Dip in frequency response, fix ideas?

[nonsuchpro]

Originally Posted by Melo theory

I can't picture this speaker.
You want to use these full range you say....but you are going to use tweeters.
And mids? What like alternate the hi waves with the mids to smooth out the dip?

I know I'm not doing this conventionally. Also I should have said super tweeter.
I wanna try something different.

[nonsuchpro]

I may not have to do EQ if I can get all the frequencies in there. That's where I want to brainstorm.

[nonsuchpro]

What about these with that narrow band filter?

https://www.parts-express.com/pe/sho...number=275-010

[Melo theory]

Hey sorry nonsuch.....I forgot all about this thread!
Are you just going to roll those tweeters in with a cap? Or a proper crossover to the mids.
What frequency if a crossover?
Are you going with a line of tweeters?
The reason I ask is that unless the line is short and you are out of the fresnel zone you will probably need eq anyway even if the crossover is low enough to avoid CTC spacing issues.
But if the tweeters are rolled in, and the hiwaves are run open, you will have CTC lobing issues.
You said you wanted to go with 12 driver line?
So that's 4 x 12 = 48"
You would have to be pretty far back to be out of the fresnel zone of higher frequencies.
Which means you will need eq to combat HF roll off due to CF.

[nonsuchpro]

I haven't got it all figured out yet. If I cross the tweeters to the BMRS it will be at the point the FR drops. I think it's like 13-14k but I would measure it once it's in the box. The room is very dead so I will have a HF issues off the bat. In the past, speakers I've made for that room I would make the tweeter louder. What I'm worried about the most is the dip at 1.5k. https://www.parts-express.com/pe/sho...number=275-010 would combat that if I do a narrow band filter. 4th or even 8th order. So there will be 3 drivers x however many. I don't think I have to do 1:1:1. I might be able to do 1:1:2. The BMRs are 4 1/4" so 12 would be 5"
Distance is only 10' back.

[Melo theory]

I wouldn't try to do this with passives.
There are just too many anomalies to consider.
You are defiantly going to get HF combing, I think that dip is only 1 piece of the CF pie....
I don't trust their graph anyway. It looks like its 1 octave smoothed, and if so, that dip wouldn't show up so pronounced. I'm sure there is a phase issue going from piston mode to flop flop mode

[nonsuchpro]

Originally Posted by Melo theory

I wouldn't try to do this with passives.
There are just too many anomalies to consider.
You are defiantly going to get HF combing, I think that dip is only 1 piece of the CF pie....
I don't trust their graph anyway. It looks like its 1 octave smoothed, and if so, that dip wouldn't show up so pronounced. I'm sure there is a phase issue going from piston mode to flop flop mode

Ok well how about the BMRs crossed over at say 1k to the 1/2" inverted dome mid/tweeter (non-comb filtering) and cross that at 15k to supertweeters (or at least tweeters that are strong to 20k+). I can do this with a 24db electronic stereo 3-way. Question is, do I need to do 1:1:1?

[Melo theory]

Well, I'm not sure if those 1/2" tweeters can boogey that low, but if you could do that why would you need super tweeters?
What's 1:1:1? Is that the ratio length of each line?
Do you want to send me one of those hi waves for measurement?
I can measure one of those tweets to if you want.

[nonsuchpro]

Originally Posted by Melo theory

Well, I'm not sure if those 1/2" tweeters can boogey that low, but if you could do that why would you need super tweeters?
What's 1:1:1? Is that the ratio length of each line?
Do you want to send me one of those hi waves for measurement?
I can measure one of those tweets to if you want.

Well, the 1/2" mid/tweeters have a 520 Hz Fs. I'm sure they will do 1k. They say that it will do up to 15k without comb filtering. I can cut them off there and add a super tweet to go up from there.
Thank you but I do have the ability to test, I just need to get a new mixer. I did WT3 10 of them, here are my results for the BMRs:



http://i39.photobucket.com/albums/e1...chpro/bmrs.jpg

[Melo theory]

Yeah, they have a low Fs but they can only produce 72 db max each.
You would need 250 of them to produce 99db at a 1kHz crossover.
Its not only CTC spacing that causes CF, its the fact that its a lIne source.
Its the different arrival times along the line that causes CF.
You need a rising response in the drivers to stay flat with a line source, or you will have to eq.