1/3 octave EQ

S

smg

Member
Been working on things using a TASCAM DP-OO8ex which has a low/high shelf EQ built in...got tired of the limitations so I looked around online and found a good deal on a Samson S curve 231....appreciate any feedback from people who use/have used this ....+ about using 1/3 octave equalizers in general...

i.e.I was reading online that even though each band has a narrow Q the frequencies around it get boosted which seems like an inherent contradiction unless it means that the specific frequency being dealt with itself is the only one affected in the immediate area but further away there is some effect i.e.boosting @ 100hz would affect 110hz and 90,hz but not directly around 100...????
 
Graphic eqs are mostly useful for live sound, but perhaps it would be of use for you. More appropriate would be a parametric eq of some sort.

"Narrow Q" is kind of a dumb mix up of high Q and narrow bandwidth, which are two ways to say the same thing. It means the filter slopes off steeply either side of the specified frequency so the cut or boost doesn't affect too much of the audio.
 
I actually looked at parametric first but they were all out of my price range so I opted for the Samson...I realize I won't be able to hit the exact area where a frequency that needs to be cut/boosted is but relative to using a high/low shelf to simulate how parametric /graphic EQ' deal w/ specific frequency areas being affected (by recording instruments to 2 tracks then eq'ng them individually so the combined effect produces a given result (i.e.the low shelf range is 32hz to 1.6kH...so I'd set one @32-150 + 6dB and the other @32-80 -10dB so 80-150 would get boosted and 32-80 would get cut -4dB...and since there's an input EQ/track EQ/mastering EQ I could keep messing around like this but.....)...its a big improvement..
.
Thanks for the info re-narrow Q...unlike what I thought was meant by what I read,using the ex.above @ 100hZ let's say 90-100 and 100-110 would be affected as two sides of a ascending slope w/100 at the top + 90/110 as the L/R bottom points ???
 
Hey thanks for the link...I'm going to take a look at the rest of the site as well...can't get enough good info as I just got started w/recording/mixing/mastering w/the DP008ex and have a lot of demo tracks made on the keyboard workstation sequencer w/drum machine recorded w/a liitle digital handheld to redo on the TASCAM......

Is that right though(above re-way 1/3 Oct EQ Q works w/slope)?
 
While it's true that 1/3 Oct EQ's are usually used for live PA work, I have at least one sitting around in the studio for the occasional EQing where needed usually to eliminate a problem, I used to run the kick drum through mine for a long while and it worked well, a bit of a throwback from my live days where I used to run the kick through a 31 band and then a compressor due to having several bands a night on stage with quick changeovers and not having the time to fix the kick sound up at the kick drum LOL. It still has it's uses in the studio.

Alan
 
Not your graphs but the way I was conceptualizing the idea of narrow Q as you suggested it is on the 1/3,octave re-slope and adjacent frequencies...
 
smg said:
Not your graphs but the way I was conceptualizing the idea of narrow Q as you suggested it is on the 1/3,octave re-slope and adjacent frequencies...
Click to expand...

Here's a parametric eq filter at max boost and cut, and at three different Q settings. The inner one is the highest Q, the narrowest bandwidth. It affects the center frequency the most and adjacent frequencies progressively less. The narrow bandwidth is the most selective, the wide bandwidth is the least selective.

tascam_pe-40_1khz.jpg
 
smg said:
Alan you familiar w/the unit I'm getting?(Samson S Curve 213)
Click to expand...

I think you mean the 231, no but I have used so many 31 band EQ's in my life that I can't remember how many brands and generally they all do the job, admittedly some better than others but it's rare to find one that is really bad.

Alan.

Just a footnote, I very rarely boost anything on a 31 band EQ, I would usually only cut frequencies, which is maybe why I don't mind brands so much.
 
Okay I think I understand it now..using the graph-1k is boosted the most @ 98dB while 900 on the left and 1.25 k on the right are boosted @90 dB and it gradually slopes down and evens out @ 84 dB on the left at 300 hz and at 3k on the left..
 
Alan yeah its 231 excuse me just ordered it today...re-"cut narrow boost wide"you could concievely boost a frequency range using it as well of course...but instead of using just one frequency band you might use a few i.e.100/125/160/200 +250 .maybe on an ascending/descending slope w/200 as the center or something....unlike cutting where you'd use one frequency band(which is one of the reasons I decided to get it as a lot of.my tracks have drums and bass and I want to use range allocation ...)...
Am I on the right track here re-using it...??
 
Last edited:
smg said:
Okay I think I understand it now..using the graph-1k is boosted the most @ 98dB while 900 on the left and 1.25 k on the right are boosted @90 dB and it gradually slopes down and evens out @ 84 dB on the left at 300 hz and at 3k on the left..
Click to expand...

That's basically it. The only thing is that the software I used is meant for measuring speakers, not processors, so the graph is meant to show sound pressure level dB rather than dB relative to a zero point. So the flat line in the middle should be 0dB and the curves should be +/-15dB.
 
smg said:
Alan yeah its 231 excuse me just ordered it today...re-"cut narrow boost wide"you could concievely boost a frequency range using it as well of course...but instead of using just one frequency band you might use a few i.e.100/125/160/200 +250 .maybe on an ascending/descending slope w/200 as the center or something....unlike cutting where you'd use one frequency band(which is one of the reasons I decided to get it as a lot of.my tracks have drums and bass and I want to use range allocation ...)...
Am I on the right track here re-using it...??
Click to expand...

Yes, you are on the right track.

The real limitation with the 31 band EQ is that it only boosts/cuts 3 notes (in 10 octaves). Essentially every G, B and somewhere between E and F. So, for example, if you are trying to fix a bass track where the low A sticks out, you will need to lower both the 100hz and 125hz to target it. (the first harmonic of the low A will be 110hz, that would probably be the frequency that jumps out) This means that you will have to lower the G below it and the B above it. Depending on the slope, the G and the B might actually be attenuated more than the A you want to target.

This is why parametrics are better for that sort of thing.
 
Jay thx for the specifics....given the slope on the specific 1/3 Oct EQ I'm getting (Samson S curve 231)if you're familiar with it,how much do those bands overlap and how much of a slope is iinvolved i..e.do you think its close to the graph Bouldersoundguy posted ?...I'm thinking of using it more for the capacity it has to zero in on a given area albeit w/less precision than the parametric type but definitely more than the shelf type I've been using...
 
Last edited:
I have no idea what the bandwidth is on that eq. Since already ordered it, it really doesn't matter what the slope is specifically, you will simply have to make it work.

They could have designed it so that it has a wider bandwidth at lower gains and a sharper band with at higher gain settings. But again, you are already getting it, so just plug it in and make it work for you.

You have a grasp on the way it works, but you seem to be looking for minute details to visualize what it's going to sound like. Don't bother. Just use it. The details won't change what that unit will ultimately do once you run your sound through it.
 
Yeah thinking the same thing...BTW checked the manual and it says it has constant Q where the bandwith stays the same irregardless of the amount of cut or boost......
 
Last edited:
Just found a graph in the manual that shows the slope of each band...

looks like it only hits the band frequency itself at the maximum amt of boost/cut(+/-12dB)...

at half that on either one (+/-6) it hits a wider area on the asc/desc slopes..I.e.for the 20hz its between 15+25....

the slope shape is consistent throughout the whole range(re-info in post just above)....

All of them overlap and intersect as was discussed here -

Here are some band frequencies and the bottom of the slope frequencies on either side as well.....
[MENTION=17048]20hz[/MENTION] the R/L points are 10+30

200hZ=175 +275

500hZ=350+750

5k =3.75k+7.5k....

If any of the people who have already replied over here have any ideas based on the info I might want to be aware of (or if anyone else does who has used the same EQ)I'd appreciate the continued insight into what I'm going to be dealing with once it gets here(UPS says 7/9)


..



..
 
Last edited:
You must log in or register to reply here.
Back
Top