I used the AudioTools app on my Ipad mini to do the RTA using the internal mic My iPhone 5 playing was playing pink noise through the USA Spec adapter. Supposedly the internal mic isn't half bad but obviously an external calibrated mic is going to be more accurate. I just wanted to see where my system was at and adjust it a little. I don't have an EQ. I played around with the crossovers and settings in the HU and this is about as good as I could get. In the HU Bass is at -5, Treble -3, Bal R2. The bass is a little high bc I like my sub gain up a notch or two I'm not sure what that dip is at 1k -1.2K and if anything can be done to bring it up. It seemed to fluctuate a little if I moved by knee away from from the drivers side speaker.
This is probably a subjective question but how much better is it going to sound if i get an eq and flatten it a little more? I'm not sure if $500 on something like a 3sixty.3 would be worth it. Opinions anyone?

 

Forgot to post the graph. lol

 

Gobs of low end was a given considering you have 2 very efficient 15's and not so efficient 6.5's in the door.

If you don't want to spend the coin for a 3sixty, minidsp is a good budget option if u can diy.

Tune for what sounds best to you, not a flat response.

 

Definitely some dsp needed. What are you running in your car?

 

Polk MM6501 components
A pair of AE IB15's infinite baffle
PPI P1000.1 and PPI P900.4 phantom class d amps. With that 900.4 amp I can go active but I'm not sure it would help.

 

bridge down to 2 channels

Once you get familiar with tuning, go active.

 

sounds good.

 

Audiotools is what I'm using for an RTA. It's a great app. The measurements from the internal microphone are pretty good, but since it's not omnidirectional, the results are a lot more susceptible to variation if you move the mic slightly or hold it at a different angle. Changing the decay setting to "average" and moving the mic around will help with the inconsistencies.

A processor is definitely worth it. Your un-eq'd response looks pretty good, but even a 3-band parametric EQ would make a big difference. If you cut these three spots I think you'd notice a significant improvement:



I wouldn't worry about the slight dip at 1Khz. It's pretty small, and narrow dips aren't very audible anyway.

A processor that allows you to EQ each channel independently will help a ton with your stereo imaging. Even if your summed response looks good, if your left-side response isn't similar to your right-side response, you will have good tonality, but your imaging will be all over the place.

 

Yeah I tried moving the mic around just to see how the signal changed. To get my measurements I reclined my seat back a little and held it pretty close to where my head would be. What made a big difference was removing the silicone case even though it had a pretty decent size cutout for the mic the signal changed quite a bit with it off. Do you use an external mic with yours?
I'm really leaning toward getting the 3sixty.3 It looks like RF is continuing to improve the software / firmware and it's a good sign when they start adding new features, not just fixing bugs. Eventually I might upgrade my front stage. I really want to get that up front bass. The Polks would then probably go in my boat.

 

If you look a the scale, that is hardly a "slight dip". If I'm reading it right it is like 7-10 dB from the points on either side. A dip that big probably has more to do w/ x-over than eq. Also, it is very possibly destructive interference from a lower harmonic (500, 250, 125 etc).

Also, did you use any weighting in the measurement? In a non-weighted reading the bass will appear visually with quite a bit more energy than what is sounds like. The mic will pick up the bass better than our ears will. Look up "Fletcher-Munson" and "equal loudness curves" for more in-depth reading.

Again, tune it to what sounds good to you. Most tuning can be done with x-over points and slopes if you have the ability to adjust them. Then you can use EQ to smooth it out, then time correct if you a) are inclined and/or b) capable of doing so.

Tuning is a really big can of worms... proceed with caution.

Also, to use an RTA to get a really good idea of what is going on you'd want to use something with more resolution than 1/3 octave. The 1/3 octave will give you a rough idea, but more resolution will help identify more issues.

 

Good info. I'll look into the weighting and other things you've mentioned here. I'm just about to order a 3sixty.3 Then the tuning begins I just hope there aren't too many worms in that can...

 

Yea, I'm using a Dayton EMM-6 mic and ART Dual Pre USB preamp with the iPad camera connection plug. (~$130 total) Audiocontrol offers a standalone mic for $200. The Audiocontrol mic would be nice for convenience and portability.

A processor will help a lot with getting upfront bass. With some tuning you should be to cross those IB subs pretty high without it pulling to the rear. The feature set looks really good on the 3sixty.3 and the price is pretty reasonable compared to some of the other processors out there. I'm not sure why it's not more popular.

After cutting those peaks, it would really only be about 5db lower than the surrounding frequencies. From what I understand, even a 7-10db dip won't be that noticeable as long as it's narrow.

I think you're right about the destructive interference. I've got the same dip around 1khz my car and I can boost the hell out of it without making much difference.

I've never really looked into using weighted measurements for tuning. I've just been using unweighted and shooting for a target curve similar to Audyssey's or Andy W's (+9db from 20hz to 60hz, smooth transition from 60hz to 160hz, flat from 160hz to 2khz and slightly downsloping from 2khz to 20khz)

Audiotools has an FFT module that can do up to 1/24 octave resolution or you can turn off smoothing altogether. I usually switch back and forth between the 1/3 octave RTA and the higher resolution to get a better idea of what's going on before making my boosts or cuts. I think I read that 1/3 octave gives a good approximation to what we hear.

Tuning can definitely take some time, but with an RTA and a good DSP you can break it down into a fairly simple process and, IMO, as long as you have decent speakers, amps, installation etc., you'll get more of an improvement than switching to super high end gear.

 

The weighting thing isn't that big of a deal as long as you know what you are looking at. I just find it a little easier to look at, especially on the old Audio Control 3050, when you can see everything on the screen at the same time. Otherwise the bass tends to shoot off the top when you are looking at say 500 and up.

The thing I always look to do first is eliminate any big peaks or dips. Just looking for smooth response. Then I eq to "salt and pepper" to taste if you will.

 

So I've been playing around with the 3sxity software in demo mode until my hardware arrives and I've got a question about xover slopes. When / where should 12db, 24, 36, 48 be used. Is there a drawback to always using the steepest available?
What's pretty cool is that I can do all my intial xover and EQing based on my RTA right now and load it on my 3sixty when I get it.

 

I recommend starting with the steepest possible and flattening out if needed. Generally, the steeper the slope the less chance you have of getting a boost where the slopes intersect.

 

Steeper slopes will mean less interaction between drivers which helps alleviate some of the phase issues you can run into with shallower slopes. 24db per octave is pretty popular for car audio. There aren't really any drawbacks to using slopes steeper than 24db/octave but its kind of a gimmick because it won't really make a noticeable difference going any steeper.

 

Sounds good and then there's crossover class. Linkwitz vs butterworth, bessel or chebychev. I'm somewhat familiar with the first two.

 

I'm not familiar with the other two either. Butterworth is 3db down at the crossover point and LW is 6b down

 

I'll stick with butterworth for now then. I probably won't mess around with t/a until I get more comfortable with Eq but just curious to what your using?

 

I'm using 24db LW, but not for any reason in particular. Different slopes and types just give you more flexibility. You kinda gotta play around with them and see what works best. Time alignment is fairly simple. To start out, just take a tape measure and measure the distance from each speaker to where your forehead would be in the drivers seat. Use the farthest distance as your reference and subtract each shorter distance from that. There are more complicated ways to do it an you may need to make some fine adjustments, but taking simple measurments will get you pretty close.

 

Thanks for all the info guys. The next RTA I post should look a lot better. Honestly this doesn't look like it's going to be that difficult. Being able to use an RTA to see the adjustments will obviously be key, for me anyway. No way I could do it by ear. I want to get a good "by the book" reference point first, by duplicating other's and then maybe customize it from there if I feel like I need to. Some of the reading I've done say there are certain frequency's in the high range that should be bumped to help bring the sound forward. Anyway this should be fun.

 

More worms...

http://www.diymobileaudio.com/forum/...rossovers.html

 

Lol Thanks for the link.
Good thing I'm not afraid to learn or try new things. In my job I'm expected to pickup a 500+ page manual for some device or product I've never seen and learn how to set it up, use it and train other people on it. And the really fun part is I get to document it all.

 

Installed the 3sixty3 today. As soon as I turned it on, without any eq or t/a the system sounded better. Here's my new RTA after about 30 minutes of playing around with the t/a and eq. I need to get a calibrated mic for more accurate measurement especially in the lower frequencies. Apple cuts out everything below 250hz / 24db.