Btw, most of the Linda Ronstadt harshness I heard with the new speakers in development in the studio/lab was absent in the old Linkwitz Orion in the house. Much larger space, and far more reflective but almost the same electronics. The upper mids of the Orion is a bit recessed, but it has a smooth silky clarity that I suspect comes from the magnesium cone 8" Seas w22ex001. It's still a pretty good speaker system. (Not that any of this is conclusive, comparing unfinished speakers to the Orions.)
Last edited:
Oh, do you have measurements of the response of Orions? The basic problem with them is wide directivity above 1kHz (loss of dipole pattern). This means that easily the room response at spot has too much energy in 1-3kHz range. If not in spl response, but decay response might most llkely show that.
No measurements ATM. It was done years ago with other software, but who knows where they are now after a couple of big moves & computer changes, too. The Orions are in the house, far from measurement tools. Either have to install REW on that PC or haul them out here -- 200' away, so a bit of a PITA, but it might be best. Ah, just remembered I have 8-conductor Speakon connectors with dedicated cables on them... no, it will have to be done in the house.
ps -- did you get my PM?
Try Prisoner in Disguise. Just about any track from her Canciones de Mi Padre album. You can hear it on most of her singing, but it's not always harsh; when it's bad, it's unmistakable. It's not clear whether these tracks were mis-recorded, and thus the harshness (HD/IMD?) embedded in the recordings -- ie, mic/mic preamp overload? -- or whether the playback system can't handle them well, or a combination of the 2. Her voice has complex overtones, not sibilance per se, but multiple harmonics far above the the primary tone. Another voice with such harmonics is Jennifer Warnes, though not as extreme as LR or maybe generally better recorded.
I have measurements made with the clio of my Orions from 2004, only 1 meter frequency response and listening point, no decay.
if you are interested I can publish.
I had problems with medium-high reproduction and a lot of listening effort after 10 minutes.
the sound was too sparkling, too crisp, too nervous.
Even the female voice was bad for example from the CD of the Sheffield Lab, Pop Experience, track 11, Thelma Houston - I've got the music in me, the voice was too nervous, it seemed like a singer who cannot modulate the voice, unheard, I listened to the same song on many other hi-fi systems and the voice was relaxed, it was a beautiful voice.
so I called a professional to my house to take measurements and understand where the problem was.
Here is an interview of Linda from 2019. Parkinson is one hell of a disease!
YouTube
Sound of My Voice trailer YouTube
Linkwitz himself went to 4-way LX521 because Orions couldn't be fixed for the low treble problem LX521 Description
"The LX521 evolved from experimentation with minimal width baffles, which can provide a more uniform dipolar radiation pattern at higher frequencies, if also suitable drivers are available. Inevitably this leads to a 4-way design, which I mostly tried to avoid in the past. The shape of the midrange/tweeter baffle was arrived at empirically for the chosen SEAS drivers and after many acoustic free-field measurements."
YouTube
Sound of My Voice trailer YouTube
Linkwitz himself went to 4-way LX521 because Orions couldn't be fixed for the low treble problem LX521 Description
"The LX521 evolved from experimentation with minimal width baffles, which can provide a more uniform dipolar radiation pattern at higher frequencies, if also suitable drivers are available. Inevitably this leads to a 4-way design, which I mostly tried to avoid in the past. The shape of the midrange/tweeter baffle was arrived at empirically for the chosen SEAS drivers and after many acoustic free-field measurements."
Last edited:
try to separate the upper baffle from the woofer cabinet.
if even small vibrations arrive at the upper baffle you will lose accurate reproduction and detail resolution.
a good piano recording with long notes will emphasize the distortion of the baffle which vibrates and lengthens the tempo of the notes, you will feel loss of articulation loss of the scoring of the piano keys.
I'm not convinced of this.
It is impossible to keep the upper baffle from vibrating in sympathy with the 8" driver mounted on it. The baffle board is not reinforced around its edges so it is not rigid enough to prevent this.
Even my 1.2" thick CLD baffle vibrates with the 8" driver. My tactile perception (with fingers) is that when the top baffle is placed atop the W-frame, vibrations from the bass box feel somewhat weaker than the vibrations from the 8" driver. (This is with bass bin running alone vs upper baffle running alone with the same signal.)
Long waves below 120Hz are still present in the 8" driver even with a LR4 xover, and that makes the baffle vibrate no matter how it is secured at the bottom.
At least with the W-frame woofers, their vibrations are mostly perpendicular to the radiation direction of the upper baffle drivers.
The Linkwitz solution of the bridge that straddles the bass bin is inelegant & clumsy, and it doesn't prevent the baffle from vibrating with the 8" driver.
Note: I might have to try a CLD with 2 layers of 12mm or 15mm BB. The MDF board used as the second layer on my current CLD baffle may not be rigid enough for this application. I also have some 10mm or 12mm aluminum honeycomb board (used in commercial airliners) that could be worth trying in a CLD between 12mm BB boards -- though that makes it a pretty thick baffle. The aerolam is very rigid, resists bending much better than similar thickness BB. Not sure if 9mm BB is rigid enough or even available here.
The Orion went through many revisions. 2004 is an early version -- did it have the rear tweeter or the magnet bracket for the midrange driver?
No problems of that sort heard in my original setup with Orions (v3.3), circa 2011~2012 with ASP, which was improved a little later with miniDSP DA8. The DSP allowed easy tweaking that improved overall presentation in my room. My main issue with the Orions in the current setup is that the midrange is a bit recessed, but not so imbalanced as valtergio describes his system. And there's too much reverb in my large sparsely furnished room, the dipole pattern isn't ideal for it, probably. Sounds is over-diffuse beyond 10~12' distance. Haven't tried any other speakers in there yet so, hard to say for sure. At 8~10' listening distance, it's pretty good.
Thank you, Mike. I've listened to the track "Prisoner in Disguise" from the album of same name (2014 edition streamed over Qobuz). It worked well on my system. The recording is noticeable 70s analog and her voice distinctive, but I found nothing unpleasant.
Speaking of Jennifer Warnes, I had added a 750Hz notch filter in my DSP system just for her, as her distinctive voice made me aware of a speaker resonance that wasn't apparent in my windowed measurements at the time.
I often compare sound with rather good headphones, these can be low in IMD and are free of time smear and room and directionality effects, I find many to be a good tool for identifying problems in loudspeaker and room.
Speaking of Jennifer Warnes, I had added a 750Hz notch filter in my DSP system just for her, as her distinctive voice made me aware of a speaker resonance that wasn't apparent in my windowed measurements at the time.
I often compare sound with rather good headphones, these can be low in IMD and are free of time smear and room and directionality effects, I find many to be a good tool for identifying problems in loudspeaker and room.
Interesting. Esp the 750Hz notch filter! What's your system?
What volume did you play LR? Obviously, if there's a problem, it's worse when played louder. Try any of the Spanish songs?
Headphone tip is a good one. I used to do the same, but have fallen out of the practise, mostly because I rarely use headphone now that I'm out in the country. No nearby neighbors to worry about. Better try headphones in the lab with LR though.
Sebastian K --
Tried Prisoner in Disguise through AKG K7XX Limited Red Edition headphones, TEAC UD-501 DAC, fed via coax directly from PC. Two versions -- 24/192 Rhino/Elektra remaster flac file & standard CD rip. Both files have much reduced harshness through the headphones compared to NEO8 but I still hear traces of it at higher volume especially when she's singing loudly.
Compared NEO8 (in test mule) to Scanspeak 10F/4424G00 (in 521 baffle) playing this tune -- both with my REW-derived EQ filters in place, LR4 HP at 500Hz. No contest -- the NEO8 exaggerates the harshness/distortions of Ronstadt's voice while the Scanspeak sounds far cleaner.
Slapped a 9kHz LR2 LP on the NEO8 -- much better! It must be that +9 dB on axis peak in the NEO8 that's causing this. I had hoped the NEO8 could be used by itself w/o a supertweeter, but that looks unlikely.
Tried Prisoner in Disguise through AKG K7XX Limited Red Edition headphones, TEAC UD-501 DAC, fed via coax directly from PC. Two versions -- 24/192 Rhino/Elektra remaster flac file & standard CD rip. Both files have much reduced harshness through the headphones compared to NEO8 but I still hear traces of it at higher volume especially when she's singing loudly.
Compared NEO8 (in test mule) to Scanspeak 10F/4424G00 (in 521 baffle) playing this tune -- both with my REW-derived EQ filters in place, LR4 HP at 500Hz. No contest -- the NEO8 exaggerates the harshness/distortions of Ronstadt's voice while the Scanspeak sounds far cleaner.
Slapped a 9kHz LR2 LP on the NEO8 -- much better! It must be that +9 dB on axis peak in the NEO8 that's causing this. I had hoped the NEO8 could be used by itself w/o a supertweeter, but that looks unlikely.
Last edited:
I have been using dipoles for 25 years, every time I have reduced the vibrations I have always obtained an improvement in reproduction.
vibration hitting the speaker and or any other equipment damaging the sound.
if you are able to hear it is another story.
if your system does not have sufficient resolution you will not be able to feel the small differences.
the LX521 design is structurally minimalist and accepts compromises, giving precedence to ease of construction, the Lx521 bridge is used only to isolate the excessive vibrations of the woofers, but even the baffle only does not have the rigidity and weight necessary to contain the vibrations of U22REX,
I read on hifiSentralen that the baffle vibrates even 2 mm, and I have no difficulty believing it if 200w channel amplifiers are used.
in fact someone more demanding made the aluminum baffle and said it made the bigger difference in sound.
when you are able to feel the vibration with your fingers your speaker is losing resolution, how much?
until you have solved the problem you cannot feel the improvement.
no the problem I felt was not the vibration of the W22, the problem was related to the not yet optimized crossover as it will be changed many times in the following years and all the various versions of Orion.
valtergio --
The point of your post is unclear. Can you reword or restate differently? Are you saying my vibration analysis is irrelevant unless I can hear the effects? Or...?
No question that reducing vibration in the enclosure of a speaker system is beneficial. My point is that separating the upper baffle from my bass bin will not serve that goal, because the bass bin has virtually no effect on the upper baffle, the bass bin hardly vibrates at all, and what vibration it makes goes up and down, not front to back like the midrange drivers. In contrast, the 8" driver swamps the whole baffle with vibration when it's working. That, to me, is a much bigger problem than the bass bin.
I double-checked all this after my last post. There's almost no vibration felt in the 521 baffle with just the bass bins running. Force cancellation really works, and the W-frame is so much more rigid than the v-frame.
In retrospect, the upper driver baffle of the Orion is far more stable mechanically, being well braced along its sides all the way to the top. The minimalist size & shape of the 521 baffle may well provide dispersion benefits but there's definitely a reduction in mechanical stability.
All this assumes that speaker drivers work best when their cones are pushing/pulling the air without their frames also moving in the same direction.
The point of your post is unclear. Can you reword or restate differently? Are you saying my vibration analysis is irrelevant unless I can hear the effects? Or...?
No question that reducing vibration in the enclosure of a speaker system is beneficial. My point is that separating the upper baffle from my bass bin will not serve that goal, because the bass bin has virtually no effect on the upper baffle, the bass bin hardly vibrates at all, and what vibration it makes goes up and down, not front to back like the midrange drivers. In contrast, the 8" driver swamps the whole baffle with vibration when it's working. That, to me, is a much bigger problem than the bass bin.
I double-checked all this after my last post. There's almost no vibration felt in the 521 baffle with just the bass bins running. Force cancellation really works, and the W-frame is so much more rigid than the v-frame.
In retrospect, the upper driver baffle of the Orion is far more stable mechanically, being well braced along its sides all the way to the top. The minimalist size & shape of the 521 baffle may well provide dispersion benefits but there's definitely a reduction in mechanical stability.
All this assumes that speaker drivers work best when their cones are pushing/pulling the air without their frames also moving in the same direction.
@Silent Mike: I have a Fane 12-250TC 12" wideband atop a damped U-frame box with a Fane 12-500LF, it's more heavy twin brother, crossed at 350 Hz acoustic LR4, which is about the lower bandwidth of the WB running nude, although I have damped the rear output with woollen rugs ("upcyling" - wool is excellent acoustic damping!) The damping of the dipole rolloff starts at about 100Hz for the bass, I'm equalizing to 40Hz (nearfield measured).
All that talk and theories about dipoles and cardiodids you can find here and elewhere are good and interesting, but after a while I found I needed to go for an experiment. I did choose the Fane widebander because
* It's inexpensive at 79€
* 12" and a bit of xmax allows for some good bass EQ
* No negative opinions found on the internets
* The rear radiation is severely compromised for smaller drivers due to diffraction at basket and magnet. This one has a steel basket but with some breathing space due to driver size.
* The data looked good: 96dB usable sensitivity, low mechanical losses even while the construction is quite basic
* data and opinions made me believe that no tweeter would be needed and complicate everything.
Previously I had put the widebander in a open baffle, then experimented with rear damping and found that bass output, bass distribution in room, imaging focus would improve but a trade-off with adding some sort of midrange distortion has to be made. I blame reflections from too rigid material behind the speaker. Sort of halfway between a nude speaker and a box speaker. (No innuendo intended!)
I'm running it all active from a noiseless mini PC with Lubuntu Linux and @Tfive's Pulseaudio Crossover Rack for IIR DSP work, plus some packages to make it accessible via UPnP and Bluetooth. Amplification is ... um, are you seated? ... a 10 years old entry level Pioneer AVR connected via HDMI. I have slightly annoying hum on one channel, and I'm having software problems now, also I learned the hard way that HDMI has drawbacks as a hardware standard, so I would not recommend that approach and are considering to go for passive XO, a higher valued stereo amp and different SW for DSP.
The sound... as said earlier I did only once find a critical voice or instrument and since then never a case where something really annoyed me. I now do find my Sennheiser HD600 headphone intransparent and with distortion issues that it'd put at around 500 Hz.
I listen at levels that the poor construction of my current dwelling allows, usually 55..65dB (smartphone "measurement"). As the speakers are really heavy duty parts I assume they can also do loud well.
I did never enjoy piano recordings before, the HD600 and any speaker I heard that over also, presents them wrong, the initial keystrokes are heavily distorted. No problem with the strange speakers, I like listening to Iiro Rantala sometimes. Good recordings are a 3D illusion happening inbetween the two speakers. Recordings of the compressed pan-pot style "made for earbuds" become unlistenable after hearing one good recording. I can listen to Adele's "25" album over headphones but not over these speakers.
However, the early setup with the single WB had a kind of "weightlessness" (can't really describe the difference, this word came to my mind while listening) that I no longer get. Might be the lesser group delay, the different room position, or both.
I have chosen the crossover point so that there would be no lobing. I did once vary the crossover point live, between 300 and 1000Hz no noticeable difference!! This, however, with some piece of music and in my seat only. I further see a funny effect that has been found in line sources as well: The Sound seems to emerge always at ear level height, for almost ground level up to about 20 degrees above vertically, then it becomes strange and "phasey".
I won't post pictures here as to not further undermine my credibility among the high end crowd. If anyone wants to further discuss please PM, I won't further discuss my setup on this board before I'm ready for it. If you haven't heard it you don't know it.
BTW, HDMI cables make a difference! I has pops and crackles, not unlike with dirty vinyl, at the beginning and experimented with buffer sizes priority settings in Linux audio to no avail. Then, I found that when I moved the HDMI cable at the AVR end just slightly either connection was cut off shortly or the noise would go away for some seconds. I replaced the came-with-some-device cable with one for 8€ cable with 20€ looks and better fit and the problem was no more.
All that talk and theories about dipoles and cardiodids you can find here and elewhere are good and interesting, but after a while I found I needed to go for an experiment. I did choose the Fane widebander because
* It's inexpensive at 79€
* 12" and a bit of xmax allows for some good bass EQ
* No negative opinions found on the internets
* The rear radiation is severely compromised for smaller drivers due to diffraction at basket and magnet. This one has a steel basket but with some breathing space due to driver size.
* The data looked good: 96dB usable sensitivity, low mechanical losses even while the construction is quite basic
* data and opinions made me believe that no tweeter would be needed and complicate everything.
Previously I had put the widebander in a open baffle, then experimented with rear damping and found that bass output, bass distribution in room, imaging focus would improve but a trade-off with adding some sort of midrange distortion has to be made. I blame reflections from too rigid material behind the speaker. Sort of halfway between a nude speaker and a box speaker. (No innuendo intended!)
I'm running it all active from a noiseless mini PC with Lubuntu Linux and @Tfive's Pulseaudio Crossover Rack for IIR DSP work, plus some packages to make it accessible via UPnP and Bluetooth. Amplification is ... um, are you seated? ... a 10 years old entry level Pioneer AVR connected via HDMI. I have slightly annoying hum on one channel, and I'm having software problems now, also I learned the hard way that HDMI has drawbacks as a hardware standard, so I would not recommend that approach and are considering to go for passive XO, a higher valued stereo amp and different SW for DSP.
The sound... as said earlier I did only once find a critical voice or instrument and since then never a case where something really annoyed me. I now do find my Sennheiser HD600 headphone intransparent and with distortion issues that it'd put at around 500 Hz.
I listen at levels that the poor construction of my current dwelling allows, usually 55..65dB (smartphone "measurement"). As the speakers are really heavy duty parts I assume they can also do loud well.
I did never enjoy piano recordings before, the HD600 and any speaker I heard that over also, presents them wrong, the initial keystrokes are heavily distorted. No problem with the strange speakers, I like listening to Iiro Rantala sometimes. Good recordings are a 3D illusion happening inbetween the two speakers. Recordings of the compressed pan-pot style "made for earbuds" become unlistenable after hearing one good recording. I can listen to Adele's "25" album over headphones but not over these speakers.
However, the early setup with the single WB had a kind of "weightlessness" (can't really describe the difference, this word came to my mind while listening) that I no longer get. Might be the lesser group delay, the different room position, or both.
I have chosen the crossover point so that there would be no lobing. I did once vary the crossover point live, between 300 and 1000Hz no noticeable difference!! This, however, with some piece of music and in my seat only. I further see a funny effect that has been found in line sources as well: The Sound seems to emerge always at ear level height, for almost ground level up to about 20 degrees above vertically, then it becomes strange and "phasey".
I won't post pictures here as to not further undermine my credibility among the high end crowd. If anyone wants to further discuss please PM, I won't further discuss my setup on this board before I'm ready for it. If you haven't heard it you don't know it.
BTW, HDMI cables make a difference! I has pops and crackles, not unlike with dirty vinyl, at the beginning and experimented with buffer sizes priority settings in Linux audio to no avail. Then, I found that when I moved the HDMI cable at the AVR end just slightly either connection was cut off shortly or the noise would go away for some seconds. I replaced the came-with-some-device cable with one for 8€ cable with 20€ looks and better fit and the problem was no more.
Last edited:
As I posted a few days ago, I decided to try my version of the AINO-Gradient 1.3 with a sealed bass box. Rather than build yet another box, I decided to make a temporary switch and listen/test in mono with one box.
A 2-layer back panel and similar front baffle with stiffening braces were built to fit the V-frame "box". The box was quite stuffed with recycled blue-jeans cotton insulation (fiberglass substitute), the edges lined with thick foam weatherstripping. The 2 panels were pressed into place, and 6 clamps used to tightly secure them to the V-frame box. One of the Dayton RSS265HF 10" drivers used in the V-frame was mounted & wires routed. The box is probably ~1.8 cu ft, agiving lower than 0.7 Q for this driver. It then moved into the lab and took position under the angled open baffle 12" and NEO8.
I was impressed at the first FR sweep I took of the closed box at 1m -- it's a nearly straight line from 40 to 600 Hz. Very easy to integrate with the other drivers, no EQ needed at all, really. I'm listening to it now. The bass and the overall sound strikes me as more natural than with the V-frame. There's a bit off bass boost that brings the -6 dB point to ~35 Hz. It sounds good at first listen. More measuring & listening...
A 2-layer back panel and similar front baffle with stiffening braces were built to fit the V-frame "box". The box was quite stuffed with recycled blue-jeans cotton insulation (fiberglass substitute), the edges lined with thick foam weatherstripping. The 2 panels were pressed into place, and 6 clamps used to tightly secure them to the V-frame box. One of the Dayton RSS265HF 10" drivers used in the V-frame was mounted & wires routed. The box is probably ~1.8 cu ft, agiving lower than 0.7 Q for this driver. It then moved into the lab and took position under the angled open baffle 12" and NEO8.
I was impressed at the first FR sweep I took of the closed box at 1m -- it's a nearly straight line from 40 to 600 Hz. Very easy to integrate with the other drivers, no EQ needed at all, really. I'm listening to it now. The bass and the overall sound strikes me as more natural than with the V-frame. There's a bit off bass boost that brings the -6 dB point to ~35 Hz. It sounds good at first listen. More measuring & listening...
