Re: Smoothing in Frequency response graphs [message #24190 is a reply to message #24187] |
Wed, 05 March 2008 17:56 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
![](http://audioroundtable.com/forum/images/custom_avatars/5.gif) |
Wayne Parham
Messages: 18752 Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
I have more experience with point sources, but I have measured arrays too. One thing I noticed right away is the absence of the floor bounce notch. Any time I measure a tower speaker or a monitor on a stand, I see a notch from floor bounce. Arrays don't have this because of the number of drivers.Because of this, ground-plane measurements took on a slightly different meaning to me for arrays. The ceiling and ground are like reflectors of the line. Jim Griffin and others describe this, and it surely makes sense. The ground reflector isn't a good thing if there is a single woofer at ear level, but it's fine if there is an array of woofers from the floor up. Do you have any other observations about measurement techniques specific to arrays? They're definitely a different breed than point source speakers. On a similar topic, one thing I suggest to guys with deep pockets is to use several subs in a room. It does sort of the same thing that vertical arrays do to remove floor notch. Using a few strategically-placed subs balances bass throughout the room. One can use a program like CARA to decide where to place the subs. Earl Geddes suggests a random placement, with one sub in a corner, one sub above mid level height and one or more subs placed randomly in the room. I agree with him on the idea of using multiple sources to improve the uniformity of bass energy throughout a room, although I prefer trying to keep the subs symmetrical and somewhat close to the mains, if possible. My suggestion is a layout that is more like an oblique array, with a few subs at different points along the horizontal and also at different heights. Most importantly, I suggest checking any proposed placement with CARA rather than depending on randomness to take care of averaging. But whichever way you slice it, the idea is to use a handful of subwoofers to smooth room modes.
|
|
|
Re: Smoothing in Frequency response graphs [message #24191 is a reply to message #24188] |
Wed, 05 March 2008 21:05 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
Keith Larson
Messages: 34 Registered: May 2009 Location: Boston
|
Baron |
|
|
Hi Ed In actuality my configuration is one step further from traditional Since I have not yet actually taken the step of converting the EQ to a standalone op-amp circuit. My EQ is in fact a purely RC circuit that filters the response coming from a first amplifer (a NAD acting as a big gain stage). I follow this with a Hafler DH500 after the appropriate attenuation. If you think of each amplifier as nothing more than a big op-amp (thats what they are), it makes sense. The DH500 does indeed drive a conventional crossover (one 12", 4" mids and 3/4" tweeters). I also have an 18" sub that extends the low end into the mid teens. Best regards, Keith Larson
|
|
|
Re: Smoothing in Frequency response graphs [message #24192 is a reply to message #24190] |
Wed, 05 March 2008 21:28 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
Keith Larson
Messages: 34 Registered: May 2009 Location: Boston
|
Baron |
|
|
Hi Wayne, If you want to get into the math behind a line source its not realy all that bad. The trick is knowing that you need to sum pressure (volts) and not power (watts). The 'power' of a point source does indeed fall 1/R^2, but the pressure is 1/R. That is, power is propertional to V^2. I learned this back in 1985 when I was in college and built my first line source system. Even though I dont think he really knew what I was up to, it was my DSP professors that helped me figure out the 1/R issue. When I mentioned the 1/R^2 power relationship he quickly picked up on the fact that you don sum power, you sum volts. The model I have predicts many of these effects quite nicely, but it lacks floor and boundary conditions. This would not be hard to add, but for best results reflectivity and absorption would need to be known. Or, you measure and then make the appropriate corrections. Hope this helps, Keith Larson
|
|
|
Re: Smoothing in Frequency response graphs [message #24193 is a reply to message #24192] |
Wed, 05 March 2008 21:51 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
![](http://audioroundtable.com/forum/images/custom_avatars/5.gif) |
Wayne Parham
Messages: 18752 Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
I've never really studied arrays, although I do like the concept very much. My focus has always been on point sources, both direct radiating and horns.I guess I may have kind of gone off on a tangent on the floor bounce thing, but we were talking about measurements and it is one of the things I noticed about line arrays. They don't have a distinct floor bounce notch even when standing upright because of multiple path lengths from driver(s) to floor to microphone. So I was thinking out loud that ground plane measurements probably make more difference in the response graphs of point source speakers than they do of line arrays. I like the idea of multiple subs for a similar reason, smoothing room modes.
|
|
|
stereo subs vs all around the room subs [message #24194 is a reply to message #24190] |
Wed, 05 March 2008 22:01 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
Marlboro
Messages: 403 Registered: May 2009
|
Illuminati (1st Degree) |
|
|
I know that bass below a certain point is supposed to be omni-directional, but I've heard the omnidirectional sub and I've heard systems where there are two subs in stereo configuration. The two sub system beats the omni hands down. Of course people have shied away from that by limiting bass in theeir MTM's or whatever, and then just using a sub for sounds below 60 hz. I still think they would do better with a threeway, and using at least two high X-max woofers calibrated to go down to 20hz. 4 would be better, 6 even better. sound volume could go up, but would go down is distortion in the bass range. I hope to do that some time in the future but don't have the funds at the moment to buy 4 more $130 woofers($520) + the cabinets. Marlboro
|
|
|
|
Re: stereo subs vs all around the room subs [message #24196 is a reply to message #24195] |
Wed, 05 March 2008 22:24 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
Marlboro
Messages: 403 Registered: May 2009
|
Illuminati (1st Degree) |
|
|
Understand what you are saying. I'm just telling you that my personal experience in comparing an omni with a stereo sub system is that the stereo sub system beats it out in perceived quality of the bass in the music presented, at least in the homes where I heard them. I've no doubt that you are right; my ears also present me with a different view. Now, I've never heard an omni system with 5 subs spread around the room. Perhaps I should keep my current system and ad 3 or so spread around the room. Of course If I were to add 4 more subs to my ccurent array bringing up a true bass line array to go with the 34 mid ranges and 60 tweeters, that would be A SOUND SOURCE THAT IS VERY LARGE. Marlboro
|
|
|
Re: stereo subs vs all around the room subs [message #24197 is a reply to message #24196] |
Wed, 05 March 2008 22:37 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) ![Go to next message Go to next message](/forum/theme/AudioRoundTable/images/down.png) |
![](http://audioroundtable.com/forum/images/custom_avatars/5.gif) |
Wayne Parham
Messages: 18752 Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
I agree with you about stereo subs. One of my favorite implementations is a three-way speaker, with horn-loaded tweeter and midrange that goes down to nearly 100Hz. The woofers are only used up to ~200Hz, and are acoustically close to the midhorns. Extra subs can be added if desired, placed fairly close to the mains but spaced enough to average room modes. These extra subs are crossed-over very low, with low-pass cutoff well below 100Hz. This prevents abnormal localization of sounds and keeps all sound sources within 1/4λ of their adjacent subsystems.Bass is omnidirectional unless the sound source is very large. But most subs touch the midrange, which really starts around 80Hz. You can localize sounds from about there upward in frequency. That's one reason why you can sometimes tell that bass is coming from a sub if it's far away from the mains. There is a completely different issue that is sometimes a problem, and that's room modes. Sometimes they're not bad, thanks to framed drywall construction which tends to damp room resonances somewhat. The walls have some give, and that helps a great deal. But room modes can make certain bass notes drop out in specific places in the room. The cause is interference between a reflected wave and the source, or between two reflected waves. It makes pockets of high and low energy throughout the room at frequencies below about 300Hz. When you use multiple strategically-placed subs, you can fill in the holes. A notch formed by interference between one sub and a boundary's reflection is energized by a different sub that doesn't have destructive interference at that particular location.
|
|
|
Did you Try EQ ? [message #24199 is a reply to message #24186] |
Thu, 06 March 2008 15:06 ![Go to previous message Go to previous message](/forum/theme/AudioRoundTable/images/up.png) |
chris
Messages: 48 Registered: May 2009
|
Baron |
|
|
That is what Roger Russell does. Hey, got an idea ??? Buy some old Bose 901's, and simply array the drivers vertically, LOL Seriously, did you TRY to eq the midrange drivers ?? What you MAY have been hearing is the drivers inherent roll off ?
|
|
|