Simplifying EQ/crossover understanding for sound Newbies

[JazzyRick]

I got to thinking after reading Pedros thread about bass cabinets and his questions about sound. So thought I'd give my shot at putting some of these items in terms joe off the street might have a shot at understanding. Here goes.

Typical 15 band EQ frequencies for a graphic equalizer:

25 40 63 100 160 250 400 630 1k 1.6k 2.5k 4k 6.3k 10k 16k

Now lets relate those numbers to say a bass guitar or Piano notes.

Low E on a Bass is about 41Hz...pretty close to that second slider on a 15 band EQ. The A string is about 55 Hz. The D string about 73.5 Hz. The G string is about 98 Hz.

The highest note on a piano is a C at about 4.2kHz, so why do we need EQ at 6.3k and up? Every sound from an instrument is colored by the wood/brass/strings/reeds etc... and even the room played in. These all contribute to what is called harmonics. Harmonics are multiples of the vibrations that is the fundamental of a note. Take a 41Hz low E on a bass. The first harmonic is 82Hz or the next octave E. Now the tricky part. The next harmonic just adds another 41Hz, so the second harmonic would be 82+41Hz or 123 Hz which is a B...not an E at all, but if you listen closely when you pluck an E string you can hear that 123Hz B note...sometimes not at all because it's too quiet, but sometimes in the right situation it may actually be nearly as loud as that 41Hz E or even louder.

Whew...

What an EQ is is a series of miniature volume controls for a certain range of frequencies. You see the 40Hz slider on an EQ doesn't just reduce or increase 40Hz exactly. It effects the frequencies near it as well to some extent.

The "smiley face" EQ setting for Bass guitar is infamous around these parts. The lower and upper groups of frequencies are increased while the midrange frequencies are lowered. Many times this results in big BOOMY bass, but you can't hear any of the definition of the notes at all...just wooom wooom wooom, instead of dooon dooon dooon. It's barely music if you ask me, but that's just an opinion.

If you crank that 63Hz slider up like many bass players are known to do what you are doing is increasing the volume near the notes B and C(second/third fret on the A string on a bass). Then you bury what makes a note have it's character....the harmonics(not to be confused with distortion harmonics a speaker produces).

Most vocals are in a range from 200 ish Hz to 1.5kHz. S and definition of words and cymbals and Cracks of snare are above that(2k-10k ish) and above that is not too much but "airy" types of sounds.

I would guess there are some mistakes above and some easier ways to describe what I'm trying to get across, so feel free to add, comment, or criticize. I can take it, as this is just meant to be a starting point.

As far as what a crossover is....

A crossover splits the output signal at a certain point measured in Hz as well. A typical 2 way PA system run by weekend warrior sound guys like myself would have a crossover point around 100 Hz or so. All sonic content below 100Hz would be sent to a "low" output. Everything above 100Hz would be sent to a "high" output. You would then take the LOW output and run it to a separate amplifier to power a sub woofer. The other HIGH signal would go to an amplifier powering a top speaker(typically in these situations the top speaker would even have a Passive crossover built into it to separate it's own signal between the woofer and tweeter).

[Strapping Young Stu]

Nice thread, to give a bit of a graphical representation of the frequency distribution I took a photo of my piano and annotated it. The main thing to notice is the way the frequencies are distributed, for every octave up you get a doubling of frequencies, which means that all the bottom EQ sliders on a graphic are working very close to each other (maybe a tone or two apart) whereas the 3khz and 5khz sliders are almost an octave apart.



Here you can see that the bottom string of a bass guitar (41.9Hz) is right down at the bottom range of a piano, and that a pianos fundamental frequencies only stretch to around 4kHz.

However there are many frequencies that go beyond this in the harmonics all the way up to 16kHz and higher and the materials of construction are solely responsible for this.

In these terms EVERYTHING makes the difference. Size of the soundboard, material of the soundboard, thickness of the soundboard, material of the strings, thickness of the strings, how many strings per note, hammer felt hardness, action of the hammer, etc etc.

The same applies to all instruments - on a guitar for example you can get a huge range of tones just from the guage of strings and the thickness of the pick used.

This is why all instruments can range hugely in price and performance, because everything about them contributes to their tone and sound qualities.

The above picture should be a huge visual aid for a sound engineer, for example... "This guy whos singing has a really weak bottom end to his voice and I need to fix it with some EQ... well if I go home and practice with a piano for a while I can find out that most guys can sing down to around 100Hz and up to 500Hz for extreme ladyboys. So now I know that I have to EQ somewhere in that range to fix the sound."

In the same way "This bass player has a really boomy sound and he seems to play a lot in the 2nd octave up from what I can see from the desk. Lets check the chart ... hmm 80-160Hz... well that boom sound quite high so lets try cutting at 160Hz....nope still not right, that boom seems higher lets try 200Hz.... getting there.... 250Hz.... Bingo!"

Stu

[bgavin]

which means that all the bottom EQ sliders on a graphic are working very close to each other (maybe a tone or two apart).

This is not correct.

My EQ boxes are all on geometric intervals, meaning the note relationship remains constant. Starting at the bottom, each slider is 1/3 octave above the previous.

[Strapping Young Stu]

which means that all the bottom EQ sliders on a graphic are working very close to each other (maybe a tone or two apart).

This is not correct.

My EQ boxes are all on geometric intervals, meaning the note relationship remains constant. Starting at the bottom, each slider is 1/3 octave above the previous.

In the case of a 31 band yes, but this is not necessarily the case with 15 band and other EQs which might be built into an instrument amp

Just to clarify

Stu

[Bill Fitzmaurice]

In the case of a 31 band yes, but this is not necessarily the case with 15 band and other EQs which might be built into an instrument amp

Just to clarify

Stu

If anything the EQ bands on instrument amps are even further than 1/3 octave apart. 15 band EQs are usually set at 2/3 octave intervals, while 5 and 7 band may have as much as two octave bandwidths. I've never seen an EQ with spacing 'a tone or two apart', and most attempts to do so would be fruitless anyway, as the bandwidth of analog filters cannot be made narrow enough for that kind of resolution. Some digital filters are made with bandwidths of 1/5 or even 1/10 octave, but those are used in automatic feedback killers. Feedback killers with manual controls are useful with subs in small rooms to tune out room modes, but that's very much a hi-fi implemenation only; you won't find them being so used in rooms large enough for pro-systems, nor will you find them on-board in a bass amp.

[JazzyRick]

I'd have to back up Bill on this one. There is usually the misunderstanding that they are closer together simply because of the numbers. 25 is pretty close to 40 and 40 is close to 63, but 1.6k and 2.5k are nearly 1000Hz apart. Numbers do not, however, equate to the distance between tones in what we hear.

If you look at that Piano diagram it's quite evident.

The 2 highest A notes you have there are 1.7k and 3.5k respectively. The 2 lowest As are 27.5Hz and 55Hz....only 27.5Hz apart as opposed to 1700Hz. They both sound 1 octave apart to our ears though.

[Bill Fitzmaurice]

They both sound 1 octave apart to our ears though.

The both are 1 octave apart. One octave is a 2:1 frequency ratio, or 1:2 if you prefer.

[JazzyRick]

I'm just a hack weekender so bear with me, but I am also a research freak, so maybe some of this will make sense.

Some nearly universal EQ settings I've noticed over the years.

The 250Hz slider nearly always will clean up the definition of the vocals/guitar/horns by reducing it. It's a super muddying frequency in my opinion. Why is this? Well, I busted out my trusty rusty website finder to end up here http://www.padrick.net/LiveSound/CancellationMode.htm(ok actually I got that link from this forum some time ago), and low and behold there's this chart called room modes. If you run sound at clubs like I do most every ceiling is 8-10 feet high. This puts the room mode frequency in the 55-65 Hz range. This means there will be a natural boost in output in that frequency range(remember those bass players who boost 63Hz....well in 90% of the rooms they play they already have a 3 dB boost or so there!!! WOOOOOM WOOOOOM WOOOOOM). Now lets do some sound guy math(most of us can barely add). Double that range to get the first harmonic(remember those?) and you get to around 110-130 Hz. Remember Bass guitar has a LOT of content in the first and second harmonics, so banging around A B C and D(that 60Hz range) puts some pretty strong harmonics in that 110-130Hz area as well. Next up the harmonic scale is(drum roll please)...220-260 Hz. So why is this such a big deal? Well the harmonics tend to effect the room similar to the fundamental, and most people don't sing low enough(at shows I do anyway) for a 120Hz boom because of room modes to effect the sound of their voice. At 250 Hz though you are right into the heart of most peoples vocal range. If your really into naturally killing some of that boom at 250 you can build a 13.5" high crate to put the monitors on top of as that's a boundary cancellation measurement for that 250Hz range as well. Doesn't science suck?

Just think what happens when that monitor speaker is not only 8-10 feet from the ceiling, but also 8-10 feet from a back or side wall???

I also have common feedback problems in different venues at 2.5k and 4k. Can anyone shed some light on this one perhaps?

[Bill Fitzmaurice]

The 250Hz slider nearly always will clean up the definition of the vocals/guitar/horns by reducing it. It's a super muddying frequency in my opinion.

So-called electric bass 'boom' is usually sourced in the midbass, for the same reasons you noted. Killing the bass tone control removes it, but removes the bass as well. What works well is a parametric, just sweep the frequency knob until the boom disappears and you don't lose your bottom end too.

[Strapping Young Stu]

I'd have to back up Bill on this one. There is usually the misunderstanding that they are closer together simply because of the numbers. 25 is pretty close to 40 and 40 is close to 63, but 1.6k and 2.5k are nearly 1000Hz apart. Numbers do not, however, equate to the distance between tones in what we hear.

If you look at that Piano diagram it's quite evident.

The 2 highest A notes you have there are 1.7k and 3.5k respectively. The 2 lowest As are 27.5Hz and 55Hz....only 27.5Hz apart as opposed to 1700Hz. They both sound 1 octave apart to our ears though.

Yep you guys are totally right, sorry thats extreme tiredness for you!

Stu

[Strapping Young Stu]

I'm just a hack weekender so bear with me, but I am also a research freak, so maybe some of this will make sense.

Some nearly universal EQ settings I've noticed over the years.

The 250Hz slider nearly always will clean up the definition of the vocals/guitar/horns by reducing it. It's a super muddying frequency in my opinion. Why is this? Well, I busted out my trusty rusty website finder to end up here http://www.padrick.net/LiveSound/CancellationMode.htm(ok actually I got that link from this forum some time ago), and low and behold there's this chart called room modes. If you run sound at clubs like I do most every ceiling is 8-10 feet high. This puts the room mode frequency in the 55-65 Hz range. This means there will be a natural boost in output in that frequency range(remember those bass players who boost 63Hz....well in 90% of the rooms they play they already have a 3 dB boost or so there!!! WOOOOOM WOOOOOM WOOOOOM). Now lets do some sound guy math(most of us can barely add). Double that range to get the first harmonic(remember those?) and you get to around 110-130 Hz. Remember Bass guitar has a LOT of content in the first and second harmonics, so banging around A B C and D(that 60Hz range) puts some pretty strong harmonics in that 110-130Hz area as well. Next up the harmonic scale is(drum roll please)...220-260 Hz. So why is this such a big deal? Well the harmonics tend to effect the room similar to the fundamental, and most people don't sing low enough(at shows I do anyway) for a 120Hz boom because of room modes to effect the sound of their voice. At 250 Hz though you are right into the heart of most peoples vocal range. If your really into naturally killing some of that boom at 250 you can build a 13.5" high crate to put the monitors on top of as that's a boundary cancellation measurement for that 250Hz range as well. Doesn't science suck?

Just think what happens when that monitor speaker is not only 8-10 feet from the ceiling, but also 8-10 feet from a back or side wall???

I also have common feedback problems in different venues at 2.5k and 4k. Can anyone shed some light on this one perhaps?

From what I have learnt about mixing records over the years which applies to live sound as well.

Midbass frequencies of 200-500Hz are almost solely responsible for the terms "mud" "middly" "boom" and many others.

Speakers are naturally getting into their "efficient" operating range at 200Hz and above and because of this we hear the harmonics a lot better than we usually hear the fundamentals.

In addition a lot of frequencies are present in that area - 2rd harmonic of bass, 1st harmonic of guitar, keyboard left hand, vocals, some harmonics of the kickdrum and the toms.

In other words - a LOT of stuff, and you have to sift through the different tones and figure out which ones sound good at that frequency - usually the vocals and sometimes the guitar. To do that you need to remove some of the frequency range from other instruments to give enough space to breathe.

So the typical scooped bass EQ sound applied to the kickdrum works wonders here, as in a live situation the general trend is to feel the kick rather than necessarily hear a specific pitch, together with a slight cut for the bass guitar just to even out the tone from the first and second harmonics.

Depending on the amp, mic, and cab you might have to play about with the guitar as well, but in my experience I tend to end up cutting the higher end of the guitar which is around 2kHz and above - usually because the guitarist has a 4x12 and doesnt realise how much gainy high end rubbish is coming out of their cab and taking peoples heads off.

Stu

[Ron K]

Normally I wont EQ for individual instruments. I noise the room at several different critical listening location. Then I add the necessary EQ to flatten the total system response. This may require boost or cut in many different bands. Hopefully when I'm done everything will sound pretty good when I run a CD through it.

My next step is to set the overall system gain or how loud I'm going to let the system get. I do this on the crossover limiters generally setting them at +6 to +10 db. across the board.

Finally I'm ready to dial in the separate instruments and vocals.I use the channel mid sweeps on the board to set each instruments tone or each vocals tone by mixing into my set system curve! If a particular channel requires more than the two mid sweeps and High and Low Tone or low shelving I simply insert an additional graph into that channel and Q to my hearts content.

If compression is needed or any other processing I normally just insert it on the associated group or channel. Far too many times do I see system operators over compress or under limit! At the end of the day they wonder why their take doesn't cover their costs for repairs! The same goes for EQ. A great tool but used improperly it can cost you plenty or just plain mess up your sound!

Addressing the 2.5K and 4K feedback in monitors. This is quite normal. Just cut it till the feedback diminishes. If you find yourself using more than 6-9 db cut consider repositioning the monitor at a different angle to the vocalist or just turn it down. Another trick is to duck it with a compressor and EQ. Usually some repositioning takes care of it. I find 400 and 4K to be my most cut freqs on monitors on harder stages.

[billynoband]

I thought we were keeping this thread simple. It started off fine.I cannot understand it unless you tell it to me like I am six.
If you do not stop I will send you all e-mail of 10000 words on quantum physics.