:) Clipping, for one.

--Rick

Excessive EQ bass and treble, sharp transients, square waves at high levels, massive amounts of DC offset, dropped microphones, recording sonic booms or jet engines at full roar up close and personal, massive amounts of electronically produced super low or super hi frequencies, etc., etc. The list is endless. Mostly it depends on level. If it is too much level all at one time the speaker can be permanently damaged beyond repair.

I use do audio in a college. We would have young composers coming in to do playbacks of their pieces and it would be all square waves off the MOOG or it would have sharp transients that were over recorded on a DAT causing distortion plus the transients which would not reproduce correctly on the sound system installed in the hall. They would want to play the stuff back at ear splitting levels and then wonder why it did not sound very good.

Tracks that are hypercompressed and then normalized to 0 dBFS (or -.1 or whatever) which represent waveforms that can actually exceed 0 dBFS by several dB, oftentimes causing distortion in the analog circuitry of D/A converters.

-Duardo

Deep bass tones (like <50Hz-ish) can easily make subs start to mechanically distort. Pretty common issue with trance and techno.

Attached files

Thank you for your comments.

How will I know what to avoid? As Thomas wrote, analog synthesizers can often create dangerous output. Is there some kind of model to keep within.

I have several songs that I wonder if they are damaging speakers. Is there an EQ slope that I should definitely not go beyond?

Also if I have pure square waves what do I have to do to dull the corners and how much do I need to dull them before they wont thrash someone bomb box?

Also, if I have a recording of a microphone that's been dropped or over-vibrated in some other way, that I want to save, can I rescue it by cutting out the low frequencies? If so how much? IF not, what else needs to be done?

I'm sorry to ask so many specific questions all at once, but this is a pretty important issue for people producing and finishing their own music, and I can't find any other sources on the matter. I've read several mixing, engineering and mastering books, but they never seem to mention this obvious issue.

Thank you again. :tu:

Originally posted by noit:
Thank you for your comments.

How will I know what to avoid? As Thomas wrote, analog synthesizers can often create dangerous output. Is there some kind of model to keep within.

I have several songs that I wonder if they are damaging speakers. Is there an EQ slope that I should definitely not go beyond?

Also if I have pure square waves what do I have to do to dull the corners and how much do I need to dull them before they wont thrash someone bomb box?

Also, if I have a recording of a microphone that's been dropped or over-vibrated in some other way, that I want to save, can I rescue it by cutting out the low frequencies? If so how much? IF not, what else needs to be done?

I'm sorry to ask so many specific questions all at once, but this is a pretty important issue for people producing and finishing their own music, and I can't find any other sources on the matter. I've read several mixing, engineering and mastering books, but they never seem to mention this obvious issue.

Thank you again. :tu:

As a rule of thumb if it sounds bad (like a dropped microphone) it probably is not too good for your speakers.

There are no hard and fast rules but if you were to play almost anything with transients in it (including square waves) at high levels it would prove difficult for most speakers to reproduce. Think of a speaker as a piston laid on its side going slowly in and out - in and out but if you put square waves into it the speaker the speaker has to reproduce something that is full on and then full off so instead of a piston you have an electrical switch (just on and off) and speakers make lousy switches since that is not what they are designed to do.

When speakers start having problems reproducing what you have created a couple of things happen (sometimes at the same time) First the voice coil can start getting hot and when it gets hot it expands. Since the gap is only so big the speaker voice coil may get so hot and swollen that it impedes the speaker from moving in freely and out by feeding more and more level into a speaker that is, for all intents and purposes, frozen in the gap the speaker will start to self destruct.

The other thing that can happen at about the same time from sharp transients is that the voice coil can leave the gap on a forward excursion and may not be pulled in correctly by the spider which can cause an off center voice coil that can then start rubbing on the sides of the gap and cause a failure.

The best advice is to use moderation in all mixing, equalization, sound effects and compression. Stay away from hi level transients such as dropped microphones if at all possible and if you want to do this for an "effect" turn down the gain to a safer level.

I have seen material coming trough my mastering room that has eq at the extremes of +30 dB at 40 Hz and 15,000 Hz. The reason this was done was that the person was monitoring on cheap speakers that did not have a wide frequency range and in order to get the "effects" he wanted in his mix he kept turning up the eq thinking that he was listening to the fundamental only to realize later in the mastering process he was indeed listening to harmonics and not to the fundamental. (Boy is that an ear opener....)

Any time you have to go over about 6 dB in equalizing anything you should stop and think about it. Maybe you are trying to eq something that is simply not there or you are unable to hear it due to your speaker's response. A lot of hip hop and Reggae use a lot of eq at 60 and 80 Hz. This is fine until someone puts it into their boom box or car stereo with the bass boost turned on and a smiley eq setting and then you get a double dose of the bass which can ruin speakers if played at high levels.

Again their are no hard a fast rules for composing music or for sound effects (just go to your local cinema and listen to the sound tracks for any of the space movies or epics like LOTR.) you will be enveloped by sound at very low frequencies but the speakers in the cinema are built to provide this very low frequency information and the people who mixed the movies are professionals and understand what they can and can't do and know that the cinema owner is not going to be too happy if he has to replace all the speakers in the theater if the mixers try and overpower the system.

If you listen to some boom cars you can hear what happens when really loud bass is played on speakers that are designed to produce very low frequencies but the car is not and it rattles. When someone tries to overpower the subs in their car they start distorting and after a while the voice coil distorts enough that even when you play "normal" level music though them they sound distorted.

If you use moderation in everything you should be fine. This does not mean that you can't use effects such as dropped microphones or square waves but you should reduce the gain when using them and not use them for long periods of time as this can cause the problems mentioned above.

Hope all this helps.

Thanks Tom, your comments are very helpful in keeping the range of tolerability in perspective.

Got a quick question. A demo I'm doing right now has a little part which is literally four tracks of AM radio noise and five tracks (plus two room ambiance) of literal double maxed-out distortion guitar through a wah.

Assuming all this goes out a speaker and into a mic (SM57 to be specific), would it be less "dangerous" than direct recorded signal? Or is it even dangerous at all?

Got a clip if necessary.

Originally posted by falkon2:
Got a quick question. A demo I'm doing right now has a little part which is literally four tracks of AM radio noise and five tracks (plus two room ambiance) of literal double maxed-out distortion guitar through a wah.

Assuming all this goes out a speaker and into a mic (SM57 to be specific), would it be less "dangerous" than direct recorded signal? Or is it even dangerous at all? Got a clip if necessary.

It is all level depend. If you play it at "ear shattering levels" yes you are going to destroy speakers and maybe someone's hearing. If you use it at a more "normal" level (below 85 dBSPL then no it will probably work ok. White noise (AM radio noise) is used for certain kinds of testing. Distorted Guitars are the norm of a lot of rock and roll so no they in themselves will not be "dangerous" but if you play them at full concert volume (+120 dBSPL) then they will be dangerous to equipment and to your hearing. One thing to keep in mind is that when you distort something it turns into square waves and speakers by and large do not like square waves so you run the risk of damage from the square waves and then if you overdrive them you have a double jepordy.

Hope this helps.

Makes sense that way. Never thought to measure up the RMS vs some of today's rock (*cough*lastestAlkalineTrioalbum@-5dB*cough*).

It *sounds* like audio torture, but the levels are pretty mild in comparison. Most of the energy falls below 6kHzs too - typical guitar ranges, so there's no cone-shattering treble.