trying to solve listener fatigue

As I understand it, my system has pretty bad listener fatigue. After a song or two, usually 5 minutes or so, my ears feel like they are wincing or retreating from the sound. It feels like the higher frequencies are mostly responsible for this. I keep the volume at a reasonable level, though it's high enough that a casual conversation would be difficult. The system is an Eclipse 7200mk II active with time delay, levels, and eq (active crossover is only highpass on the midrange, and lowpass on the sub). Amplifier is an alpine pdx-5 that is feeding 4x100w to Morel Hybrid Ovation 6" components. The Morel passive crossover is bi-amp capable and feeds each of the tweeters and midranges separately (to keep time delay and levels intact). The tweeters are soft domes, not metal. Subwoofer is an Infinity 10" inside a transmission line. I have two guesses. I think the EQ could to be dialed in better, though it "measures" well on TrueRTA with my calibrated mic, and it sounds great too for short periods of time. The other idea I have is to better install the tweeters. I've read lots about diffraction by sharp edges very close to the tweeter (there are many). I also think reflections could be making things worse, there are plenty of hard surfaces nearby and the tweeters are sitting on the far back of the dashboard near the windshield. I plan on building a spherical enclosure for the tweeters with smooth transitions and a large radius, mostly to play and see if I can tell the difference. I wonder if this will help, because it will also bring the tweeters out from the back of the dash and let them sit more "out in the open" and further away from other hard surfaces. Any other ideas? -J Malcom: "This is the captain. We have a...little problem with our engine sequence, so we may experience some slight turbulence and then...explode."

Jayne: "We're gonna explode? I don't wanna explode.

Even with time delay, levels, phase, and all that other stuff accounted for ... not having the mids and highs in the same physical location will always have fatigue? I see many builds separating the midbass from a mid/high cluster in a 3-way system. Do I understand this is a frequency dependent issue? A different location at a lower crossover point (~200hz) is acceptable, whereas a higher crossover of (~2khz) is not? I plan to sell my current car soon, so I can't do anything special just yet. I love having the stage up on the dashboard in front of me, so I'm afraid putting the tweeters in the doors will move it too far down. If I want to make the best of the gear I have now, is fighting diffraction and moving the tweeters out from the corner of the dashboard an improvement? Or would I be chasing pipe dreams... and a 3-way setup is the way to go? Many thanks, -JMalcom: "This is the captain. We have a...little problem with our engine sequence, so we may experience some slight turbulence and then...explode."

Jayne: "We're gonna explode? I don't wanna explode.

Yes, it's done often, but that doesn't mean it works well.  A rule to live by is "where ever you place a tweeter, place a mid and vice-versa." the12volt • Support the12volt.com

I suggest you might also want to try attenuating the tweeter.  The Morel crossover has a built-in -3db pad you can engage which might help.  But as mentioned the best solution is to mount the tweet within 1/2 wavelength of the crossover point to the mid.  The crossover in this system is at 1800Hz, so 1/2 wavelength would be approx. 4 inches.Support the12volt.com

Hmm, good to know. I haven't heard that one before. Does that also mean if the midrange/midbass crossover is say ... 300 hz, they should also be within 1/2 wavelength of eachother? (~2ft.) Thanks again guys, -JMalcom: "This is the captain. We have a...little problem with our engine sequence, so we may experience some slight turbulence and then...explode."

Jayne: "We're gonna explode? I don't wanna explode.

teenkertoy wrote: Hmm, good to know. I haven't heard that one before. Does that also mean if the midrange/midbass crossover is say ... 300 hz, they should also be within 1/2 wavelength of eachother? (~2ft.) Thanks again guys, -J

Correct, if the goal is proper imaging.  And frequencies below about 80Hz tend to be non-directional so that rule does not apply to subwoofers.

Support the12volt.com

*I* expect that your fatigue is from the corner loading of those tweeters, IN ADDITION TO the extreme mid/tweeter separation. The corner loading WILL cause major FR anomalies, most especially in the lower registers being fed it. Move them. Really. My mid/tweeter pods are in the doors in custom pods, and I promise you, I have no fatigue... Others on the board have heard the system, and can vouch for the image. Even mid way up the doors, the TD, the careful time alignment and crossover frequencies and slopes, put my image SQUARE in front of me.

haemphyst wrote: *I* expect that your fatigue is from the corner loading of those tweeters, IN ADDITION TO the extreme mid/tweeter separation. The corner loading WILL cause major FR anomalies, most especially in the lower registers being fed it. Move them. Really. My mid/tweeter pods are in the doors in custom pods, and I promise you, I have no fatigue... Others on the board have heard the system, and can vouch for the image. Even mid way up the doors, the TD, the careful time alignment and crossover frequencies and slopes, put my image SQUARE in front of me.

Yep.  Haem's car sounds *great.* 

Support the12volt.com

haemphyst! I am still looking forward to hearing your car someday and comparing notes on transmission line subs : ) I remember you were a big fan of using very steep electronic crossover slopes. What is the advantage compared to a more gentle slope or 6? or 12db? Do you hope the mid and tweeter have less overlap, and therefore less interference with eachother? Thankfully my tweeters are easy to move, so I'll try a few different places on the door and near the mids. I also found Jon Whitledge's sprinter van build and leaned a ton from his notes. In particular it helped me realize I was tuning for a flat unweighted response, whereas a b-weighted response is a better goal. This little java applet helps to visualize the interference patterns between the mids and tweets at various crossover frequencies and distances. I found it useful. -JMalcom: "This is the captain. We have a...little problem with our engine sequence, so we may experience some slight turbulence and then...explode."

Jayne: "We're gonna explode? I don't wanna explode.