ALK Universal Crossover Squaker Attenuation

[Rudy81]

In the past few weeks I have been on a journey to better my system. This all started when I took the V-trac demo, which led me to get Al's Trachorns. I use the Integra DHC 9.9 pre/pro, which has Audyssey EQ built into the system if you choose to use it. Folks rave about Audysseys capabilities to flatten room frequency response. After many calibrations I feared I may have some serious problems in my room, one is too much absorption in the high frequency range and the other was really bad bass standing waves. I suspected those issues since the sound difference with Audyssey vs. direct was very noticeable. That led to building several bass traps and reducing the number of high frequency absorbers. I also wanted to KNOW what the frequency response in my room was in order to either trust Audyssey or not use it if I found it to be faulty. This has been an eye opening experience. The bottom line is that Audyssey was right on the money, although that is a topic for another thread. In order to measure the room response, I got a Behringer ECM 8000 microphone and the Room EQ Wizard software.

One of the things I noticed on my very first runs was that my mid range, originally set to -6.2 attenuation, was reading rather low in my room. I decided to make several measurements with the mic and software while changing the attenuation settings on the ALK universal. My goal was to get my room response as flat as possible before running the Audyssey calibration in order to have it affect the signal as little as possible.

What I found was very interesting. In my room, the -6.2 attenuation is too much relative to the 75dB reference signal. The Room EQ Wizard is one nice piece of software....best of all its free.

This is the 20Hz-20kHz comparison:

Here is a closeup of the frequency range that changes with the attenuation changes:

This last graph is the effect of my first Audyssey calibration after changing the attenuation to -2.6 dB, the least attenuation I can set on the ALK network.

I have taken many readings with the Room EQ Wizard and most relate to my room acoustics and the changes I have made to create a listening environment with a flatter frequency response. One of the benefits for me is that I got to see exactly the effect of changes made on the ALK and its effect in my room. These results are objective and the good news for me is that they back my original mid range settings that some said were too "hot". Based on my interpretations of these and several other graphs, the low attenuation is not too 'hot' in my room, it is closer to just right. I am really enjoying finally getting to see just what happens acoustically in my room.

I'd love to hear from the experts in the acoustic field. My experience is little, although I have read the Master Handbook of Acoustics several times and have made various acoustic treatments over the years.

[sootshe]

Rudy,

Can you describe to us whether you feel that these new settings actually sound better to you or not. Sometimes I feel that we strive for a flat frequency response at the expense of musicality.

Sometimes flat frequency response is not always the best sound or the sound that our ears appreciate.

[Al Klappenberger]

Rudy,

Only one thing bothers me. Your correction has filled in an 8 dB dip at 400 Hz. Your power amp needs to work 6 times harder there. Be sure you have enough head room to do that safely. One of those one-lunger SET amps, for example, couldn't handle it! Also, it is right at the woofer / squawker crossover. Try inverting the phase of the squawker, just as a test to see if it is actually driver interferance. A crossover glith that is. The overall improvement looks good though.

Al K.

[Rudy81]

Rudy,

Can you describe to us whether you feel that these new settings actually sound better to you or not. Sometimes I feel that we strive for a flat frequency response at the expense of musicality.

Sometimes flat frequency response is not always the best sound or the sound that our ears appreciate.

In all honesty, I have not had much of a chance to just sit down and listen. I was busy for two days making the bass traps and the last day I had off was playing around with the software. What I do know from before starting all this is that the difference between playing music with vs. without Audyssey was very noticeable, particularly in the upper bass region and the high frequency region. This difference was what led me to work on all this.

I'm not sure which I will like more, I would just like to start from a known quantity. I do know that my room had some problems in bass as predicted by room mode calculations. I could hear that at times and Audyssey did take care of that 'boomy' response. It is nice to be able to see it. Based on my reading about room acoustics, I am pretty convinced that room acoustics is just as important if not more so than the gear you are using. I beleive a properly treated room can make a mediocre system sound pretty good, but poor acoustics can make a top tier system sound poorly.

I will have to live with this for a while and see how I feel about it. In a week or so I will likely post my impressions.

[Rudy81]

Rudy,

Only one thing bothers me. Your correction has filled in an 8 dB dip at 400 Hz. Your power amp needs to work 6 times harder there. Be sure you have enough head room to do that safely. One of those one-lunger SET amps, for example, couldn't handle it! Also, it is right at the woofer / squawker crossover. Try inverting the phase of the squawker, just as a test to see if it is actually driver interferance. A crossover glith that is. The overall improvement looks good though.

Al K.

Al, thanks for looking at this. Audyssey is making the corrections automatically. I have no way of seeing what settings Audyssey uses. I only set the system to automatically calibrate and run a series of eight runs. The professional version runs thirty two runs....I keep reading that it is extremely accurate....for $500 more.

Anyway, that dip at around 400 Hz can be seen in every graph I have made of my front speakers, so I don't think it is a a crossover problem. The left main, right main and center speaker results show a dip, or null, in that frequency range. Thus, I suspect it is a room acoustic problem. When I originally ran a room resonance spreadsheet, you can see that there are many axial resonances between 350 - 450 Hz along the room length and width. I hope to be able to do something with better bass trap placement to prevent this big dip in response.

As far as amplification, I should be ok. I have 7 channels of 125 watt power. The amps are Parasound HALO amps and should handle the loads very well with the Khorns, La Scala and Cornwalls.

If, as you suggest, I invert the phase; would that invert the graph response? I don't know enough to guess at the result. The other thing I could do is take a measurement in a different location. If in fact it is a room created problem, the dip will change if I move the mic to a different location ourside the null. I will try that when I get home next week.

[Al Klappenberger]

Rudy,

That 400 Hz dip coud be anything! Some industrial equalizers do not allow positive correction. They provide cut-only equalization. The fact is that dips usualy are not ofensive and usually do NOT need to be corrected. It's the peaks that you hear. I would take readings at a few places to see if the dip is real.

Al K.

[Rudy81]

Al: I will make another set of tests when I get a chance. What do you think will happen to the plot if I reverese the phase and the problem is driver interference? Also, if it was driver interference, would Audyssey not be able to correct the problem? Just for comparrison, here are the plots for the center and right speaker. As you an see in these uncorrected graphs, they so show a problem area in that frequency range. The center is a La Scala and the right is a Khorns, both with ALK universals crossovers and identical drivers. Although, the center has the K400 and the right and left have the Trachorns.

Center speaker:

Right speaker:

[Al Klappenberger]

Rudy,

When and identical sound comes from two sources in two different places, they add to each other in some places in the room and null each other out in others. That shows up as peaks and dips in the frequency response randomly around the room. If it's a room resonance, it would be less likely to change when you reversed the phase of one of the sources. If it's a sum or difference between two sources it will change or the peak will move to another place in the room. I am no expert on room resonances, so I won't attempt to go any further that that to predict what will happen with any authority. I will say that a 1st order network, like the Universal woofer / squawker crossover,I would think would generate interference peaks somplace in the room. This is why I advocate extreme-slope networks. They only allow a single driver to make any give sound so room reflections are all that is left. I would run a curve to use as a reference, then withOUT moving the mike, invert the squawker and run another curve looking for a change in the 400 Hz dip. You might also disconnect the woofer, then the squawker and see if the dip right at the crossover frequency seems to change. Of course, you need to look at the changes only in the frequency range where both are capable of making a good sound. That would be between, say, 300 and 500 Hz. If the dip at 400 Hz shows up with either driver operating, it's a room resonance. If you see it only when BOTH drivers are working, it's driver interferance. AGAIN though: I am no expert in this area!

Al K.

[Rudy81]

Al: That is a great explanation, thank you. The software has the ability to select the frequencies tested, so I can certainly test only the frequencies in question. I will work on this in a few days and post the results. Very interesting stuff, and it makes the case for the ES networks.

[Rudy81]

Al: I just finished running some tests on the 300Hz to 600 Hz area as we discussed a few days ago. The good news, at least the way I interpret it, is that the dips in that frequency range are due to room standing waves at my listening position. I am including a graph where I tested just 300-600 Hz. The green line is with a full driver compliment in my left Khorn. The blue line is the K55 and your trachorn only, woofer was disconnected. The purple line is just the woofer, with the K55 disconnected. My interpretation is that both drivers compliment each other in that crossover range for overall system benefit.

You may notice that compared to my earlier graphs, the dB level is much higher, although the plot is very similar. The reason for this is that I had skipped one step in the software setup. I had not calibrated the ECM8000 mic volume level against a SPL meter to set the proper baseline in the software. Once I did that, the plots reflected that calibration, although the general shape, peaks and dips are the same.

I hope my interpretation of the graphs is correct. If so, I am glad the crossover is not causing the drivers to fight each other. I am really pumped about the improvements I have been able to make in the last few weeks between your Trachorns and an entire rework of my acoustic treatments. I now have large bass traps and just a few high and mid absorbers. My last project is going to be a 1D QRD diffuser for the back wall.

[Al Klappenberger]

Rudy,

I agree. The two drivers are adding up as they should. The other thing you might try is running the same or a slightly wder frequency range at several places in the room. If it's a standing wave null it will show up as a peak at a differnt place in the room. In that case, I wouldn't even try to correct for it. It would be something you should just live with.

Al k.

[Deang]

Great explanations Al and nice work Rudy!

[Rudy81]

Dean: The real story is the work I have been doing on my media room. Long story, but the V-trac demo led me to work on the acoustic treatments in my room. I first found that I had added way too many high frequency absorbers and no bass absorbers. So, I took everything down to bare walls and started over. I needed a way to actually measure any changes to see if in fact they were improvements or not. I didn't want to make the same mistake again.

That's when I found the REW software (Room EQ Wizard). I then built the bass traps, re-worked the high frequency absorbers just for the side image points and today I hope to finish the back Quadratic diffusers. This has been a great learning experience. A side benefit was being able to graph the mid performance for each attenuation level. I am very pleased with how this is turning out.

[Rudy81]

Rudy,

Can you describe to us whether you feel that these new settings actually sound better to you or not. Sometimes I feel that we strive for a flat frequency response at the expense of musicality.

Sometimes flat frequency response is not always the best sound or the sound that our ears appreciate.

I have been thinking about your comment for a while now and now that I am just about finished with my room treatments, I figured I could answer with a little more conviction. The quest for flat frequency response is to prevent your room from modifying the sound originally recorded due to room modes, comb filtering, modal ringing etc. Speaker and equipment manufacturers spend quite a bit of time and money striving for as flat a frequency response as they can get on their gear. Anything other than a flat response is coloration of the sound originally recorded, whether it be from your room, your speakers or other gear. Now, if you prefer whatever colorations exist in your room, great, but that is not necessarily what the artist intended.

I have spent just about every free minute of the last three weeks working on the acoustics in my room. It was time well spent, and probably one of the best improvements I made to my 'system'. Mind you, I have the ALK Universals, ALK Trachorns and Crites tweeters. I have made a lot of changes, but I dare say that the changes I made to my listening environment were just as, if not more, important. Now I can really enjoy the improvements brought on by the ALK upgrades etc.

The beauty of this endeavor, is that it is not just my psychoacoustic brain telling me it sounds great.....I have measured the room response and I can graphically show the improvements to the system. I have spent just a little bit of time listening, but so far I'm extremely pleased with the results. I have removed the old, improper, high frequency absorbers, added a large amount of bass traps, properly placed high and mid frequency absorption to prevent early reflections and added an entire back wall of Quadratic diffusers. I will be started a new thread in the near future with pictures and graphs.

In any event, my thoughts are that obtaining a flat frequency response will only enhance the musicality, not reduce it.