Full range line array + Linkwitz Transform?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have a 12 driver per Channel open baffle line array that uses 4" Fullrange drivers which have opened my ears as to what is possible with fidelity. The are an immense improvement over my Linn Isobariks. They have there own characteristics but when eq'd correctly and with the right material the speakers just disappear....completely.
I have another 32 of these drivers that I would like to use in another line array but this time a sealed box version. I have put the drivers through WIN isd Pro and come up with the following. The drivers have an EBP of 86.3 so strictly speaking are suited to a vented alignment, but putting them through a sealed alignment in WIN gives a box volume of 293.13 litres with a Qtc = 0.795 and an Fsc = 82.83 and a -3 db point of 75 Hz (approx)
Now a box volume of 293.13 litres is way to big so I reduced the box volume to 40 litres and that gave a new Qtc = 1.135 and an Fsc = 125.96 and a -3 db @ 94 Hz. This is obviously unacceptable so I tried the LT. on the WIN filter function and got this.

Fsc (Fo) = 125.96......LT. ......Fp = 40 HZ
Qtc (Qo) = 1.135.......LT........Qo = 0.707

I checked the LT. on one of the online excel spreadsheet calculators and it all looks good with K = 1.2 and the DC gain at 19.93 (could reduce this by lifting Fp to 50 Hz)

Now the -3db is at 40 Hz the response plot is flat (Qo = 0.707) the cone excursion is good to 40 Hz at 5 watts which gives 104 db
In reality I'll rarely, if ever, go beyond 1-2 watts so......

Is this doable or am I missing something. It's not the usual application for the LT. I know but it would appear to be effective for this.

Any thoughts?
 
Have you seen my thread?
http://www.diyaudio.com/forums/full-range/242171-making-two-towers-25-driver-full-range-line-array.html

That's exactly what I'm doing, and my vote would be: Yes, it's doable. Though a lot may depend on the actual x-max of the fullrange drivers you're going to use. Got any specs?
My box volume is 52 litres with a lot of thought and time (experimenting with test box) spend on internal damping.
I use the 3.5" Vifa TC9. Chosen because of it's size and the ability to get them as close as possible. Yours are a bit bigger but get them as close together as possible and you'd probably be alright.

I am doing all processing in my PC, from EQ right down to using FIR filters. Most of it documented in the above thread.

I don't need the right material, they always disappear for me. All that is left is a gorgeous musical landscape. Seriously! It performs really well with all kinds of genres.
My placement is fairly close to the walls to aid the low frequency of these tiny drivers. They actually go down to 20 Hz quite easily. I had expected to get no more than 30 Hz after I ran my own sims. The room matters though, be prepared to work with it.
 
Hi wesayso,
I have seen your thread and have followed it with great interest. It is one awesome build that you have been working on. Tremendous work, way beyond what I can even contemplate.
The drivers do butt up right next to each other so there is no gap between them and they have square frames which is good. The main aim for this next build is to compare open baffle and sealed applications for this driver in a line array. The open baffle line array excels in vocal reproduction and acoustic and live performance as there are zero box resonances and no crossover issues. They run Fullrange with no crossover but with eq.

Hi richie00boy,
you're right about the gain but I can reduce this as mentioned by raising the Fp to 50 or 60 Hz. I was running through the possibility to see if it would work and...... The excursion profile below 40Hz looks fine as it does not go throught the roof suddenly which is what I expected. Again I can lift the Fp to whatever I need to to help in this manner.
In order to do a sealed line array with these drivers I had to force them into a much smaller box than that given by WIN isd Pro and the LT. is the only way. It's going to be a hard sell getting a pair of 16 driver speakers past das wife at 40 litres let alone 293 litres.

So in principle you both think that it will work and worth the effort? I've ordered an LT. pcb from Rod Eliot at ESP so I'll be good to go in the not too distant future.

Cheers
 
Expect to need more than just a Linkwitz transform to get the lines flat in a room.
Here's my pré EQ before I let DRC-fir sort out the rest:
eq1.jpg


WinISD plots only show a 2 dB gain (due to box dimensions) at ~250 Hz but in real life that turns out to be quite a bit more than that due to how line arrays work ;). In my case I did not need to boost the high frequencies but you might need that too.
 
Hi wesayso,
I will no doubt need some eq. I have my line arrays eq'd reasonably flat at my listening position using a Behringer 31 band equaliser which will have to do until I make the move into something a whole lot more sophisticated at a later date (if ever) I'm really using the LT. primarily to remove the increase in Qtc given by the reduction of box volume. I will of course have to measure the built speaker to see what the actual Qtc and Fc are as richie00boy suggested and do the LT. afterwards to get an optimised result.

I have done other LT. experiments with subwoofers (on paper/screen) and it seems to me one has to be "very" careful in its application. Several subs that I put through had massive excursion issues below 30 Hz I was therefore surprised that it seemingly worked so well with the line array, even when asking to do 40Hz. I'll certainly back off from this number when it comes to the build for safety.
 
Don't back off the gain too much... 16 drivers share that boost, I presume in a parallel/series wired configuration leading back to the impedance of a single driver. What seems like huge gain is suddenly not that huge if you look at what it does for one single driver in the array.
If you can, build the box first and measure. Then determine the LT. you need. Depending on where you place your speakers it can vary quite a bit. One of my speakers is in a real corner. The other is not as it's just a chimney beside the speaker. The difference in the bass response without any EQ is huge.

Do you plan to have them close to the wall or more out in the room? Close to the wall is not that bad of a place for line arrays. You just got to watch out for the first reflections, place damping there if you can get away with that :p.

I'm not allowed to place damping panels everywhere I want but you migh have better luck. I've got one huge panel now and two more are planned (actually got permission for them) without cluttering up the room.
2 Of them will be out of sight, one will be/look like a poster.
 
Here's another option - corner loading to bring up the bass https://www.trueaudio.com/array/MCLA_design.htm

To your original question, yes, you can use a linkwitz transform to boost a full range array - if the drivers can take the boost . I have an old audioexpress article that described just such a beast using small full-ish range drivers. Google about for "Unique Stereo Column System" by Roger Russell
 
Hi wesayso,
yes, 4 x 4 series parallel. Although one day I will try all parallel with added series resistance to hear what happens. ��

Hi zobsky,
thanks for the link. I have seen it before but have never bothered to read it, so I did. Very interesting read. My next line array could employ the same principles of construction whether or not it ends up in a corner. Thanks for that. I'll check out Roger Russell as well.
Cheers
 
Hi richie00boy,
yes I realise this will happen so would have to do the LT. accordingly. I have read many different opinions regarding the pro's and con's of series/parallel wiring versus parallel
( + series resistance ) wiring and there does not seem to be a definitive answer. Most come down in favour of series/parallel connection because of the load presented to the amp, but instinctively I cannot help feel that having all the drivers driven in parallel will remove any possible interactions and effects caused by the alternating series/parallel connections before and after any "one" particular driver in a line array.
I must try this on my open baffle line array where an increase in Q will might actually be beneficial.

Cheers
 
Hi wesayso,
yes, 4 x 4 series parallel. Although one day I will try all parallel with added series resistance to hear what happens. ��

Hi zobsky,
thanks for the link. I have seen it before but have never bothered to read it, so I did. Very interesting read. My next line array could employ the same principles of construction whether or not it ends up in a corner. Thanks for that. I'll check out Roger Russell as well.
Cheers

There's also the TLAH, . which uses basically uses the length of the (tapered) corner array as a stuffed transmission line to bring up the bass enough to mate with a subwoofer. Not quite what you asked for, but worth mentioning in the same breath.

TLAH
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.