The large PT guides have been fan favorites for some time. Over at avs forum there's a 223 page thread on the 4722 (their threads do go VERY long, and I hate the forum software with ads everywhere).
The JBL 2453h-sl seems to be an excellent driver choice for them- cheaper than most other JBL parts by a long shot, with a coated/embossed titanium diaphragm. I've previously used coating on an embossed diaphragm in the selenium D2500Ti, to great effect. Only thing I'd want to add would be allowing some significant rolloff and using a supertweeter rather than trying to get top octave content from a 4" phragm.
The JBL 2453h-sl seems to be an excellent driver choice for them- cheaper than most other JBL parts by a long shot, with a coated/embossed titanium diaphragm. I've previously used coating on an embossed diaphragm in the selenium D2500Ti, to great effect. Only thing I'd want to add would be allowing some significant rolloff and using a supertweeter rather than trying to get top octave content from a 4" phragm.
The 'PT' guides also resemble some old EV horns, which will be even further out of copyright (and JBL aren't shy about the similarity - they reference EV horns in their tech notes).
I suppose when a group of manufacturers all:
-stop using diffraction slots / sharp angles
-use horns with largely flat walls
...the look, profile and sound of their horns is going to converge.
Attachments
Last edited:
+1 on the JBL 2453H-SL - excellent driver for the money. Replaced the 2432H on the 2384's with these last weekend. Measure and sound really nice - lower distortion, smoother response.
I had a 3-way with a BMS 4540ND on the QSC 152i waveguide, that measured well past 20 kHz. The 2453H-SL last 1/2 octave out to 20 kHz isn't too pretty, but sounds good to my ears. I use DSP to straighten it out a bit. I am too old to hear that last 1/2 top octave anyway
I had a 3-way with a BMS 4540ND on the QSC 152i waveguide, that measured well past 20 kHz. The 2453H-SL last 1/2 octave out to 20 kHz isn't too pretty, but sounds good to my ears. I use DSP to straighten it out a bit. I am too old to hear that last 1/2 top octave anyway
Not really. He let it expire (stopped paying maintenance fees). Geddes wasn't exactly getting rich from selling speakers - he made his living as a consultant.LOL, that explains the gedlee crankiness
Those are the Electro Voice CE horns - designed by Keele. I have the big ones. Th HR9040 and HR6040. The HR6040 is my favorite of the two. CE stands for conic exponential. The 9040's are around 40" wide at the mouth and 6040 is around 32" wide. They both are tall so the vertical radiation is somewhat controlled compared to most horns.
Last edited:
Hello Jorg. Sorry for the late reply. I was taking vacations in the middle of nowhere, without internet access.
Yes, the CMCDs look interesting. In fact those were the ones that got me started down this path. Pooh tried several cone driven horns including the CMCD 61H and 81H (I now seem to recall he tried both). My recollection is the larger one was good up to about 2kHz and the smaller up to about 4kHz. And I've been experimenting with a Faital M5N12 which he thought worked as good.
But then he also said the JBL compression driver and 2384 sounded a lot better than the CMCD from 650 to 2000Hz...so the bug bit me
BTW, thanks for the pointer to the Seos 24-4". I had seen it before, but was put off by their note of minimum recommended xo at 1200Hz, which is way too high for me. But then again they also say 1500Hz for the 24" and people have successfully implemented it lower than that. I wish they added some frequency response graphs for reference.
Hey Scott.
These look like one of my prototypes, but for 2" CD instead of the 4" throat I used on my 5" cones. Certainly an option, although for something like this I would build them.
I may have shared this info with you before. If so, sorry for the duplication, but if not, I hope you will find this interesting, and of use:
Volvotreter Homepage
http://volvotreter.de/downloads/Edgar-Midrange-Horn.pdf
The Azura horn website is good for this - he has measured 300-950mm mouth horns. You can use his plots to make guesstimates of what cutoff you can get from a given horn size.
NOTE: most of the Azura charts have mic at horn mouth, so their extension is slightly exaggerated. The 425 is the only one with a 1m response.
Building a 2" throat is a fairly delicate job. Even the nice Volti horns have a few (tiny) chips and wiggles in the throat of the horn they used for the photoshoots. Unless you are a patient craftsman (I'm not!) with access to great equipment (as good or better than Volti) I heartily recommend starting with a cheap generic 2" horn and just extend the mouth to whatever size you need. Minor imperfections in the mouth matter a lot less.
(Versus buying a Volti horn) you might only save $800 rather than $850 by doing it 'my' way, but I think the saving in effort is worth it.
Hey Scott. Yeap, I've been all over that website. My tractrix prototype was based in good part on Edgar's article.
Exactly. That is the website I had in mind when I posted my reply.
Would the extrapolation you suggest be worth counting on, though? Isn't mouth size just one [key] factor? Horn length, profile, throat and chamber size being others.
Is this a delicate job if I'm looking at frequencies up to 2kHz? I was under the impresion the frequencies I'm looking into aren't sooo picky. Maybe I'm wrong.
I only have experience with the CMCD 8" driver, I never managed to find the 61H phase plug (not available new anymore, the 81H was/is).
The 2384 is a great horn - probably the best horn for use all the way up. BUT it is a little bit too small. Yes, it's about 30 inches wide, but the effective mouth is probably not larger than about 24 inches. Unfortunatly this horn is a little bit small and does not load very good below 700-800Hz. But if you can cross over at 800Hz, it's exceptional! Thought this is often too high for people that use midbass horns below. The same for the Seos 24, a little bit small to load a large format compression driver crossed over at 600Hz. The Seos 30 is perfect in size, but unfortunatly it's constructed wrong for 1,5" drivers, so you can't use it all the way up. The problem is that they have constructed it for 2" exit and just built in an longer horn start to 1,5" exit. That makes the horn start not suited to play over 10k KHz.
The crossover frequencies for Seos 24 and Seos 24-4 on Auto-Tech's web sites is totally wrong and thats obvious. You know something is wrong when the Seos 12 says 1200Hz and the Seos 24 says 1500Hz - the 24 is twice as wide and can then be used twice as low - about 600Hz. But as said earlier, it does not load 100 % that low. Just look at polar plots and you will understand. Same for the Seos 12 at 1200Hz, you should probably cross it a bit higher - 1500Hz+.
Large format compression driver in 2384 sounds better than CMCD from 650 to 2000Hz? Well, I would say the opposite. The 450-2000 range is just where the CMCD is best. If it was not for the fact that I have big loobing problems because I am sitting so close the CMCD would still be here. Now I am looking for a horn to use all the way from 600Hz and up. The 2384 is for now the best we have tried, but not large enough. Next step is a new horn constructed by a small team from Norway, soon a prototype is finished. Hopefully that will be the solution, if not the next step is to 3D scan the 2384 and upscale it big enough
For you, if you want the best option from 450-2000Hz - stop looking and buy the Seos 24-4+CMCD-81H. A friend use the same driver/phase plug in a custom Tractrix 180 Wood horn because he rather wants the round horn profiles of esthetic reasons. Works really well that too!
The 2384 is a great horn - probably the best horn for use all the way up. BUT it is a little bit too small. Yes, it's about 30 inches wide, but the effective mouth is probably not larger than about 24 inches. Unfortunatly this horn is a little bit small and does not load very good below 700-800Hz. But if you can cross over at 800Hz, it's exceptional! Thought this is often too high for people that use midbass horns below. The same for the Seos 24, a little bit small to load a large format compression driver crossed over at 600Hz. The Seos 30 is perfect in size, but unfortunatly it's constructed wrong for 1,5" drivers, so you can't use it all the way up. The problem is that they have constructed it for 2" exit and just built in an longer horn start to 1,5" exit. That makes the horn start not suited to play over 10k KHz.
The crossover frequencies for Seos 24 and Seos 24-4 on Auto-Tech's web sites is totally wrong and thats obvious. You know something is wrong when the Seos 12 says 1200Hz and the Seos 24 says 1500Hz - the 24 is twice as wide and can then be used twice as low - about 600Hz. But as said earlier, it does not load 100 % that low. Just look at polar plots and you will understand. Same for the Seos 12 at 1200Hz, you should probably cross it a bit higher - 1500Hz+.
Large format compression driver in 2384 sounds better than CMCD from 650 to 2000Hz? Well, I would say the opposite. The 450-2000 range is just where the CMCD is best. If it was not for the fact that I have big loobing problems because I am sitting so close the CMCD would still be here. Now I am looking for a horn to use all the way from 600Hz and up. The 2384 is for now the best we have tried, but not large enough. Next step is a new horn constructed by a small team from Norway, soon a prototype is finished. Hopefully that will be the solution, if not the next step is to 3D scan the 2384 and upscale it big enough
For you, if you want the best option from 450-2000Hz - stop looking and buy the Seos 24-4+CMCD-81H. A friend use the same driver/phase plug in a custom Tractrix 180 Wood horn because he rather wants the round horn profiles of esthetic reasons. Works really well that too!
Hello Jorg. Sorry for the late reply. I was taking vacations in the middle of nowhere, without internet access.
Yes, the CMCDs look interesting. In fact those were the ones that got me started down this path. Pooh tried several cone driven horns including the CMCD 61H and 81H (I now seem to recall he tried both). My recollection is the larger one was good up to about 2kHz and the smaller up to about 4kHz. And I've been experimenting with a Faital M5N12 which he thought worked as good.
But then he also said the JBL compression driver and 2384 sounded a lot better than the CMCD from 650 to 2000Hz...so the bug bit me
BTW, thanks for the pointer to the Seos 24-4". I had seen it before, but was put off by their note of minimum recommended xo at 1200Hz, which is way too high for me. But then again they also say 1500Hz for the 24" and people have successfully implemented it lower than that. I wish they added some frequency response graphs for reference.
JBL specs the 2384 waveguide to crossover at 630Hz (active) for the 2-way cinema JBL 4722: https://www.jblpro.com/ProductAttachments/SS4722_0509.pdf
I use digital XO crossing at 630 Hz and don't hear or measure any real issues... Maybe the vertically stacked double 15"s assist... Effective mouth area is around 26"
I would love to try CMCD 8" driver and maybe some day I will get a chance. But pretty happy with the 2384, especially using the upgraded JBL 2453H-SL aquaplased driver - it measures quite smooth, and has very low distortion from 600Hz on up - much less measured distortion than the stock 2432H. The 2453H-SL is one of the few JBL neo drivers that doesn't cost an arm and leg, at least in the US. They go for $250 each.
I use digital XO crossing at 630 Hz and don't hear or measure any real issues... Maybe the vertically stacked double 15"s assist... Effective mouth area is around 26"
I would love to try CMCD 8" driver and maybe some day I will get a chance. But pretty happy with the 2384, especially using the upgraded JBL 2453H-SL aquaplased driver - it measures quite smooth, and has very low distortion from 600Hz on up - much less measured distortion than the stock 2432H. The 2453H-SL is one of the few JBL neo drivers that doesn't cost an arm and leg, at least in the US. They go for $250 each.
Well, all speaker systems have compromises and the JBL engineers probably found out that the 630Hz crossover frequency was a good compromise in that application. Personally I am trying to minimize the compromises and a 600Hz crossover frequency with the 2384 is not ideal. I really wish it was the solution so I could stop looking for the "perfect horn", but unfortunately not.
If you had compared your 2384 to for example the larger 2360/Seos 30 horns crossed over at about 600Hz I can almost assure you that you would agree that these horns have more "growl" in the 600-800 area. But again, no one of these two can be used all the way up.
Measure your 2384 horns 0-15-30-45-60-75-90 degrees and you will probably understand without trying other horns.
I have never tried the 2453H-SL but I am sure they are better than f.ex. the Radian drivers. Personally I use the 2450SL with Truextent Beryllium diaphragms. I also have a pair of TAD ET-703, but I hope that I don't need to use them if I can find the right horn to play all the way from 600 and up. The 2384 have constant directivity up to about 17KHz and really don't need any tweeter.
If you had compared your 2384 to for example the larger 2360/Seos 30 horns crossed over at about 600Hz I can almost assure you that you would agree that these horns have more "growl" in the 600-800 area. But again, no one of these two can be used all the way up.
Measure your 2384 horns 0-15-30-45-60-75-90 degrees and you will probably understand without trying other horns.
I have never tried the 2453H-SL but I am sure they are better than f.ex. the Radian drivers. Personally I use the 2450SL with Truextent Beryllium diaphragms. I also have a pair of TAD ET-703, but I hope that I don't need to use them if I can find the right horn to play all the way from 600 and up. The 2384 have constant directivity up to about 17KHz and really don't need any tweeter.
Oh, good.
Yes indeed. Very loosely: length is what sets LF limit and mouth size supports this / sets the amount of ripple.
Almost all modern horns are pretty short*, in the same ballpark as the Azura horns. Between looking at the Azura plots, and trying a sim in Hornresp, you should be able to get a pretty good idea. I like the plots as a (literal) reality check.
Good point. I think you're right. If you use a grill cloth and a low pass filter, you can probably be quite sloppy
I don't have the tools for compound mitre joints, but I do have an eye for found objects. A humble wooden fruit bowl from Ikea makes a good conical horn for small cone drivers.
http://www.diyaudio.com/forums/full-range/283083-horn-fast-timber.html
A small cone driver + horn like this is not as efficient as a big compression driver - but is still enough to keep up with a direct radiator woofer. It's also a cheap option, and good for < 500Hz. The main demerit is the bigger mid-to-tweeter spacing (than an equivalent rectangular horn) ...and there's the ongoing subjective debate about relative levels of 'detail' and 'impact' of cones vs. CDs.
Many good options < $100
1) Tang Band W5-2143 (fairly robust, 2.5mm xmax)
2) PRV Audio 5MR450NDY (very robust, 2.0mm xmax)
3) P Audio SN4-60F (I've used it; not bad)
4) Scan-Speak 15M/4624G00 5.5 (looks good; I have some on order)
*The Inlow Paper horn is one exception, it is ~triple the length of an average PA horn ...but he also uses old drivers, which have a much lower flare rate than modern units.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.