Back Ground Sound
Home > Home Recording Forum > Recording Techniques > Back Ground Sound
Posted on Oct 16, 2004 09:32 am
Bleak
A small pie will soon be eaten
Member Since: Aug 26, 2004
I am picking up my computer's fan when i record through the mic. At the moment the mic is about 2-3 feet away from the pc. I know this is not ideal but it is un-avoidable at the moment. I know, given this proximity, i wont be able to cut the fan sound completely but i was just wondering if there are any techniqes to minimize this.
I have a hardware compressor with a limiter. Is there a frequency or threshold that i can set so as to minimize back ground noise?
I should add that this isnt a HUGE problem as when i actually play the guitar does seem to overwhelm the pc sound.
[ Back to Top ]
Oct 16, 2004 09:41 am Well, one option would be to hang or pace something between the PC and the mic, like a blanket, pillow, board of some sort or something.
Alternatively, you may want to record JUST the fan, save it as a track, then record the track with the fan and instrument. Then try reversing the phase of one of them while playing them both back, the phase inversion might get rid of it, then, if that doesn't work, you have a track of just the fan to do a noise analysis on and use a notch EQ or a noise remoer to remove it from the take.
ILLbizzelILLbino - one of WA's finestMember
Since: Feb 10, 2004
Oct 16, 2004 08:54 pm Db, can you explain both those suggestions some more: phase inversion, noise analysis?
BleakA small pie will soon be eatenMember
Since: Aug 26, 2004
Oct 17, 2004 10:01 am The blankets, and so on, do work!
but as for the other suggestions, can they be used in the long term for good results?
Oct 17, 2004 10:17 am if they work once, they can work every time. It'd just be a pain in the butt to do everyt ime, the better long term solutions is fixing the problem.
Illbizzel, by noise analysis, I mean plugins such as Sonic Foundry's NR plugs can listen to the track with just the fan and "learn" the frequency spectrum of that sound, then apply it to the other track to remove that spectrum, thereby removing the noise (in theory, it takes some tweaking sometimes)
As far as phase inversion goes, I know what it is and how to do it, but not really how to explain it. In theory, two track of the same sound, one with normal phasing and one with reversed phasing, cancel each other out. Therefore, if you record the fan noise, then the real track, play them together reversing the phase of the fan noise track, could remove that sound from the audio...in theory...
LokiCone PokerMember
Since: Apr 07, 2002
Oct 17, 2004 10:31 am phasing can be a nightmare when recording drums too :( what happens is you have two waves, and one cancels the other out... I can't explain it either, geez
ILLbizzelILLbino - one of WA's finestMember
Since: Feb 10, 2004
Oct 17, 2004 05:39 pm There are different phase settings though. How do you know which one to use?
Noize2uCzar of MidiAdministrator
Since: Apr 04, 2002
Oct 17, 2004 07:02 pm Ill, I have found using a piece of foam such as packing foam, not the aquirt in type, but the kind found in some guitar cases and such. More like studio foam. I use a piece of that behind the mic to eliminate most noise that comes from behind the mic. You simply cut a slot in it, and hang it over the mic stand directly behind the mic, and viola a nice baffle. They also have them at Auralex here is what thye look like. www.auralex.com/sound_iso...on_xpanders.asp
Oct 17, 2004 07:51 pm just for future reference, a compressor would bring up the noise floor, not make it quieter.
ILLbizzelILLbino - one of WA's finestMember
Since: Feb 10, 2004
Oct 17, 2004 08:11 pm Yea, I just bought 36 pieces of 2 inch thick wedge foam. I got a great deal on it. Less than $1 for each! I know that is going to help considerably for better vocals.
I dont have fan noise problems, but was curious how people removed noise. I've used noise reduction, but not in the way DB was suggesting.
Oct 17, 2004 08:41 pm the phase inversion thing is pretty straightforward in theory but i have yet to try it. basically all it is...
imagine a perfect sine wave. a peak followed by a trough of the same depth. any natural sound is made up of scores of these, at different frequencies.
now go get in the tub and start making waves. when you see one comin' at ya, send another one toward it. bam! where the crest of your outgoing wave runs into the trough of the oncoming wave, the water there turns momentarily flat. they cancel out.
same deal with sound. if you have an exact copy of a sound and phase-reverse the copy and play them together, every peak cancels out every trough. but for this to work perfectly, the recordings must be identical. that won't necessarily be so with a fan recorded at two different times. i bet if you tried it, chance would guarantee that some of the sound would cancel out, but i can't say how much.
with drum phasing, do you get a sort of a flanging sound? is that cause the different mic distances create little notches of cancellation, the frequencies of which change over time because of decays and stuff, and so you end up with an effect like a sweeping notch filter?
WaltChief Cook and Bottle WasherMember
Since: May 10, 2002
Oct 17, 2004 09:13 pm Ya Forty,
In the digital world an algorithm simply adds the signals and sends the result to the A/D converter. If a peak and trough are added the net sum is zero. Obviously every point along the waveforms will be exact oppisite and also cancel each other.
In the analog world things are not quite as descreate. Each signal will pass through some amplification stages independently prior to being mixed. Per tolorance differences in amplification stages some signal is usualy generated post mix per stage matching differences.
Once to the speakers you now have quite a bit of room for error as speakers have even greater tolroance differences than the analog stages that proceed them. You also have acuoustic effects of air travel in the room. Obviously the end result is still a diminished resulting sound.
Extropolating on those postulates; it is easy to see why higher frequency signals are so impervious to phasing effects. Their probability of lining up sufficently to create phasing effects are stisticly rare. Low frequencies however being quite slow can easly line up to create phasing opertunities. This explaines that one low tone that will just not shine through a mix. Often it is necessary to isloate just that one tone and reverse that portion of the waveform only. Also as you noted there is always that freek sub-harmonic of some mid tone that has a tendency to be real mystic, hence the 80hz cut on the getfiddle to let the bass shine through as a precautionary or proactive measure.
Take all of that and add oslicitory effects in both anolog circuitry and in natural acoustics and WoW, what a matrix of things to consider. Great fun for those who have a lot of patience.
Noize2uCzar of MidiAdministrator
Since: Apr 04, 2002
Oct 17, 2004 09:28 pm OH sorry I guess it was Bleak looking to quiet fan noise.
WaltChief Cook and Bottle WasherMember
Since: May 10, 2002
Oct 17, 2004 09:44 pm hahahahaha! Touche' my friend! Ya, I'm playing around with some polyester fill that is inexpensive and together with my ole Singer is making some very nice, light, fairly rigid absorption panels. In designer fabric no less! Matches my purse!
