|
Home » Audio » Speaker » Interesting New Line Array Design
| Re: Interesting New Line Array Design [message #61160 is a reply to message #61159] |
Mon, 12 October 2009 07:20   |
Marlboro
Messages: 403
Registered: May 2009
|
Illuminati (1st Degree) |
|
|
Does anyone know the limit of actual descernment for a healthy adult human of maybe 35 years old in +/- db in a frequency response in music?
I mean, is it possibly to actually hear the what no smoothing and 1/3 octave smoothing shows? Maybe the 1/3 octave smoothing is actually more like what we can hear anyway.
Marlboro
|
|
|
|
|
|
| Re: Interesting New Line Array Design [message #61163 is a reply to message #61162] |
Mon, 12 October 2009 11:19 |
 |
Wayne Parham
Messages: 18787
Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
Yes, I agree. Posting response of an equalized speaker indoors makes any comparison to other systems impossible.
About smoothed verses unsmoothed, I think that the overall spectral balance is most important, which does sort of correlate to a 1/3rd smoothed chart. But I also think that subtleties show up in the unsmoothed chart, and those are definitely audible in an A/B comparison. Even without an A/B comparison, I think a speaker with a lot of ripple often sounds more interesting at first, but becomes less appealing when lived with for a while.
The coloration that's "interesting" often makes some things sound better to the listener, other things sound worse. And every listener is different, so what may sound good to you may not sound good to someone else. To me, the best policy is to be as true to the source as possibe. Sure, it's all about creating the illusion of being there but I think that illusion is usually enhanced when coloration is reduced.
|
|
|
|
|
|
| Re: Interesting New Line Array Design [message #61166 is a reply to message #61165] |
Mon, 12 October 2009 14:39 |
|
Wayne Parham
Messages: 18787
Registered: January 2001
|
Illuminati (33rd Degree) |
|
|
Yeah, I noticed that too. Measurements taken at different locations and averaged together will tend to smoothe the curve.
This is sort of how the multisub approach works to smoothe the modal region. Where there would be dip from destructive interference between a single sound source and a reflection, adding another sound source placed appropriately fills in the dip. With enough sound sources, the overall sound field through the room is made smoother.
The thing is, when a single speaker is placed in the room, you don't get that effect. Measurements taken at various locations and averaged together tell you what the sound field would look like if there were multiple loudspeakers in the room. But if multiple loudspeakers aren't used, I'm not sure it is valid to average the measurements as if there were. It provides smoothing from dense interference that that wouldn't actually be there.
Not only does this approach smoothe the effects of room modes (which would then make it closer to an anechoic measurement) but it also smoothes the effects of destructive interference between drivers in the loudspeaker, itself. I think it sort of makes sense to try and remove the effects of the room, but by averaging the charts from multiple locations together, you don't see nulls that may form even when the loudspeaker is in a true anechoic environment. I think it is probably better to measure anechoically to begin with, as a sort of baseline. Indoors measurements might be useful too, but for a different purpose.
Then again, I think the indoors/averaged method is interesting, sort of like what Keith Larson was talking about in the thread called "Comb Filtering Misconceptions". Keith was talking about averaging over time, but averaging using measurements taken at several locations also shows the average power response, sort of an average distribution of energy in the room. It won't show where the lobes and nulls lie, of course, but it will give an idea of the general spectral balance.
|
|
|
|
| Re: Interesting New Line Array Design [message #61232 is a reply to message #60571] |
Mon, 19 October 2009 10:49 |
Rocket
Messages: 5
Registered: August 2009
|
Esquire |
|
|
Please let me save you a lot of heartache. Please.
I built a Russell IDS25 clone - 'used the HiVi B3N drivers and a Behringer EQ. It was cool for a while by virtue of "hey cool, I built an array and it sounds impressive." And it was... for a little while.
But look, you are far better off with a 2 way array.
After A/B comparos to my little Usher X718's, there was just no competition. The little cones trying to reproduce that entire spectrum were just distorting like crazy, but you cant really hear it until you A/B em, unless you are golden eared and can get over the "wow" factor of you 1st array. - So I added a sub, and cut the low pass down to 80 Hz. It got a little better, but only marginally. The upper frequencies we just not there, even with gobs of EQ. (and its a killer EQ, BTW.)
Do yourself a favor, get some super cheap woofers around 5" or so, and a line of the PE 5/8" or 3/4" domes within the listening height, you don't have to go top to bottom of the line, and build a 2 way array of modest price. It will be exponentially better than the full range units, and you wont have to pawn off 50 drivers that you have no use for, or sell them to your friend who doesn't know a hill of beans about speakers, just that they are "impresssive and cool looking."
Just don't do it, Man.
|
|
|
|
| Re: Interesting New Line Array Design [message #61264 is a reply to message #61162] |
Sat, 24 October 2009 13:05 |
John L. Murphy
Messages: 3
Registered: October 2009
|
Esquire |
|
|
I've now posted in-room frequency response measurements of the non-equalized corner line arrays. I also expanded the section explaining image analysis.
Here is a link to the new measurement data:
http://trueaudio.com/array/MCLA_array_test_results.htm
I am in the process of putting together distortion data. Initial indications are that the array has greatly reduced distortion compared to a simgle driver. The distortion performance benefits from the -3 dB/octave slope due to the array effect.
Regards,
John
////////////////////////
John L. Murphy
Physicist/Audio Engineer
www.trueaudio.com
|
|
|
|
|
|
| Re: Interesting New Line Array Design [message #61448 is a reply to message #61265] |
Mon, 23 November 2009 10:00 |
John L. Murphy
Messages: 3
Registered: October 2009
|
Esquire |
|
|
| AudioFred wrote on Sat, 24 October 2009 15:33 | John, thanks for posting the response data. I would also be interested in your subjective reaction to their sound.
|
In my subjective opinion they sound excellent, just as the measurements imply. Over the years I've grown to love the midrange clarity of small full range speakers but I had to tolerate the limited output level of my single driver systems. Now I have it all. The MCLA is a crossover free full range system with the midrange clarity of a small full range driver but with bass extension and output level to spare. The listening does not get any better than this, IMHO.
I've posted more comments on how they sound on the Parts Express forum here:
http://techtalk.parts-express.com/showthread.php?t=214300
Cheers!
John
////////////////////////
John L. Murphy
Physicist/Audio Engineer
www.trueaudio.com
|
|
|
|
| Re: Interesting New Line Array Design [message #61450 is a reply to message #61232] |
Mon, 23 November 2009 11:04 |
John L. Murphy
Messages: 3
Registered: October 2009
|
Esquire |
|
|
| Rocket wrote on Mon, 19 October 2009 11:49 | Please let me save you a lot of heartache. Please.
I built a Russell IDS25 clone - 'used the HiVi B3N drivers and a Behringer EQ. It was cool for a while by virtue of "hey cool, I built an array and it sounds impressive." And it was... for a little while.
But look, you are far better off with a 2 way array.
After A/B comparos to my little Usher X718's, there was just no competition. The little cones trying to reproduce that entire spectrum were just distorting like crazy, but you cant really hear it until you A/B em, unless you are golden eared and can get over the "wow" factor of you 1st array. - So I added a sub, and cut the low pass down to 80 Hz. It got a little better, but only marginally. The upper frequencies we just not there, even with gobs of EQ. (and its a killer EQ, BTW.)
Do yourself a favor, get some super cheap woofers around 5" or so, and a line of the PE 5/8" or 3/4" domes within the listening height, you don't have to go top to bottom of the line, and build a 2 way array of modest price. It will be exponentially better than the full range units, and you wont have to pawn off 50 drivers that you have no use for, or sell them to your friend who doesn't know a hill of beans about speakers, just that they are "impresssive and cool looking."
Just don't do it, Man.
|
Rocket, what frequency response did you achieve with your B3N array? I would think they would sound great. Could you show us a measured frequency response? Given the "pink" response of an array it is entirely possible that the high end really was "just not there".
I have to disagree that a 2-way array would sound better...especially one made with low quality transducers. I've been listening to an excellent 2-way line array for better than 30 years but I now prefer the new full range array for its midrange clarity. There are no missing highs in my opinion. Here is my measured frequency response averaged over the listening area:
You said your system was "distorting like crazy" but I wonder if it was actually the loudspeakers drivers distorting. Pro audio equalizers can be difficult to interface to home hi-fi systems due to the different operating levels and it is not unusual to overdrive the source feeding the EQ. The ND90 array has very low distortion at even the highest sound levels. Here is what I measured from one array driven at 50 Hz with 1 Watt:
With 94 dB SPL of output (at 1 Watt) the 2nd harmonic is down -41 dB from the 50 Hz fundamental for 0.9% distortion while the 3rd harmonic is at 1.4% distortion. This is as good as or better than the distortion performance of many subwoofers.
Cheers!
John
////////////////////////
John L. Murphy
Physicist/Audio Engineer
www.trueaudio.com
|
|
|
|
Goto Forum:
Current Time: Fri Nov 22 22:17:42 CST 2024
|
Posting Rules
Members
Search
Help
Register
Login
Get a T-Shirt!
