Need some measurement help

[Mazeroth]

It’s a little long, but I could really use the help :B

I have a pair of line arrays in my basement listening room. I’m using a DCX2496 to handle crossover duties and have it set at a Butterworth 24db/oct at 2.25khz for both the woofers and the tweeter. The sound is incredible. What I did next was purchase the Behringer ECM8000 measurement microphone, an M-Audio Mobile Pre which is an external soundcard for my laptop, and TrueRTA with the 1/24th octave resolution. I then used the Quick Sweep function in TrueRTA on each speaker to get their curve and captured this into TrueRTA. I then used the EQ built into my DCX2496 and got both speakers almost perfectly flat, all the way down to 40hz (they cross to my IB at 60hz). The before and after is absolutely astonishing. I have my wife turn the EQ on and off on the DCX2496 and I can’t believe how much better the EQed setting sounds compared to the unEQed. I thought they sounded terrific before I EQed them, but now they sound better than anything I’ve ever heard in my life. Now, my questions:

I’m sure my methods of measuring and EQing my speakers aren’t optimal, but the results are absolutely staggering. I did only EQ for my listening position, and I know most people recommend doing a few different spots, but I’m the only one that listens to music and since it is my spot, I figured it would be best to EQ it for that. I’ve also read that an RTA isn’t optimal for measuring.

With that, how could I go about measuring the “right way”? I love the DCX2496 and couldn’t imagine taking a step back to passive crossovers, and I know a lot of the measurement packages are for measuring and designing passive circuits. Yes, this is the first pair of speakers I’ve ever built and I jumped right in, built some line arrays, and got the DCX2496 because of the ease of use and all the features.

Also, I’ve read that applying anything more than 3-6db of EQ is a bad thing. I’m at work right now but I know that some of my EQ settings, mainly in the bass region, are for more than 6db. For my ribbon tweeters (Fountek JP-2.0) I didn’t need to apply any EQ at all above 4khz, which is pretty nice. I think there was a small 2-3db dip around 3khz that I cleaned up with a small filter.

Well, I had a hard day at work and some of this mess probably doesn’t make sense, but if you could digest it and leave me some feedback I’d really appreciate it.

Thanks!

[Mazeroth]

No one loves me? :cry:

[ThomasW]

No sure what to say.......

I use a combination of nearfield and 2m measurements. Nearfield for actually creating the XO, and ~2m for EQing the room

EQing for a single position is problematic since the in-room variations can be huge just moving the mic 18".

There's not much of a problem using more than +6dB for the bass, unless you run out of excursion or amplifier power. Try to avoid boost were the filters are narrow and steep. Those change the "Q".

TrueRTA is okay, but having gated measurements is more accurate.

[Mazeroth]

Try to avoid boost were the filters are narrow and steep. Those change the "Q".

TrueRTA is okay, but having gated measurements is more accurate.

Sorry for the newbie question, but could you clarify what you mean about avoiding boost were the filters are narrow and steep? Do you mean if I have a null of say, 6db at 160hz, and I apply a 5.0Q +6db parametric boost? I'm at work now but I do know I used a few 5+ Q parametric boosts/dips to get my flat response. If this is a no-no I can go home and mess with it to see what I can get out of a lower Q boost.

Also, can you recommend a program that will allow me to do gated measurements on the cheap?

Thanks! ;x(

[ThomasW]

A notch filter is an example of this, it of course is using cut not boost. So avoid a boost filter that has the passband of a notch filter

Play with the freeware program Speaker Workshop.