid (a type of electromagnet) to release a spring which opens the contacts. ANY electromagnet (connected to an AC circuit) has the potential to "buzz" (mechanical vibration), if some of the components are not firmly seated (next to impossible). All transformers buzz.

Also, the magnetic coil in a residential/commercial thermal magnetic breaker will never make any noise unless it is overloaded. The thermal element in the breaker trips on overloads and the magnetic coil provides the instant trip function for short circuits and extreme overloads ~100 amps for a 20 amp breaker. Look at the trip curve for say a square D QOM115. At normal operating current there is not enough current flowing through the coil to create much of a magnetic field.

Just to update this tonight I replaced both 15a with 20a. I already had 15/20 light switches.

There is no difference at all in the noise. The noise is coming from each breaker because I can turn off either one and it about cuts the noise in 1/2. Sounds like something to just get used to??

 

[mrb]

Just to update this tonight I replaced both 15a with 20a. I already had 15/20 light switches.

There is no difference at all in the noise. The noise is coming from each breaker because I can turn off either one and it about cuts the noise in 1/2. Sounds like something to just get used to??

did you try unscrewing half of the bulbs in each circuit?

 

[mrb]

i only had a 13w CFL, i hooked it up to an ammeter that is able to capture the peak current draw. 1.33 amps. thats a 13w bulb. I bet those 105w cfls peak at 5 or more amps a piece. What make / model are they? If theyre not too expensive I might buy one to do some testing on.

 

[gearheads78]

did you try unscrewing half of the bulbs in each circuit?

OK I did that tand the sound is still there but at 1/2 or less of the sound. It might be quiet enough that I would have never noticed it if i was not already listening for it.

 

[gearheads78]

i only had a 13w CFL, i hooked it up to an ammeter that is able to capture the peak current draw. 1.33 amps. thats a 13w bulb. I bet those 105w cfls peak at 5 or more amps a piece. What make / model are they? If theyre not too expensive I might buy one to do some testing on.

These are the bulbs I have. I love how they work. With my white painted walls and ceiling. Its really bright in my 42 x 50 shop.

http://www.1000bulbs.com/product/6051/FC105-S65.html

 

[theoldwizard1]

Also, the magnetic coil in a residential/commercial thermal magnetic breaker will never make any noise unless it is overloaded. The thermal element in the breaker trips on overloads and the magnetic coil provides the instant trip function for short circuits and extreme overloads ~100 amps for a 20 amp breaker. Look at the trip curve for say a square D QOM115. At normal operating current there is not enough current flowing through the coil to create much of a magnetic field.

I still disagree.

Any electromagnet in a 60 Hz AC circuit can cause a mechanical buzz. Electromagnets are typically made up of "plates" (laminations) of steel with copper windings around them. At 60 cycle per second they can vibrate.

For a better picture of the internals of a typical breaker see this article on How Stuff Works From the 3rd section

The basic circuit breaker consists of a simple switch, connected to either a bimetallic strip or an electromagnet. The diagram below shows a typical electromagnet design.

The hot wire in the circuit connects to the two ends of the switch. When the switch is flipped to the on position, electricity can flow from the bottom terminal, through the electromagnet, up to the moving contact, across to the stationary contact and out to the upper terminal.

The electricity magnetizes the electromagnet. Increasing current boosts the electromagnet's magnetic force, and decreasing current lowers the magnetism. When the current jumps to unsafe levels, the electromagnet is strong enough to pull down a metal lever connected to the switch linkage. The entire linkage shifts, tilting the moving contact away from the stationary contact to break the circuit. The electricity shuts off.

Your typical breaker does not "sense" thermal overload (although these type of breakers do exist and are typically self-resetting).

Power factor has nothing to do with the buzz from inside the breaker.

 

[mrb]

I still disagree.

Any electromagnet in a 60 Hz AC circuit can cause a mechanical buzz. Electromagnets are typically made up of "plates" (laminations) of steel with copper windings around them. At 60 cycle per second they can vibrate.

For a better picture of the internals of a typical breaker see this article on How Stuff Works From the 3rd section



Your typical breaker does not "sense" thermal overload (although these type of breakers do exist and are typically self-resetting).

Power factor has nothing to do with the buzz from inside the breaker.

sorry but youre wrong.

From square D QO brochure: "QO® (plug-on) and QOB (bolt-on) one-, two- and three-pole thermal-magnetic circuit breakers provide
overcurrent protection and switching on ac and dc systems." look up homelinw, CH or whatever other residential breaker you like and you will see the same.

next lets address the electromagnet. I see what youre saying, BUT you need enough current flowing through the coil to create a magnetic field to get a buzz. The magnetic trip coil in a breaker is different than say a solenoid. It doesnt do anything until you get the right amount of current going through it, which is quite a bit more than the breaker rating. Once you approach the lower boundary of this current level, the magnetic trip element will not generate a field strong enough to trip the breaker but will make noise. Again, look at a breaker spec sheet, view the trip curve and where you find the instant trip range -thats where the electromagnet comes into play, everything before that is thermal.

now ill explain the three types of miniature circuit breakers and their applications.
thermal-magnetic -these are the most common and are used in 90% of branch circuit applications. The thermal element provides a time delay function that prevents the breaker from tripping on inrush from turning on cold heating elements and from starting motors. The magnetic trip element provides an instant trip function that clears faults and extreme overloads.

hydraulic-magnetic -this type of breaker replaces the time delay function of the thermal element with a plunger moving through a viscous fluid. The advantage is immunity to temperature both ambient and internal, as well as precise calibration and repeatable operation. You will find these used where ambient temperature is an issue (one very common place is in the portable power distribution used on movie sets where circuits are heavily loaded and the breakers are sitting directly in the sun) , and they are used to protect circuits supplied through semiconductors such as in professional lighting dimmers. They are also used in telecommunications and industrial settings.

magnetic only - i cant give much application info on these, but they have no time delay function. Exceed the rating by even a fraction of an amp and they trip instantly. Maybe theyre used in airplanes or something.

With regard to your comment about power factor having nothing to do with the buzz, in this situation it does. The noise is coming from the magnetic trip element which makes noise when a ton of current is passing through it. The CFL with poor power factor draws a ton of current, but does it so fast that the thermal element isnt heating up and tripping the breaker and he is under the magnetic trip threshold. Look at my 13w CFL with a peak current draw of 1.33 A. If I put 50 of these on a circuit, my load would be 650 W (plus whatever the ballasts eat) but the breaker is going to be seeing current spikes of ~66 amps and the magnetic element is going to make quite a racket.


note: the breakers (actually called supplimental protectors) in outlet strips, push to reset breakers on appliances and the such are thermal only. These are not suitable for or listed for branch circuit use -entirely different UL class.

 

[gearheads78]

So bottom line if I leave the 20a breakers in and ignore the noise the is no real dangerous situation

If you are willing to return it I can send a spare bulb to test.

 

[mrb]

So bottom line if I leave the 20a breakers in and ignore the noise the is no real dangerous situation

If you are willing to return it I can send a spare bulb to test.

thanks, but for the $10-$12 in round trip shipping (plus the chance of it getting broken in shipping) ill just go buy one.

as far as if its ok, i dont want to answer that definitively but if I were to guess the only downside would be the breaker eventually wearing out and starting to trip.

 

[theoldwizard1]

sorry but you're wrong.

I stand corrected. Square D (and likely all other brands) use both thermal and magnetic tripping mechanism.

next lets address the electromagnet. I see what you're saying, BUT you need enough current flowing through the coil to create a magnetic field to get a buzz. The magnetic trip coil in a breaker is different than say a solenoid. It doesnt do anything until you get the right amount of current going through it, which is quite a bit more than the breaker rating.

But it is doing something ! It is creating a magnetic field that is not yet strong enough to cause the breaker to trip !!

Why do transformers buzz/hum ? Same answer here on a much smaller scale.

With regard to your comment about power factor having nothing to do with the buzz, in this situation it does. The noise is coming from the magnetic trip element which makes noise when a ton of current is passing through it. The CFL with poor power factor draws a ton of current, but does it so fast that the thermal element isnt heating up and tripping the breaker and he is under the magnetic trip threshold.

Okay after reading Wikipedia on Power Factor and Switched-mode Power Supplies (it has been 30 years since I had any electronic classes and they barely mentioned power factor) I understand what you are saying.

Look at my 13w CFL with a peak current draw of 1.33 A. If I put 50 of these on a circuit, my load would be 650 W (plus whatever the ballasts eat) but the breaker is going to be seeing current spikes of ~66 amps and the magnetic element is going to make quite a racket.

Yes, but why ! What mechanical thing inside the breaker is vibrating ? Anything mechanical that is vibrating for a long time is going to fail (sooner than if it was not vibrating).

If this (poor power factor from fluorescent with electronic ballasts) is an issue, do power companies make large buildings with hundreds or thousands of electronic ballast do auxiliary power factor correction ? How about computer centers ? Again Hundreds of switched mode power supplies.



Bottom line for the original poster, add more lighting circuits so you have less bulbs on each circuit OR live with the buzz and replace the breakers when they fail.

 

[mrb]

to answer a couple of your questions:
switchmode power supplies in computers and other electronics have good power factor, in the high 90% by including power factor correction. A long time ago it was optional, now its pretty much standard. same goes for electronic fluorescent ballasts. CFLs have gotten away with bad power factor by pretty much flying under the radar.

here is whats going on with the magnetic trip element and associated mechanism that is making noise but not tripping: during normal operation (under the rating of the breaker) the trip mechanism is not seeing enough current to generate a magnetic field. If you look at the scope picture of the CFL current spike I posted a link to a while back you will see a huge current draw, but for only a fraction of a cycle. So every 1/120 of a second you have a high current spike that only lasts for a couple milliseconds?. Now look at the trip data for a breaker. You will see that it takes a full cycle to trip. Thus the current spike isnt 'energizing' (or putting enough current through to create a magnetic field) the trip element for long enough to trip the breaker. Thus the noise but no trip.

I agree with you on the potential for early breaker failure as I mentioned earlier in the thread.