Line array project

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm scratching a m&k clone idea for the bose L-1 clone. I think my crossover needs will be much easier to reproduce. I'm looking at
264-1062 Vifa TC9FD-18-08 3-1/2" Full Range Paper Cone Woofer
and the
Peerless 830986 3" Full Range Woofer



I notice the bose angles every other driver opposite the other. If I go with the peerless I was thinking just cut it at 80hz and let the sub handle the rest. Any thoughts and how many drivers should be used? Bose say's the use 24 but I was thinking 10-12 would be good for my purpose.


 
I was looking at trying this one out.
 

Attachments

  • Bose_L1-Compact_Array-2.jpg
    Bose_L1-Compact_Array-2.jpg
    120.2 KB · Views: 366
That looks like a disaster of edge diffraction to me, holy crap.

Bose uses a special anti-edge diffraction cone and anti-diffraction coating since they spend so much money on their "research".

Before you build an array, I have a few pages of fun things to read before buying speakers and cutting wood. First up, the different types of them and how/how they don't work.

Live Sound International | Tech Talk: Understanding Line Array Systems

If you are going to build them for home use in a room, they act a bit differently up close than they do at distance. Here are the fun parts of listening to them up close.

http://diy-audio.narod.ru/litr/nflawp.pdf

Here are the results adason achieved when playing around with close out drivers.

http://www.diyaudio.com/forums/full-range/180902-keele-line-array-bad-start-10.html#post2920768

A good comparison between one speaker and a dozen of them--notice how the frequency response shifts.

sony buyout - Page 14

Others have reported the Vifa 3.5 inch full range when stacked around 25 tall sound very good when used with parametric EQ with a DSP. I've completed building a 3-way line array with 12 five inch woofers, 20 three inch full ranges for mids and 48 tweeters...a major pain in the butt. If I had to do it again (under threats of death) I'd go for the Vifa 3.5 inch and EQ to taste.
 
I'm unclear on why the freq response changed.

I think it did not, perhaps what was happening is that due to the higher freq energy beaming, and the LF energy combining, that effect was seen.

The ears might disagree with the measurements, not sure.

One could tell for pretty certain with a very close to the cone (6"?) measurement of a single and the array. Probably would try at the top and dead center of the array, just to see...

at least that is a theory...

_-_-bear
 
...actually if accurate measurements from the same distance are/were made at the same input level (that might be tricky) per driver, then one would see if the response at HF was constant and the LF rises due to addition...

_-_-bear

or maybe the way to go is to normalize at say 1Khz for level at the measurement mic, then compare, etc.
 
Last edited:
I'm unclear on why the freq response changed.

I think it did not, perhaps what was happening is that due to the higher freq energy beaming, and the LF energy combining, that effect was seen.

The ears might disagree with the measurements, not sure.


_-_-bear

I call it the "tilt" effect

Once the center-to-center distance starts to intrude into the frequency response, the drivers start to destructively go out of phase with each other. The line then starts to drop 3dB per octave after that point--I learned that the hard way with my 2-way line array with 5" woofers.

Since the design is for garage use, I didn't want to go DSP just yet so went with a 3-way design to strongly boost the output of the midrange. TV speakers tend to have increasing output as the frequency increases, I used those and they came out smooth in the end. Rising response countered by falling response equals out for the most part.

The destructive phasing issue will show itself in treble response--I just bump up the 10K up a touch on the receiver--done. They work well in a garage or party speakers--have not tested them in the house since the WAF is calculated in negative numbers.
 
Thanks for all the good reading. I am still curious to know if the angle of each individual driver changes comb filtering, phasing and diffraction issues. I was going to use a behringer ultra curve pro for my dsp. Any more thoughts would be much appreciated.
 
Last edited:
...actually if accurate measurements from the same distance are/were made at the same input level (that might be tricky) per driver, then one would see if the response at HF was constant and the LF rises due to addition...

_-_-bear

or maybe the way to go is to normalize at say 1Khz for level at the measurement mic, then compare, etc.

A nearfield measurement of the array will look pretty much as it would for a nearfield measurement of one woofer. The difference is that with a line you can get farther away and still be "nearfield".

The LF gain at a distance is from the in-phase sum of all the elements. You still get gain at high frequencies but it is a random phase summing so it is 3dB for twice the elements rather than the 6dB you see at lower frequencies.

David S.
 
I want to live in theory--everything works there!

Lessons learned by messing around with these funky speakers. They do work in theory the problem is I'd didn't know all the theories! On paper, you stack up 48 tweeters at 90 to 93dB at one watt each--they should be very loud. Compare them against 12 five inch woofers at 85dB at one watt and I was expecting a massive treble response. That did not happen. I had bass response, the mids were falling fast and I had weak treble.

Used the Aurasound 3" full ranges for mids--crossed them at 375Hz and 6K stacked 20 high and ran them at 13 ohms. Their over 10dB rise in output as the frequency went up was tilted down by the phase interference. Cranked the tweeter lines up by running them at 5.6 ohms at 6K and used a bessel filter to throw a mild bump there since the mids were starting to roll of over 5K.

My measurements indicate a +/- 3dB window from 200 to 12KHz at 1.8 meters mic distance. I was not concerned about bass since they couple and I was in a garage and all is well--in theory. A tapped horn will be built soon to shore up the 70Hz and under point so no worries.

Fired them up and the theory train derailed when operating on the reality tracks. My bass response had a thin sound--it was there but at over 3 meters away--it did not sound as full or loud as it did at 1.8 meters away. Threw a quick and nasty sub together and ran it at 80Hz on down...my sub bass was fine but the midbass was thin. I could smell a formula I missed, it was acting very predictable.

It was the critical distance between near field and far field that depends on frequency. The height of my peaked roof in the garage is 4 meters and the arrays are 1.8 meters tall--my bass was rolling off as a point source below around 600 Hz at my over 3 meter listening position. The passive solution was to increase the woofer line height to 3 meters tall or install a ceiling at 2.5 meters so the line height would be over 70% of the ceiling height.

Reality steps in (also known as my wife) the above options would not happen so it is time for EQ adjustments. Flip the loudness control on, adjust the 100Hz bass knob a touch....screw around with the subwoofer controls and I'm close. Since the tapped horn sub build will require a delay, it would be much easier to fix it with parametric EQ and DSP.

My buddy who is a studio sound guy summed it up this way "Line arrays are great speakers as long as you can fix the treble, mid range and bass response oddities". The science fair experiment will continue in the fall when the tapped horn is complete--we will haul them to the local church and power them with a Crown amp that has the dbx driverack to play with. Testing will be as a pair and stacked in a vertical mono stack. The studio guy loves the nearfield sound of them--I told him to get the CBT from Don Keele if he wants nearfield listening in his studio. The column line arrays are
party speakers for the garage after all. :cool:
 
It's hard to say what really is going on there without posted data on individual drivers and summed up response. Perhaps I just didn't look hard enough in your threads.
I've build a number of straight arrays, heard curved arrays and waiting to hear CBT. Like any type of a loudspeaker out there, array can sound great or horrible. It depends on the design.
One of the unique advantages of line array is the stability of the vertical image and perhaps artificial recreation of a life size virtual image.
I certainly disagree that column arrays are Party Speakers only. If you are anywhere near Brooklyn NY, come by to next NY-DIY show/competition. There will be couple of arrays playing.
 
I agree with you R

I originally built them for the garage, mainly to not reflect off the floor/ceiling since it has metal walls. They have that huge sound I love from large speakers and the small variation in volume is great for working in the garage--no getting blasted by output when getting close to them while working on things.

All I'm saying is vertical line arrays have issues--once you are familiar with all the things they do and why they do it--then proceed if the downsides are less than the upsides. In my case, they have a few downsides but the juice was worth the squeeze.

To limit the issue I'm having, the bass output rolls off with distance--make them as tall as possible. The array should get close to the ceiling so the bass reflects from the floor and ceiling junction "making the line array 3 times longer". That detail is something I can't perform in my garage unless I put in a suspended ceiling. The easy way is to put them in the 8 foot ceilings of my house--until my wife spots them then banishes them back to the garage. :eek:

One day I'll drag the things into the house and do some listening tests in a proper room, metal and concrete in the garage is acoustic hell. For now, I just tweak the receiver to get it close enough.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.