RSS
Ask MAKE is a weekly column where we answer reader questions, like yours. Write them in to mattm@makezine.comor drop us a line on Twitter. We can't wait to tackle your conundrums!
A few weeks ago, we looked at why dimmer switches cause incandescent light bulbs to buzz. We subsequently received a number of questions about why flourescent bulbs vibrate, such as this comment by snarkyFish:
It might be nice to have another one of these that explains why fluorescent lights / ballasts hum and buzz as well. I imagine it's a very similar explanation, but a much more complicated fix.
You are right, they do buzz for a similar reason, which is the mechanical components being vibrated by electromagnetic waves. In this case, however, it is probably not the bulb itself that buzzes, but the ballast.
Fluorescent bulbs are made up of a glass tube that is filled with a low-pressure inert gas and a small amount of mercury, have a phosphor coating on the inside, and have an electrode on each end. When a large enough voltage is applied across the electrodes, the gas begins to conduct, which allows an electric current to flow from one electrode to the other. This current causes collisions between gas molecules and the mercury, which creates UV energy that is then converted to visible light by the phosphor coating on the tube.
This is all well and good, however as the gas heats up, it's resistance goes down, which means that it becomes a better conductor and subsequently draws even more current. If this were left unchecked, the bulb would quickly heat up and self destruct, so a ballast is placed in series with the bulb to limit the current draw. There are a number of different kinds of ballast design, but the simplest one is to just use an inductor. An inductor will certainly limit the rate at which current can travel to the light, however it does so by absorbing magnetic fields into it's core. This absorption, which causes magnetostriction, is probably the source of your hum- it literally causes the inductor to expand and contract at twice the AC frequency, which creates an audible sound wave (at 120Hz in the US or 100Hz in most of Europe).
I don't think there is a safe way to fix this besides getting a new ballast. Newer ballasts use an electronic controller instead of an inductor for the ballast, so they shouldn't be susceptible to the same buzzing problem. Making your own is certainly a possibility, but as with any high-voltage project, you better know what you are doing first.
Any other tips for how to fix a buzzing fluorescent light? Let us know in the comments!
[CC licensed photo by adotjdotsmith]
Its essentially the same thing that happens with electrical distribution transformers (the ones you see at substations or sometimes underneath grates on city sidewalks)
Its taking place in a slightly different setup than an induction coil, but the effect of the electromagnetic waves on the metal is the same: really frequent expansion and contraction.
Reply to this comment
Ah cool, that makes sense. I suppose that mechanical engineering starts to play a major part in the design of the transformers, both so that the design minimizes vibrations in the audible frequencies, and so that they don't just vibrate themselves apart after a short period. Interesting!
Reply to this comment
Magnetostriction is only part of the actual noise. You'd get noise also without an iron core because the coil itself vibrates. Why? The magnetic field (from the coil) acts on moving charges (current in the wire). Since the electrons are in the wire, the wire is subject to a force, and it moves.
Reply to this comment
Electronic ballast!
Reply to this comment
Electronic ballast and the starters might make a good band name.
Reply to this comment
That seems another reason against the ban we have here in Europe on ordinary light bulbs
- and that is coming to the USA.
As a commenter noted above, and as you say in the article, electronic ballasts (outside dimmers) shouldn't give the buzzing
problem, but of course leaves plenty of others, not just the mercury risk, but for example the fire risk of CFLs in enclosed spaces
(more http://www.ceolas.net/#li18ex )
Another "cure" to buzzing etc is of course to use LED lights instead,
but again, that doesn't mean having to ban ordinary lights,
and the supposed savings of a ban end up marginal anyway, in an overall context.
(About the industrial politics behind banning the light bulbs
http://www.ceolas.net/#li1ax )
Reply to this comment
Also, because the load isn't sinusoidal, you'll get a lot of harmonic content, which makes the hum louder, and with higher frequency content.
Electronic ballasts theoretically aren't audible, but I've noticed that a lot of them are. They just tend to make a super high pitched staticy hiss, instead of the familiar buzz.
Reply to this comment
Ah, excellent point. I wonder if anyone is using them as a sound generating source to do subtractive synthesis?
Reply to this comment
I recognize the low frequency hum of around 100 - 130 Hz you write about but I also usually notice a very high frequency sound from fluorescent bulbs when they are just turned on. I think the sound starts at around 15000 Hz and increases to over 17500 Hz which is the limit of what I can hear. I notice the sound at the same time as light gradually appears in the bulbs after they are turned on, the sound is quite intense but many people are not able to hear the high frequency. Can you explain this sound?
Reply to this comment
If it only occurs when the lamp turns on, it is probably caused by the mechanism that pre-heats the gas in the lamp. For newer electronic models, the electrodes themselves act as heating elements, and I suspect that the circuitry that drives them is causing the high-frequency buzzing.
Reply to this comment
Try holding an oscilloscope probe up close to a fluorescent light. Some of them ring well up into the radio frequency range. I've seen some oscillations on old lamps well above 1 MHz!
The glow discharge inside the lamp is inherently non-linear and represents a negative resistance. Couple that with a fairly low resistance and distributed capacitance found in the wiring, and you have a RF oscillator.
Dave
Reply to this comment
Ah, right! That probably means they can cause all sorts of havoc with other electronic devices that aren't properly shielded. I'll give the 'scope experiment a try next time I am in the shop.
Reply to this comment