Hi guys,
I am reviving this post after a long time. Sorry for getting back after a long time.
Today I was revisiting this post from the new perspective of lots of things done and I think the feedback has been very valuable.
Of course because of inexperience, availability, budget and other reasons I didn't followed all the advices.
I want to show how my first Horn proyect turned out:
Fisrt thing was to have a decent look so I was allowed to have these monsters in the living room. they were acepted 🙂
I am reviving this post after a long time. Sorry for getting back after a long time.
Today I was revisiting this post from the new perspective of lots of things done and I think the feedback has been very valuable.
Of course because of inexperience, availability, budget and other reasons I didn't followed all the advices.
I want to show how my first Horn proyect turned out:
An externally hosted image should be here but it was not working when we last tested it.
Fisrt thing was to have a decent look so I was allowed to have these monsters in the living room. they were acepted 🙂
It is far from perfect, but they are my horns, and I think the most rewarding thing now is the whole process of experimenting and learning. So in advance sorry for the design mistakes.
This is the system:
Horn: JBL 2360A with BMS 4590 coaxial compresion driver 8ohms with default BMS crossover.
Woofer: Eminence Omega 18a with Open Baffle U-frame (recycled from an old Lampizator Saba greencones proyect) So it costed zero, just I cutted the upper open baffle structure and painted it.
Amplification:
Horn: Tube 45 Single ended amplifier
Woofer: 25 Watts 151 Wadia Class-D amplifier (also this is what was available).
Crossover:
BMS coaxial driver with default pasive crossover at 6khz with 18db/oct i think.
also a Minidsp for the crossover between the horn and the woofer. i've experimented with diferent slopes 500hz, 300hz.
This is the system:
Horn: JBL 2360A with BMS 4590 coaxial compresion driver 8ohms with default BMS crossover.
Woofer: Eminence Omega 18a with Open Baffle U-frame (recycled from an old Lampizator Saba greencones proyect) So it costed zero, just I cutted the upper open baffle structure and painted it.
Amplification:
Horn: Tube 45 Single ended amplifier
Woofer: 25 Watts 151 Wadia Class-D amplifier (also this is what was available).
Crossover:
BMS coaxial driver with default pasive crossover at 6khz with 18db/oct i think.
also a Minidsp for the crossover between the horn and the woofer. i've experimented with diferent slopes 500hz, 300hz.
Some general conclussions:
I am still looking for my perfect cup of tea.
I think that mids sound wonderful. they are special.
Without EQ highs are weak and not enough when listening to different material.
but when apply Constant Directivity compensation in the MiniDSP, I get good SPL in the high frecuency (measured flat response) but I think detail is not smooth, and sound quality suffers.
My theory is that even though the BMS drivers is coaxial is suppoused to reach 20khz, the horn is not designed to reach high frecuency with Hifi sound quality .
I want to try a supertweeter with better dispersion pattern for the highs (and I presume less distortion at that high frequency?).
I haven't been able to buy a JBL 2405 yet. hard to justify the expense yet....
This week a friend will lend me some 1inch compression drivers with small Tractrix horns he uses with his Fostex BLH. I want to see if those tweeters will point towards the transparency and quality I am looking for the highs.
The low frequency area haven't been solved yet either.
I mean, with certain XO and EQ configuration I can get a nice sound, but is not the autoritative sound quality I would like to get. but well, it's a hobby, so I plan to have fun with the process and experimentation.
Trying to get flat extended response from 40hz to 20khz tend to result in not very pleasant sound.
If I try to have a more shy bass, with horn that is more present, this seems to be way more pleasant (magic experince), because as I said, mids seems to be of great quality.
I am assuming a 15 in woofer is required (not current 18in). and maybe a sealed enclosure (now open baffle), but I am not ready to get a 15in woofer yet.
Any suggestions are more than welcomed.
Best
I am still looking for my perfect cup of tea.
I think that mids sound wonderful. they are special.
Without EQ highs are weak and not enough when listening to different material.
but when apply Constant Directivity compensation in the MiniDSP, I get good SPL in the high frecuency (measured flat response) but I think detail is not smooth, and sound quality suffers.
My theory is that even though the BMS drivers is coaxial is suppoused to reach 20khz, the horn is not designed to reach high frecuency with Hifi sound quality .
I want to try a supertweeter with better dispersion pattern for the highs (and I presume less distortion at that high frequency?).
I haven't been able to buy a JBL 2405 yet. hard to justify the expense yet....
This week a friend will lend me some 1inch compression drivers with small Tractrix horns he uses with his Fostex BLH. I want to see if those tweeters will point towards the transparency and quality I am looking for the highs.
The low frequency area haven't been solved yet either.
I mean, with certain XO and EQ configuration I can get a nice sound, but is not the autoritative sound quality I would like to get. but well, it's a hobby, so I plan to have fun with the process and experimentation.
Trying to get flat extended response from 40hz to 20khz tend to result in not very pleasant sound.
If I try to have a more shy bass, with horn that is more present, this seems to be way more pleasant (magic experince), because as I said, mids seems to be of great quality.
I am assuming a 15 in woofer is required (not current 18in). and maybe a sealed enclosure (now open baffle), but I am not ready to get a 15in woofer yet.
Any suggestions are more than welcomed.
Best
Last edited:
At the moment, I am very excited exploring the idea of a 1inch compression driver with, for instance a Tractrix flare with a cutoff frequency of 1500hz. and crossoed over with a 1rst order XO at arround 8khz.
Not sure how this experiment is going to turn out.
Not sure how this experiment is going to turn out.
looks good. I bet they sound really nice to boot. Cd compensation starts around 6k and goes all the way up. I also use the minidsp between my bass horns and compression. You can find these curves and implement it yourself in the parametric sq on the minidsp. But to be honest you'll do better by running a tweeter. I don't know what the bms goes up too, but the 2360 loses pattern control at the higher frequency. If it were I, I'd only use the up to 7 to 10k. And at the bottom it all depends on you.
there are other options besides super tweeters. For instance I use altec Lansing 812-8d 1 inch compression drivers for tweeters. I tried different tweeters but alway came back to a 1 inch compression driver. I just liked them better then a tweeter. More dynamic in my opinion. But I also never tried a 2405.
I see I missed your second post today. Were pretty much doing the same thing. I am crossing my tweeters at 7k with a first order filter.
Have fun listening to the fruits of your labor. And they do look quite nice. You should be proud.
there are other options besides super tweeters. For instance I use altec Lansing 812-8d 1 inch compression drivers for tweeters. I tried different tweeters but alway came back to a 1 inch compression driver. I just liked them better then a tweeter. More dynamic in my opinion. But I also never tried a 2405.
I see I missed your second post today. Were pretty much doing the same thing. I am crossing my tweeters at 7k with a first order filter.
Have fun listening to the fruits of your labor. And they do look quite nice. You should be proud.
Many thanks Nhuwar!!
As stated, my hipotesis is that I need better horn tweeter or super tweeter to do a better HF job. I think maybe the tweter also will allow me to not use Constant Directivity EQ. but the main objective is improve soudn quality. I am also speculating I would be able to use a first order crossover (single cap) to blend tweeter and Mid horn.
I have some questions in my mind, maybe some of you guys already know som answers:
1) Not sure if a small horn with compresion driver will perform better or worse than a jbl2405 in my system. What should be de advantages / disadvantages of each approach?
2) If I go for a small horn for the highs. I am assuming 1 inch Compresion Driver is ideal. A few friends said they got great results with the Selenium 220ti that is available here so it would b e easy and cheaper to choose that driver.
Considering I have a biradial 90x60 2360A horn in the mids.
What should be the best profile for the tweeter horn. Tractrix, Lecleach, other?
I guess it would be important to match front wave patterns? dont have real experience with this.
3) I started modeling in CAD a small tractrix with a cutoff frequency of 1500hz and 1 inch. My thinking is that 1500hz is low enough to implement a crosover arround 5-6khz with a single capacitor. that sounds good to me but Am I missing something? what about beaming? wondering if a jbl 2405 would behave much better?
4) this is my first model:
Is the sharp angle at the mouth of the horn going to cause difraction problems with sound quality?
should i be simulating directivity vs frequency with a software?
Thank you !!
Best,
Carlos
As stated, my hipotesis is that I need better horn tweeter or super tweeter to do a better HF job. I think maybe the tweter also will allow me to not use Constant Directivity EQ. but the main objective is improve soudn quality. I am also speculating I would be able to use a first order crossover (single cap) to blend tweeter and Mid horn.
I have some questions in my mind, maybe some of you guys already know som answers:
1) Not sure if a small horn with compresion driver will perform better or worse than a jbl2405 in my system. What should be de advantages / disadvantages of each approach?
2) If I go for a small horn for the highs. I am assuming 1 inch Compresion Driver is ideal. A few friends said they got great results with the Selenium 220ti that is available here so it would b e easy and cheaper to choose that driver.
Considering I have a biradial 90x60 2360A horn in the mids.
What should be the best profile for the tweeter horn. Tractrix, Lecleach, other?
I guess it would be important to match front wave patterns? dont have real experience with this.
3) I started modeling in CAD a small tractrix with a cutoff frequency of 1500hz and 1 inch. My thinking is that 1500hz is low enough to implement a crosover arround 5-6khz with a single capacitor. that sounds good to me but Am I missing something? what about beaming? wondering if a jbl 2405 would behave much better?
4) this is my first model:
An externally hosted image should be here but it was not working when we last tested it.
Is the sharp angle at the mouth of the horn going to cause difraction problems with sound quality?
should i be simulating directivity vs frequency with a software?
Thank you !!
Best,
Carlos
Last edited:
Hey Carlos,
I think that horn looks like a fun project. I can't answer all your technical question though. But I do have one piece of advice. If you use a 1 inch driver for highs make sure its a aluminum diaphragm. At those frequency your right where titanium breaks up. You might not have a problem with it, or you might not. But titanium can be a bit harsh up there. Depends on the driver.
How were you planning to make your horns?
Nick
I think that horn looks like a fun project. I can't answer all your technical question though. But I do have one piece of advice. If you use a 1 inch driver for highs make sure its a aluminum diaphragm. At those frequency your right where titanium breaks up. You might not have a problem with it, or you might not. But titanium can be a bit harsh up there. Depends on the driver.
How were you planning to make your horns?
Nick
About building the horn I am very excited about the idea of using a 3D printer. I called a company here, and it should be reasonably expensive.
First I am testing the mentioned Selenium 220ti from a friend. he made his tractrix from modelling wood and papier mache 🙂 very clever guy:
He published his adventure at our local Hifi forum at chile:
Poor man tractrix - HiFi Chile
I am waiting to listen to them, he said he was very happy with them, powered with a 71a SET.
First I am testing the mentioned Selenium 220ti from a friend. he made his tractrix from modelling wood and papier mache 🙂 very clever guy:
An externally hosted image should be here but it was not working when we last tested it.
He published his adventure at our local Hifi forum at chile:
Poor man tractrix - HiFi Chile
I am waiting to listen to them, he said he was very happy with them, powered with a 71a SET.
Last edited:
My observation is you are doing the opposite what I do. I will give you suggestions but you must decide your own directions of course
First the bass, the bass is the foundation and is where you need efficiency most. Using a single 18 with no back panel is nearly the the least effective way to make bass. I would use a bass horn and get efficiency from that.
The midrange horn you are using is mainly designed for very large rooms and large distances for many people. In the home we have short distances and few people. The JBL horn (I had the 2360 very shortly) has severe problems with diffraction coloration and requires a lot of EQ. EQ takes away efficiency or requires power. You are driving it with a 1.5 watt amplifier. I would use a different midrange horn. A horn that works better in a small room. A tractrix horn may be the answer with a 300-500 cycle low frequency limit. Then as the frequency goes off hand off to the smaller horn you proposed above. I like the Celestion 1425 for treble over the Selenium it sounds much smoother and natural.- use the little amp to drive the tweeter, use a bigger amp to drive the midrange and triamplify with the biggest amplifier on the new bass horns.
First the bass, the bass is the foundation and is where you need efficiency most. Using a single 18 with no back panel is nearly the the least effective way to make bass. I would use a bass horn and get efficiency from that.
The midrange horn you are using is mainly designed for very large rooms and large distances for many people. In the home we have short distances and few people. The JBL horn (I had the 2360 very shortly) has severe problems with diffraction coloration and requires a lot of EQ. EQ takes away efficiency or requires power. You are driving it with a 1.5 watt amplifier. I would use a different midrange horn. A horn that works better in a small room. A tractrix horn may be the answer with a 300-500 cycle low frequency limit. Then as the frequency goes off hand off to the smaller horn you proposed above. I like the Celestion 1425 for treble over the Selenium it sounds much smoother and natural.- use the little amp to drive the tweeter, use a bigger amp to drive the midrange and triamplify with the biggest amplifier on the new bass horns.
thanks Pooh, I agree with almost everything you said. Main reason for this "reverse" sequence is that I want to go relax with audio budget for a few months but still want to have fun. Bass is yet outstanding but keep thinking about it. I am also aware the 2360a could be not ideal, I think I would be able to exchange my horns by a pair of Altec 811, they sounded much more musical to me, but not sure yet, what do you think?.
Those tweeters maybe still be very useful in any horn system i guess...
only thing I didn't agree is removing the 45 SET from the mids 🙂 , I am too attached to that amplifier and want to play music with it, at least mids and highs for a while... I know people have a different opinion about amplifiers, but for me not all the amplifiers sound the same, and not all tubes sound the same. so it's not so easy for me to switch amplifiers so easily when you feel you have a piece that is sounding outstanding.
Best,
Those tweeters maybe still be very useful in any horn system i guess...
only thing I didn't agree is removing the 45 SET from the mids 🙂 , I am too attached to that amplifier and want to play music with it, at least mids and highs for a while... I know people have a different opinion about amplifiers, but for me not all the amplifiers sound the same, and not all tubes sound the same. so it's not so easy for me to switch amplifiers so easily when you feel you have a piece that is sounding outstanding.
Best,
I have 45 amps that sound good on mid horns too. My point should have read not on mid horns where you need to eq because it may tale away 12 db of efficiency or more and with 45 amps 12 db or more means a lot. sometimes i put them in my system (as well as 300b or 2a3) but now i use a 6BQ5/EL84 amp not because I eq my mid but because I use a lower efficiency (around 103 db/watt/meter) horn with a cone driver and the PP amp sounds just perfect to me driving them
You could alway throw you 18 inch driver in a back loaded horn. Then you could have some real fun, lol.
Hey Carlos, I have a JBL 2380a bi-radial clone for my mids with a Selenium D405 driver. Personally, I like the combo, sounds smooth and I measured the limits of the driver from 300 Hz to 7 kHz. In my system, I XO at 500 Hz (to a 15" Bob Crites woofer in a Cornwall cab) and 5 kHz.
With respect to the top, I did have a Selenium D220Ti, while a good driver for the price, it always sounded a bit harsh to my ears. I switched to a BMS 4540 and it is a very detailed sounding compression driver, almost ribbon tweeter detailed, so I am quite happy. Then again, it was 3x the cost of the D220Ti...
With respect to horns, call me crazy, but the biradials do a nice job. I am happy with the 2380a. It has a 90x40 pattern. For the top, I was using a Selenium HM11 25 60x60 exponential for the BMS 4540. Great on axis fr, but the horn sounds a bit beamy to me compared to the biradial (I have a 3 way active system so I can just listen to single or groups of drivers).
For the top, I am switching to a QSC waveguide with a 90x60 pattern for 5 kHz and above as it exhibits excellent constant directivity - see post 70. I feel it matters a great deal for the mid and top horn to be not only of a similar horn type (i.e. both have constant directivity) and but also similar pattern control.
If the QSC is not available, the Dayton and Pyle also measure similar results.
Best of luck with your project.
Regards, Mitch
With respect to the top, I did have a Selenium D220Ti, while a good driver for the price, it always sounded a bit harsh to my ears. I switched to a BMS 4540 and it is a very detailed sounding compression driver, almost ribbon tweeter detailed, so I am quite happy. Then again, it was 3x the cost of the D220Ti...
With respect to horns, call me crazy, but the biradials do a nice job. I am happy with the 2380a. It has a 90x40 pattern. For the top, I was using a Selenium HM11 25 60x60 exponential for the BMS 4540. Great on axis fr, but the horn sounds a bit beamy to me compared to the biradial (I have a 3 way active system so I can just listen to single or groups of drivers).
For the top, I am switching to a QSC waveguide with a 90x60 pattern for 5 kHz and above as it exhibits excellent constant directivity - see post 70. I feel it matters a great deal for the mid and top horn to be not only of a similar horn type (i.e. both have constant directivity) and but also similar pattern control.
If the QSC is not available, the Dayton and Pyle also measure similar results.
Best of luck with your project.
Regards, Mitch
I tried the small blue horn this weeked with the Selenium 220TI. Did't like it. Had the "harsh" feeling you guys described. So I guess I will keep the 2360a BMS4590 combo for a while. I felt it so harsh that I didn't feel like trying with other XO points or doing additional measurements 🙁 I think it was in a lower category compared with the highs I get with the BMS4590 2360a combo. So that research line was suspended.
On the other hand, I have been hitting my head against the wall a lot about measuring flat response and having a sound that I really dislike.
Since I measured flat, I thought the horn was the harsh and that I needed a smaller tweeter.
Well, I looked a lot in the internet and it's very interesting, lots of people mentions that (including this forum) that flat response up to 20khz it not supoused to be pleasant.
And that a soft roll-off from 400hz to 10khz of -6db tend to give great results. This may vary depending on the horn type.
This talks with what I have been experimenting, so I felt more comfortable with not having a flat response at the high end, and giving a more soft CD compensation. of arround 5db in the 5khz - 10khz zone.
So now I am just looking for a pleasant subtle CD EQ arround this gentle roll-off, but not anymore with a flat response in mind
Since I measured flat, I thought the horn was the harsh and that I needed a smaller tweeter.
Well, I looked a lot in the internet and it's very interesting, lots of people mentions that (including this forum) that flat response up to 20khz it not supoused to be pleasant.
And that a soft roll-off from 400hz to 10khz of -6db tend to give great results. This may vary depending on the horn type.
This talks with what I have been experimenting, so I felt more comfortable with not having a flat response at the high end, and giving a more soft CD compensation. of arround 5db in the 5khz - 10khz zone.
So now I am just looking for a pleasant subtle CD EQ arround this gentle roll-off, but not anymore with a flat response in mind
Last edited:
So this should leave the HF area as it is for a while.
I want to focus now in the low-mid frequencies.
With 2360a and BMS 4590 I think I can go to 500hz very easily.
I also can have a XO at 300hz, but maybe frequency response won't be ideal.
If later I choose (as suggested) to use a smaller horn, I guess a typical XO fecuency should be 800hz.
So in general terms I think I should have a Bass enclosure that can be croosed over bettwen 400hz-800hz range.
I also would like to have bass down to 40hz.
And most important, a sound that sounds big, dinamic but very clean to match the compression driver.
No idea what driver/enclosure to do this.
Also, I am using a separate amplifier for the woofer.
I want to focus now in the low-mid frequencies.
With 2360a and BMS 4590 I think I can go to 500hz very easily.
I also can have a XO at 300hz, but maybe frequency response won't be ideal.
If later I choose (as suggested) to use a smaller horn, I guess a typical XO fecuency should be 800hz.
So in general terms I think I should have a Bass enclosure that can be croosed over bettwen 400hz-800hz range.
I also would like to have bass down to 40hz.
And most important, a sound that sounds big, dinamic but very clean to match the compression driver.
No idea what driver/enclosure to do this.
Also, I am using a separate amplifier for the woofer.
Last edited:
Right now I am using a Eminence Omega Pro 18a (18 inches) in a Open Baffle U-frame.
Not sure I am I am measuring it well. Don't know how to properly measure bass response, so what I am doing is to measure it is to put the microphone in the listening spot. and getting a measurement of full response, including the room interaction with a 1/3oct smoothing.
The woofer response is very weak at the 150hz-400hz region, and very strong in lower frequencies.
What gave me better results is to do serius EQ to cut down the lower frecuencies to try to get a flat response.
Not sure I am I am measuring it well. Don't know how to properly measure bass response, so what I am doing is to measure it is to put the microphone in the listening spot. and getting a measurement of full response, including the room interaction with a 1/3oct smoothing.
The woofer response is very weak at the 150hz-400hz region, and very strong in lower frequencies.
What gave me better results is to do serius EQ to cut down the lower frecuencies to try to get a flat response.
In general, I feel drums don't sound very present, but if I try to increase the Woofer SPL (relative to the horn SPL), it doesn't sound very clean and voices get ruined, so I am sort of trapped in this configuration.
I feel like trying a 15in sealed enclosure.
But first I would like to know if I am doing something wrong with the current setup, or I just have to move to a 15in woofer for better results.
Best,
Carlos
I feel like trying a 15in sealed enclosure.
But first I would like to know if I am doing something wrong with the current setup, or I just have to move to a 15in woofer for better results.
Best,
Carlos
Last edited:
Or perhaps, just my current Eminence Omega 18a will give much better results in a sealed box with proper EQ. maybe cleaner sound?
I am confused, and very sure I don't have the knowledge/experience required.
Maybe the cheaper is to build a pair of sealed boxes for my current woffers and do EQ. and see what the difference is!
I am confused, and very sure I don't have the knowledge/experience required.
Maybe the cheaper is to build a pair of sealed boxes for my current woffers and do EQ. and see what the difference is!
Yes, perhaps not having the sound of the reversed polarity rear midrange open baffle U-frame's output bouncing off the back wall and combining in an interference pattern requiring drastic EQ to be reasonably flat in the listening area may be preferable 😉.