Recommended driver for line array?

Jump to Latest
#22
wesayso said:
Digikey over here has over 1200 stock of the TC9 at this moment. Very recent update, new batch of TC9's has recently been made.
Just for your information, you get to choose...
Click to expand...
Yes i know but as i say before, from argentina i cant purchase more than 3 pcs/order and just can get 5 orders/year
Can purchase FaitalPRO locally, so think that is the only option
Searched for other small 3"/4" drivers in argentina without luck
 
#23
I would go for the smaller 3FE22, it is one nice little driver. I have around 10 in different uses and I had to open the box with the 26 pieces I ordered on sale already. I have recently discovered Beyma 3FR30 - that one looks really interesting, I need to get a pair one day. Maybe you can find Beymas as well in your country.
 
#24
pelanj said:
I would go for the smaller 3FE22, it is one nice little driver. I have around 10 in different uses and I had to open the box with the 26 pieces I ordered on sale already. I have recently discovered Beyma 3FR30 - that one looks really interesting, I need to get a pair one day. Maybe you can find Beymas as well in your country.
Click to expand...
No Beyma 3FR30 in argentina :/
 
#25
i made a triple horn for 4x 3FE25, excellent in my case useful over 80 Hz! with Schalmei
 

Attachments

  • 3FE25 16 Ohm x4 mit Schalmei web.jpg
    3FE25 16 Ohm x4 mit Schalmei web.jpg
    113.9 KB · Views: 116
#26
Well im undecided after watch this video
🙁
Will be hard to eq that problems?
Maybe will search for some 2way or 3 way design using Faital drivers :/

Will spend many bucks building the line array 50 drivers for fronts + 3 drivers for center channel + 2 drivers for rear ones. 🙁
 
#27
If I were you, I wouldn't worry too much over something Danny has to say in that video. It won't be smart building array's using large drivers.
Just look up some opinions about the GR channel over on ASR. Most of the time he has an agenda. I remember putting in a few words into the discussion
beneath that video a few years ago. He's putting out "distorted" information with that video, but it isn't his first time. In an exchange we had
due to that video he wasn't abrasive though. He's just a man making a living. Not in a way that I would ever wish to do, selling cables, connectors...

Here's an overview of why full range arrays can work pretty good in a domestic situation:
https://www.rsr-concepts.com/vandermill/?p=1307

While it's mostly about me adding filters to my full range arrays, I do compare them to a single driver in-room proposal.
The array has advantages that don't seem all that obvious looking at it from the outside. Will there be combing in a full range
array's top end? Yes. But you'll also get combing from any other speaker that is placed in a room. Due to the reflections acting
as virtual sources. A long enough array (often referred to to as a floor ceiling array) will take advantage of those mirror sources.

But feel free to change your mind. It is your free choice, I hope I can at least make it a more informed choice.
Much much more information in my thread though. If you really want to know more.
 
Last edited:
#28
Here's the first part of the discussion I had with Danny:

Danny Richie​

2 jaar geleden
There is nothing wrong with a well designed line source. I've designed many of them. This episode is about using full range drivers in a line source and the problems caused from do so.

Danny Richie​

2 jaar geleden
Line sources:

Wesayso​

2 jaar geleden (bewerkt)
And it's (this episode) quite incomplete. Too bad you've never heard a proper implementation. What you show here is that 4 drivers in a row don't make up a usable line array. A proper solution would be a CBT for a finite array, or a floor to ceiling straight array to mimic the infinite type, both with the proper DSP applied. They do work, you know... The fact that you didn't hear a proper one yet does not change that.

Danny Richie​

2 jaar geleden
Actually what you see here in the example with only 4 drivers is much less comb filtering that what you get if the line is longer. The longer the line is the greater the delay between drivers. And curving the line like the CBT lines increases the time differentials further and can make comb filtering even worse. Note that this is referring to using full range drivers (running full range) and not versions supplemented with tweeters using closer center to center spacing. And you can not correct the time delay differentials using DSP. You can only adjust amplitude. If separate amps are used for each driver then you can adjust the time delays but only for one point in space. Change the reference point and again you have a mess.
Minder tonen

Wesayso​

2 jaar geleden (bewerkt)
@Danny Richie that answer shows you don't quite 'get it' yet. Never too late to learn though. Start with some proper theory. May I suggest to start with the papers from Dave Smith from the period he worked at McIntosh...

Danny Richie​

2 jaar geleden (bewerkt)
@Wesayso I'm pretty sure that what I have said has been well illustrated above. I'm not much of a theorist, but I would theorize that there will also be a keyboard worrier out that that thinks they know more than the industry professionals.

Wesayso​

2 jaar geleden (bewerkt)
@Danny Richie I'm not worried, are you? (lol) I specifically pointed towards the work of an industry professional. Others already brought up Don Keele. You've said a lot in that video and in your comments. Sadly, not all of it was that well informed or true. The part about 4 drivers in a row? Yes... it is problematic. I'll give you that. Just don't diss the work of your well established colleagues based on that flawed demo. It could hurt your reputation.

Danny Richie​

2 jaar geleden
@Wesayso Well there was nothing flawed about the demo. Real measurements were taken and the comb filtering was real. If you notice in the video the comb filtering got worse when four drivers were used verse two. That is because the acoustic centers of the drivers were getting further apart as the length of the line increased. It gets worse as the line gets longer. That is the simple facts and simple physics of it. Nothing is going to change that.

Wesayso​

2 jaar geleden (bewerkt)
@Danny Richie Next time (I know there won't be a next time) consider using a real listening distance. You'll see that the combing won't be as severe as this obvious closely measured demo, as the distance from each driver to the microphone (or one's ear) determines the combing at that point in space. With actual reasonable listening distances these concepts can work. (but a finite array of 4 drivers still won't be a good concept, unless you introduce something like the CBT concept) The combing from all stereo speaker pairs is more severe than the combing from an actual line of full range drivers at it's designed listening distance. You know, cross talk from the right speaker to the left ear and vice versa. Browsing your online catalog, maybe you could pay more attention to diffraction. You should really do a video on that.

Danny Richie​

2 jaar geleden
@Wesayso Correct, at a distance further away the distance differentials aren't as great and the comb filtering effect will be more limited to the upper ranges. It will still knock the top octave to so right out of it though. And compensation will be needed to balance out the coupling that will be taking place down low. So the response would still be very unbalanced. That was the point of this video. You can't just make a line of full range drivers and expect the response to be like that of a single driver but with higher sensitivity. And again, as the line gets longer the effect become more like what is seen in the example of four drivers from 1 meter. And curving the baffle back doesn't elevate those issues. It actually causes more delay in longer wavelengths and can make matter worse. And I actually pay a lot of attention to diffraction and surface reflections. I covered those issues in another Tech Talk episode. Check this one out: https://www.youtube.com/watch?v=J5xUioAQ3p4&list=PLUFNGRKZZWXzCt2Syx4yjR4Oy7V-uiePB&index=6&t=728s And thanks for contributing.

​

Click to expand...

(Burried beneith in the comments, starting after a first remark of someone called: Porkchop's Papi)​

 
Last edited:
#29
And the second part: Unsencored:

Wesayso​

2 jaar geleden
The curved type of CBT arrays with full range drivers isn't without it's own tricks, with filtering, EQ and shading, quite nifty actually. It's in line with the changes we see in the industry, moving towards new technology like beam steering seen in the Beolab 90 and that new Lexicon: https://www.stereophile.com/content/soundsteer-technology-lexicon Nor are the full range straight arrays we see today simply a line of drivers. Personally, I have 5 years worth of measurements and DSP corrections on line arrays, I think I know just enough to be dangerous. The game changes with processing, not to fix combing mind you, but to be able to try and mimic an infinite line source. I don't deny there's comb filtering with arrays like these. I've never denied it and before I plunged into my own project of building straight line arrays I had all kinds of back-up plans ready to go, because I was armed with the same rational thoughts and doubts you are displaying here. I had backup plans to use a tweeter, or only use one full range driver for the top end etc. etc. Even plans to do a gradual filtering along the line, so all drivers play the bottom end, using less and less drivers to do the higher end. Sort of a frequency dependent shading. However, I've learned much more once I had build my arrays and started measuring them, from all angles. I published most all of that data on DIYaudio and it isn't anywhere near as severe as what you've shown in this demo. Actually, it blew my mind and I had to re-think and adjust all my previous notions about what was really going on. In the end, out in the room it becomes important how well it can all work together, no point in making the perfect speaker and just dragging it into a room and expect great things. So one has to look at the room + speakers as a system to get ahead. Impossible for the industry that builds speakers, as they are not building our homes, but very doable for the DIY type of crowd. (though the room much too often still is an afterthought in general) If you're willing to do what's needed, these type of arrays can do much much more than most people would hold possible, the DSP can't fix combing, it can't fix the room either. But it can assist in getting a lot of things right, if you're willing to use proper room treatment etc to help make the room and speakers work towards that goal. That's definitely different from stuffing a line of drivers in a box and expect that to make up an array. Cheers...
Meer tonen

perceval2000​

2 jaar geleden
@Danny Richie I'm sorry, I don't have the reserve Wesayso has.... Let me grasp the concept here.... You stacked 4 full range driver enclosures (thus increasing C-to-C even more than a regular line array), measured them at 1m (hardly the 2.5m+ listening position anyone would be at, except for computer speakers, but then again, why would you stack 4 boxes as computer speakers?) and from this very flawed setup, you can surmise that any kind of line array, be it 4 or 40 drivers, are bad at any listening position for any kind of setup. Wow. That's a leap off the deep end! You ignored all the research that has bee done in the last few years, your setup is flawed and not representative of real world application, your measurements are then flawed, and you went into this with old preconcieved ideas, only looking to find a way for the data to match your own dogma. This was not an experiment. this was a setup, making sure the data could be made to follow the author's way of thinking. In science publications, we need to submit our findings to a panel of peers, which will review the data collected and either confirm to publish or scrap for the lack of scientific methodology. I guess there's no such thing on Youtube!
Meer tonen

Danny Richie​

2 jaar geleden (bewerkt)
Wesayso: I agree with what your saying. I have tried everything you just mentioned somewhere along the way. I even worked with some of the best DSP systems available and some not so well known to the public designs that I was involved with in the beta stages. Too bad you aren't closer. Maybe one day we can chat and compare notes and experiences.

Danny Richie​

2 jaar geleden (bewerkt)
perceval2000: It sounds to me more like the measured data that I presented didn't conform to your preconceived ideas. The line of 4 drivers measured at 1 meter is very much like a line of 20 drivers measured at a listening distance. Been there, down that, won a Golden Ear Award for it. I've designed dozens of line arrays and have been doing this for a while. http://gr-research.com/images/2021.jpg

Danny Richie​

2 jaar geleden (bewerkt)
Another one of my line source designs: http://gr-research.com/images/valin.jpg

Danny Richie​

2 jaar geleden (bewerkt)
This full open baffle line source was a really good one too. It took "best sound at the show cost no object". http://mockingbirddistribution.com/mockingbird-audio-lsx-speaker-system/

Danny Richie​

2 jaar geleden
I also did an episode on line sources: https://www.youtube.com/watch?v=-GVdYAQkD8I&list=PLUFNGRKZZWXzCt2Syx4yjR4Oy7V-uiePB&index=14&t=0s
Click to expand...
As said, quite a civil talk and in the end, no foul words from Mr. Richie. Just don't buy his "tube connectors" or any wires 😉.
 
#30
Well, he isn't 7-12' away either.

For my little 4 x tg9, I find you have about a foot of vertical listening area 8' away.
Bigger drivers would also have less horizontal dispersion.

Add more drivers and you will lose high end.......
But you can eq it back in (according to others).

I've only made (had) 2 x 4', 4 x 4" focused, 9 x 4" focused, 4 x fe87 flat array, and lastly 4 x 3.5" flat vertical array.
 
#34
wesayso said:
25 driver unshaded array for me for 5+ years, converted to frequency shaded array's last year. Next move is to replace the 50 TC9's with Scan Speak's 10F 8414G00. I wouldn't do that if I believed the concept was that flawed. Drivers were ordered last year, waiting on delivery.
Click to expand...
big difference between shaded and unshaded? Thinking of a in-wall line array for my LCR in the future home theater. 25 pcs 3fe25 per speaker. Will use DBA up to about 100 Hz were the room dimensions set the limit.
 
#35
Cleaner sound with slightly less room influence with the shaded lines.
(in-wall would avoid diffraction even more, might want to treat first reflection points)
Less combing as well. So yes, I'd go that way for HT. Even though I have no real center channel, HT is a blast with these speakers.
I could use a bigger screen (only have a 55" TV) so it should be fun with a real 'big screen'.
Be sure to plan the needed EQ, preferably with FIR filter capability.
 
#37
On this page: https://www.diyaudio.com/community/...y-for-wall-or-corner-placement.337956/page-28 you can find a model for VituixCAD for an array with the TC9. It shouldn't be too hard to edit it (things like impedance plot etc.) to fit your wants/needs and start simming. This model probably still has the ground mirror as a separate array, with the floor and ceiling setting in Vituix, this isn't needed except if one wants to simulate an in-room IR. Just rebuild it to fit your needs.
It probably is worth it to read part of that thread and bit's and pieces are in my huge thread as well. Search for Vituix or even Virtuix as I misspelled it quite often in my early days of running sims 😳.
 
You must log in or register to reply here.
Top Bottom