Well, that certainly looks promising! IslandPink, distances in the UK aren't that great, maybe you can visit Jonjin and draw your own conclusions about the system. I'd think a PP triode amplifier would be just the ticket for driving an Altec-sourced system - WE movie theater design philosophy and all that.
Jonjin, thanks for the pix and encouragement. It was John Atwood who told me that the Tangerine version of the 288 had pretty decent treble to 12~15 kHz, which is kind of astounding for a 2.88" aluminum diaphragm. I've only heard the older versions, which had really good mid/lower treble on a 1505 multicell, with soft/rolled-off treble above 10 kHz.
Three are three differences between an old-school circumferential-slot 288 & 1505 and the new system:
1) Tangerine instead of circumferential phase plug - according to Altec data, this is good for another 2~3 kHz at the top before the phase response becomes disrupted. Subjectively, both phase plug designs have their partisans and detractors. Atwood prefers the Tangerine.
2) LeCleac'h low-diffraction horn instead of the multicell. There's an awkward transition in the multicell between the circular exit of the compression driver and the array of square-section entrances of the multiple sub-horns. This acoustically complex transition inevitably creates reflections and diffraction with resulting ripples in the time-domain and frequency response. The really big version used in the A2 speaker that had two 288's driving a common 1505 must have been even worse. (In fairness, the Altec A-series theater speakers were designed in the mid-Forties, when all movie soundtracks were mono optical, and rolled off at 8 kHz.)
3) The AH425 Azurahorn has the 8-degree internal flare of the 288 as an integral part of the design equation, unlike traditional Tractrix or standard-formula conical horns. Bjorn Kolbrek's FEM analysis shows that a detail as small as omitting this compensation can have a significant effect on the smoothness of the response in the 1~5 kHz region.
The net effect of all this? I don't know, but the impulse response should be better. As for the Altec bass, the sound I heard at the last RMAF from the OMA speaker sold me on Altec/GPA drivers - honestly, the best bass and midbass I'd heard in many years.
Jonjin, thanks for the pix and encouragement. It was John Atwood who told me that the Tangerine version of the 288 had pretty decent treble to 12~15 kHz, which is kind of astounding for a 2.88" aluminum diaphragm. I've only heard the older versions, which had really good mid/lower treble on a 1505 multicell, with soft/rolled-off treble above 10 kHz.
Three are three differences between an old-school circumferential-slot 288 & 1505 and the new system:
1) Tangerine instead of circumferential phase plug - according to Altec data, this is good for another 2~3 kHz at the top before the phase response becomes disrupted. Subjectively, both phase plug designs have their partisans and detractors. Atwood prefers the Tangerine.
2) LeCleac'h low-diffraction horn instead of the multicell. There's an awkward transition in the multicell between the circular exit of the compression driver and the array of square-section entrances of the multiple sub-horns. This acoustically complex transition inevitably creates reflections and diffraction with resulting ripples in the time-domain and frequency response. The really big version used in the A2 speaker that had two 288's driving a common 1505 must have been even worse. (In fairness, the Altec A-series theater speakers were designed in the mid-Forties, when all movie soundtracks were mono optical, and rolled off at 8 kHz.)
3) The AH425 Azurahorn has the 8-degree internal flare of the 288 as an integral part of the design equation, unlike traditional Tractrix or standard-formula conical horns. Bjorn Kolbrek's FEM analysis shows that a detail as small as omitting this compensation can have a significant effect on the smoothness of the response in the 1~5 kHz region.
The net effect of all this? I don't know, but the impulse response should be better. As for the Altec bass, the sound I heard at the last RMAF from the OMA speaker sold me on Altec/GPA drivers - honestly, the best bass and midbass I'd heard in many years.
i'm using my Onkens with a TAD 2002 driver, this also work pretty well ..
An externally hosted image should be here but it was not working when we last tested it.
Okay... been listening to the Onkens all weekend and admittedly still surprised by how good it is (if I may blow my own trumpet). The combination drivers in the horns/enclosues really worked well. I really fluked this build.
The Onken mid compression driver is simply superb. I would describe is as having a crystal clarity but with taste. For example drinking the finest spring water. Another example is comparing a Bosendorf to a Steinway - the Bosendorf has a warm character to it; the Steinway has a much clearer sound but still has character. There's lots of detail coming through the system... and everything has this air and ease of portrayal. Listening to classical (or any other music for that matter) is a joy and it is so easy to follow the flow and ebb of the music.
The Altec 414 works well in the system. As others have said, it's strength actually is it's mid. The bass doesn't go that low but down to 45ish in the Onken enclosure which is plenty for me. What's there is taut and speedy... and has tone, not some indistinct bass note. Actually using a First Watt F3 on the bass. I've been recommended a full valve amp as it seems to prefer higher amp impedences which will give it a fuller bass.
Been using the system with a Behringer DCX2496. Can't make any comments about sound degradation. At the moment it's really sublime. Cross-over points are:
Bass: 6dB Butterworth 1000Hz
Mid: 12dB Butterworth 1000Hz
Treble: 6dB Butterworth around 10kHz
Bass is powered with the First Watt F3. Mid and treble is powered by a 2W Decware Zen Triode. The plan is to emulate the Behringer with a passive but that is another story...
JJ
The Onken mid compression driver is simply superb. I would describe is as having a crystal clarity but with taste. For example drinking the finest spring water. Another example is comparing a Bosendorf to a Steinway - the Bosendorf has a warm character to it; the Steinway has a much clearer sound but still has character. There's lots of detail coming through the system... and everything has this air and ease of portrayal. Listening to classical (or any other music for that matter) is a joy and it is so easy to follow the flow and ebb of the music.
The Altec 414 works well in the system. As others have said, it's strength actually is it's mid. The bass doesn't go that low but down to 45ish in the Onken enclosure which is plenty for me. What's there is taut and speedy... and has tone, not some indistinct bass note. Actually using a First Watt F3 on the bass. I've been recommended a full valve amp as it seems to prefer higher amp impedences which will give it a fuller bass.
Been using the system with a Behringer DCX2496. Can't make any comments about sound degradation. At the moment it's really sublime. Cross-over points are:
Bass: 6dB Butterworth 1000Hz
Mid: 12dB Butterworth 1000Hz
Treble: 6dB Butterworth around 10kHz
Bass is powered with the First Watt F3. Mid and treble is powered by a 2W Decware Zen Triode. The plan is to emulate the Behringer with a passive but that is another story...
JJ
Nice work there fine fellow .
I may just get down your way before I go too far with this project, but it's moving along here. I think the horns will be posted to me next week, and hopefully the 288's not too far after.
I'm going to do some sims with various BR options . Not sure just what the Onken does that other BR would not do . I'm wondering if two side vents near the floor would be better .
Good news from you and JAtwood that it can sound pretty good as a two-way . Your comments about 45Hz sounding fine are encouraging .
I've got an option to buy a pair of Supravox 285GMF ( which I have here loaned ) so will be using those first . They have a very nice 20g moving mass .
Lynn, would you favour an alternative BR option, perhaps with a narrower rounded-off front ?
Mark
I may just get down your way before I go too far with this project, but it's moving along here. I think the horns will be posted to me next week, and hopefully the 288's not too far after.
I'm going to do some sims with various BR options . Not sure just what the Onken does that other BR would not do . I'm wondering if two side vents near the floor would be better .
Good news from you and JAtwood that it can sound pretty good as a two-way . Your comments about 45Hz sounding fine are encouraging .
I've got an option to buy a pair of Supravox 285GMF ( which I have here loaned ) so will be using those first . They have a very nice 20g moving mass .
Lynn, would you favour an alternative BR option, perhaps with a narrower rounded-off front ?
Mark
IslandPink said:
I'm doubtful the Onken alignment is all that special. What makes it novel is the clever combination of double-wall construction acting as vents. Spaced double walls are very lossy compared to single-wall construction, and the narrowband resonances radiated by the cabinet-walls tend to be strongly attenuated by the low-pass characteristics of the vent.
It would be interesting to combine the 45-degree slant edges of the HF Olson RCA LC-1A speaker with the double-wall Onken vents. This would substantially lower diffraction and retain the double-wall lossy vents of the Onken. Another clever variation would be to have different thicknesses of cabinet wall for the inner and outer surface, which would stagger the wall resonances, and make acoustic coupling from one to another less efficient.
My experience is that vents that are very close to the floor (2" or closer) couple in some mysterious way with the room, giving much deeper and more powerful bass, while vents that are 18" or higher end up sounding very bass-thin. Vent locations matter, although this doesn't show up in the simulations.
One of the possible merits of the odd-looking tall and narrow vents of the Onken is the room coupling is more diffuse than a small point-source circular vent at an arbitrary height from the floor. The sensitivity to room location for a point-source vent might be one of the reasons that stand-mounted compact speakers are so fiddly and hard to set up, with the location for best bass not being the same as the location for best imaging.
As for the bugaboo of Baffle Step Correction, I think people spend way too much time on this. The standard model that calls out for 3 dB of BSC assumes free-space radiation, literally a cabinet suspended in mid-air, at least 20~50 feet from any other surface. This is an absurd assumption that is nothing like a typical listening environment.
Real speakers in real rooms always have floor reflections (requiring a mirror image in the model), and back and side wall reflections that aren't that much further away. These have a collective effect of making the radiation angle considerably less than a hemisphere, and a rather lumpy transition region from the half-space seen by the tweeter to the complex environment seen at 20~50 Hz.
(Sorry if this has already been mentioned before.)
According to Jacques Mahul (Mr JMLab & Focal) who wrote about it in the french magazine l'Audiophile in the early eighties, the original Onken cabinets are classic ported boxes. Both the regular Onken with Altec 416 and the small Onken with Altec 414.
Doing some little reverse engineering he concluded that the cabinet is a fourth order BR with a Qt value of 0.254 and a alignment value of 6.35, very close to the maximally flat value of 5.7.
He also mentioned that Mr Koïzumi used his own calculus method to get these results. He couln't knew about T&S method which was published only later.
According to Jacques Mahul (Mr JMLab & Focal) who wrote about it in the french magazine l'Audiophile in the early eighties, the original Onken cabinets are classic ported boxes. Both the regular Onken with Altec 416 and the small Onken with Altec 414.
Doing some little reverse engineering he concluded that the cabinet is a fourth order BR with a Qt value of 0.254 and a alignment value of 6.35, very close to the maximally flat value of 5.7.
He also mentioned that Mr Koïzumi used his own calculus method to get these results. He couln't knew about T&S method which was published only later.
Basically true but the very large surface area of the vents results in very low port velocities compared to most conventional BR type speaker designs. (No chuffing or port noise with these babies.)
The basic box configuration of course was also a bit stiffer than the typical large box design of the early 1950s when these were originally appeared as the Jensen Ultraflex and this must have contributed positively to the sound.
The relatively low cone mass of the types of drivers used in these designs along with typical tuning results in good perceived transient response and speed on the bottom end which seems to help in integrating them with horn mids and highs. Mine cross at 800Hz.
Port induced comb filtering can be a problem and there is a surprising amount of mid range energy exiting those ports in close in microphone measurements, but it does not seem to be audible - or perhaps is an artifact of the pick up pattern of my instrumentation mic.
One change I do want to make is to change the fiberglass wool in mine to real lambs wool, this is supposed to do a better job damping mid range reflections in the cabinet. (Some day)
I've been running mine for about 2yrs now, and am very pleased overall with their performance. They don't really sound like most BR types I have heard either.
I have good usable output down to about 35Hz, and an efficiency that approaches 100dBSpl/W/M.. Drivers are Iconic 165-8G.
The basic box configuration of course was also a bit stiffer than the typical large box design of the early 1950s when these were originally appeared as the Jensen Ultraflex and this must have contributed positively to the sound.
The relatively low cone mass of the types of drivers used in these designs along with typical tuning results in good perceived transient response and speed on the bottom end which seems to help in integrating them with horn mids and highs. Mine cross at 800Hz.
Port induced comb filtering can be a problem and there is a surprising amount of mid range energy exiting those ports in close in microphone measurements, but it does not seem to be audible - or perhaps is an artifact of the pick up pattern of my instrumentation mic.
One change I do want to make is to change the fiberglass wool in mine to real lambs wool, this is supposed to do a better job damping mid range reflections in the cabinet. (Some day)
I've been running mine for about 2yrs now, and am very pleased overall with their performance. They don't really sound like most BR types I have heard either.
I have good usable output down to about 35Hz, and an efficiency that approaches 100dBSpl/W/M.. Drivers are Iconic 165-8G.
Lynn Olson said:
I thought it was 6dB for free space and more or less 3dB as a practical BSC in room. Did I get it wrong?
kevinkr said:
Yep, some of the 1st speakers I made used 100% wool felt for damping. And all the Onken boxes I've heard did too (as well as some others). Great stuff, but as you know- not cheap!
Interesting stuff, thanks . So maybe the long vents give a bit of advantage in reducing room nodes , like the beginnings of a line array ?
Ok, beginners question here - what's this aligment value of 6.34 calculated from ? - I mean in other software for instance Winisd you just have a volume and a port tuning frequency . I imagine the 6.35 comes from some simple equation from the other parameters ?
I'm going to get some better software and check out a few things . It would seem the 285GMF in typical Onken or similar BR would roll off quite a bit below 45Hz from the calcs, but I reckon it won't so much when floor reinforcement etc are taken into account .
MJ
Ok, beginners question here - what's this aligment value of 6.34 calculated from ? - I mean in other software for instance Winisd you just have a volume and a port tuning frequency . I imagine the 6.35 comes from some simple equation from the other parameters ?
I'm going to get some better software and check out a few things . It would seem the 285GMF in typical Onken or similar BR would roll off quite a bit below 45Hz from the calcs, but I reckon it won't so much when floor reinforcement etc are taken into account .
MJ
"It would be interesting to combine the 45-degree slant edges of the HF Olson RCA LC-1A speaker with the double-wall Onken vents. This would substantially lower diffraction and retain the double-wall lossy vents of the Onken. Another clever variation would be to have different thicknesses of cabinet wall for the inner and outer surface, which would stagger the wall resonances, and make acoustic coupling from one to another less efficient."
This being a good start point then - ?
http://www.supravox.fr/kits/jensen.pdf
http://www.supravox.fr/anglais/kits.htm#Jensen
MJ
This being a good start point then - ?
http://www.supravox.fr/kits/jensen.pdf
http://www.supravox.fr/anglais/kits.htm#Jensen
MJ
There's also an encouraging pic halfway down this page, for the 285GMF , looks like it's from Martin King's software . Seems only down about 2dB at 40 :
http://www.diyaudio.com/forums/showthread.php?threadid=78407&perpage=25&highlight=&pagenumber=2
Sorry for banging on about my own project, JJ !
Mark
http://www.diyaudio.com/forums/showthread.php?threadid=78407&perpage=25&highlight=&pagenumber=2
Sorry for banging on about my own project, JJ !
Mark
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.