My real OB speaker project

Discussion in 'Speakers' started by Serious, Mar 13, 2017.

Tags:
  1. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,519
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
    Looks like all the posts from the last two days or so vanished. Just going to repost what I wrote, including the shitty pic that I took at night with my phone:

    New.jpg


    As you can see I replaced the FE83En with 8" an widebander with whizzer cone. These are the Voxativ AC-1.6 drivers.

    I never really liked them at the High End shows, but I wanted to give Voxativ another chance. So I drove all the way to Berlin to give them a listen with my own gear on friday. There I knew within minutes of listening that they were for me. Simply put, these are the most realistic sounding, most resolving speaker drivers I've heard. The clarity, speed and articulation are incredible. The tonality of the drivers is almost spot-on, just that the different cabinets all add their own colorations. On-axis they have a rising FR, but it's smooth. The rising FR is needed so they don't sound dull in-room. Off-axis at maybe 15-20° it's flat. The whole off-axis response is fantastic. The dynamics are exceptional and I think beat every headphone I've heard. Transient response feels very natural: Not too sharp, but not dull like the FE83En. The paper adds very little coloration of its own and the little it adds adds a slight sweetness to the sound. The wooden cones use some special damping magic to keep them from ringing like mad. The treble is smooth, not only for an 8" widebander, but in general. I didn't believe that it was possible to make an 8" widebander with no resonances that I can hear, but they really sound like that. Very clean sound. Actually the FE83En has a rougher, grainier treble. The treble extension is great. I've seen measurements with -5db at 40kHz. The Lowthers don't even go to 20kHz. The imaging is actually more precise than the FE83En for some reason. Maybe my FE83En speaker needed more diffraction control (felt). I think even the bass quality was very good for an 8" driver and it didn't fall apart at higher levels.

    I also listened to the smaller driver, but thought it was less resolving and clear, not as fast, etc. I didn't find the smaller one to be more neutral, but I'm sure it measures better.

    I really can't emphasize enough how involving these drivers sound. I paid for the drivers on the spot and took them home with me. I stayed up listening to music all night. When I got up I started listening to music again. It's only now that I'm finally writing this, which I planned to do in the morning. Once you start listening you really don't want to stop. This is what it's about.

    What I was looking for was essentially a modded HD800 in speaker form. With the Voxativs I'm pretty much there.


    Two other options that I also considered were:
    1. A three way OB with an Accuton C173-6-90 midrange and either one of the ceramic Accuton tweeters or one of the high sensitivity Scan-Speak ones.
    2. A refurbed Quad ESL 63
    I have no doubts that I'd prefer the Voxativ over either one, but I generally find multi-ways to chop the music into pieces, which sounds less natural than a properly implemented widebander. The coherence is real. With other speakers you can hear that the tweeter is not where the midrange is. The 2-way I'm building gets a pass because it's essentially a full range with sub speaker, but done right with a real crossover. This way I can get closer to a full-range point source minimum phase speaker than with a BLH, though I will build a BLH cab for the driver eventually.
    Electrostats just don't sound as dynamic and resolving. They don't grab me as much.


    So where am I now with my project?

    The Voxativ driver is extremely efficient. I have my doubts if a 2nd woofer in parallel (+6db) will be enough. I thought that there was no way they're 100db efficient and expected closer to 95db, but it seems 100db is actually sort of accurate from what I've measured. I made the baffles slightly higher to bump up the response below 1kHz a bit, which seemed to work well in making the mids extremely neutral. This means I need more bass, but I also want to keep it passive.

    The crossover I have is still from the old speaker. Obviously it needs a complete redesign now. Actually there's a suckout at around 200Hz now, because the bigger driver goes so much lower. When I invert the woofer the bass is gone because the Voxativ driver still puts out a lot of bass, even in its small baffle. With my baffles I'm aiming at around 200Hz for my crossover, but I will have to measure to be sure.

    Because the driver is bigger the listening height is too high now. The center of the driver is about 95cm from the floor and I want to make it about 80-85cm. This is a matter of personal preference, but I ideally want to make it symmetrical, which means for 80cm listening height the speaker would be 160cm (5'3") tall. That's pretty tall, but I also don't want to make a speaker for midgets like Pure Audio Project did. I need to find a better way to mount the widebander baffle behind the woofer baffle.
     
    Last edited: Mar 13, 2017
  2. numbercube

    numbercube Acquaintance

    Joined:
    Sep 29, 2015
    Likes Received:
    71
    Dislikes Received:
    0
    Trophy Points:
    18
    Location:
    Germany
    Mono, so far? (Mono rules.) Voxativ's expensive, but their presentations at audio shows always impressed me the most, depsite their problems.
    You're driving them with the Rag, René?
     
  3. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,519
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
    Lol, not mono. I have a stereo pair of the AC 1.6s. Matching seems great, especially considering the low volume production. It's a lot of money, but they're exactly what I was looking for. Yes, I'm driving them with the Rag. The insane efficiency begs for a SET amp.

    I think I was too put off by their rising FR at shows to really hear what they could do. You absolutely have to sit in the right spot and I never listened to them with my own music.
     
  4. Audioviel

    Audioviel Rando

    Joined:
    Sep 18, 2016
    Likes Received:
    7
    Dislikes Received:
    0
    Trophy Points:
    3
    Location:
    Essen, Germany
    Did you try AER BD speaker?
     
  5. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    46,273
    Dislikes Received:
    72
    Trophy Points:
    113
    Location:
    Antarctica
    Holy crap! You actually got the Voxactiv drivers. I've always suspected bad implementation or bad setup at shows.

    You can do this for for a dual woofer "OB":

    Code:
    |   |----|   |
    |   |    |   |
    |   W    W   |
    |   O    O   |
    |   O    O   |
    |   F    F   |
    |   |    |   |
    |___|    |___|
    
    
     
    Last edited: Mar 13, 2017
  6. Daveheart

    Daveheart Friend

    Friend
    Joined:
    Jan 10, 2017
    Likes Received:
    602
    Dislikes Received:
    0
    Trophy Points:
    93
    Location:
    Dallas, TX
    I'd heard their Ampeggios twice at shows. I liked them a lot the first time, and thought they sounded terrible the second time. Though rather than developing any dislike for the speaker itself, I just assumed that the guy who set the room up was a boob.
    I've definitely seen good and bad implementations of Voxativ speakers at shows. One of the bad ones had them paired with super-tweeters. I actually think they'd pair well with super-tweeters provided the boob setting up the room* was capable wiring them in phase.


    *Said boob was a cable manufacturer. You'd think he'd be capable of getting phase correct, but you'd be wrong.
     
  7. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,519
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
    No, haven't heard them (except at hifideluxe last year). From what I gather the BD1 is almost twice as expensive as the AC 1.6.
    I'm not entirely sure what happened between Ines Adler and AER. I think Ines tried to sell her own drivers under the AER name, before founding Voxativ? Anyway, the AER drivers are probably very good, too.
    There's a lot not to like about Voxativ, but I think their drivers are fantastic. Honestly it wouldn't be hard for me to believe that the Voxativs are "better" than the AERs, but it's probably a matter of personal preference.

    This is what their subwoofer is. I didn't really like the sound of the woofer, but I might experiment with it one day, possibly as a subwoofer for the OB. I think the ripol concept gives you more subbass (moves Fs lower) at the cost of efficiency. Holger told me how 12" woofers sound faster than bigger woofers because of their lower moving mass, but what I heard didn't sound particularly fast. I switched the woofers off immediately after I started listening. 12" woofers will have more distortion, especially when asked to go down to 20Hz. I suspect that it's more an issue of 15" woofers being too large to pair with the Pi. That's probably the main advantage of the ripol woofer: saving space

    In my case I want to use the woofers up to frequencies that are relevant for the imaging, so I will mount one woofer above the widebander and one below, both in an H-Frame. That probably puts WAF at an all time low, but I think it gives me the bass I'm looking for. I can hear that the image shifts as it goes down in frequency. Hopefully this time around I can design a real 1st order crossover. The problem is getting anechoic measurements at 2 meter distance down to 200 or even 100Hz. The other option is to rely entirely on simulations.

    Might not be easy to get the phase right as the wavelengths are so short and the FR depends a lot on the position in the room.
    My ears tell me they don't need supertweeters. With the measurements it depends a lot on where you measure, though the whizzer and phase plug gives you good treble extension even at extreme angles. At the more normal angles I seem to get some sort of destructive interference around 15-16kHz, but listening to sweeps I don't think I can hear it and extension seems to be very good to the end of my hearing. It's not as airy as the HD800, but the HD800 might be too airy. It does sound more extended than the FE83En.
    I do think the treble generally sounds very realistic and would compete with expensive tweeters (at the right angle), but that might not be fair since I really believe the "best" tweeter is no tweeter.


    BTW: Holger told me they're working on a headphone. That might be something to look out for, but I fear that it'll cost 10k and sound bright as shit, considering how he set up his speakers in his room.
    On the other hand the Voxativs remind me very much of my modded HD800 the way I have them set up right now. I'll get measurements up later, but they're pretty darn good. These OB speakers already sound pretty different from the Voxativ speakers.
     
    Last edited: Mar 13, 2017
    Xecuter likes this.
  8. purr1n

    purr1n Finding his inner redneck

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    46,273
    Dislikes Received:
    72
    Trophy Points:
    113
    Location:
    Antarctica
    Yeah, it saves space. A pure OB will give you the best transient response. The transient response gets worse as the wings of the H frame get longer. Dunno why this happens. Probably why the ripol didn't sound as fast. Also, I've noticed that more efficient drivers (the 15") tend to sound faster. Doesn't Voxactiv use Eminence drivers for their sub unit, but with revised motors? Haha.

    Yeah. Sat next to him at a show with the Ampeggio Duos (those big panels) which were screeching bright. Looked over and I was like "are you serious?" while he was bobbing head to music.
     
  9. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,519
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
    I think it has to do with the H-frame essentially decreasing Fs and increasing Qts, but since it also gives you larger baffles it increases efficiency. The bass sounded slightly faster with the OB, but now I absolutely need the H-frame for efficiency.
    I think so.
    The thing with the Duos is that they use the field-coils, which I think are the strongest motors, so the drivers will be the brightest. On the other hand I think Holger told me the field coil motors sound warmer than the magnets.
    Obviously the baffles should help with the lower mids, but for some reason I've never been impressed by the Ampeggio Dues.

    BTW: Here's a FR comparison of the FE83En vs the Voxativ in the OB at 1m on axis in room. I tried to reference it to about 2.83Vrms, but it might not be perfectly accurate. This is with the woofer playing.
    FE83 vs Voxativ on OB.png
    I think the Voxativ driver was inverted with this measurement, which is why I got less bass. Baffle rolls off below 500Hz. Without the slight dip between 600Hz and 1.5kHz, which is probably from reflections somewhere and not the driver, the efficiency really seems to be close to 100db/W (keep in mind the AC 1.6 is really a 10 or 12 Ohm driver, so 2.83V is less than 1W). There seems to be a slight peak in the mid-treble and a tiny bump around 2kHz, but both go away at the right angle.
    Might be slightly high as the FE83En is more like 89db efficient, but the AC 1.6 is almost 10db more efficient.


    FR listening position 2.png
    This is the FR at the listening position. About 15° off-axis at 2m distance. It's not easy to get accurate measurements because it's much more directional than your typical speakers. Finding the right listening position is easy while listening, but not that easy when measuring. The deep dip at 250Hz is from the crossover that doesn't work yet.

    They're flat at the listening position, without a B&K target compensation. I used to be against this, but with these speakers I find it to sound right. It sounds natural, regardless of how it measures. This is how I set them up without measuring. The last octave is hard to measure with a microphone. I don't hear such a big dip around 15/16kHz.
    The driver matching is better than it looks. It's just not easy to get two measurements that look exactly the same. The whizzer cone itself is incredibly free of resonances and I think it crosses over to the main cone around 4-5kHz. The big cone might have a slight 2kHz resonance, but I don't think I can hear it. If it sounded shouty I wouldn't have bought it. The midrange is very good on these drivers.

    I'll try to get distortion measurements up later. Much better than the tiny FE83En. Distortion was very low across the board without any spikes.
     
    Last edited: Mar 13, 2017
  10. Donald North

    Donald North Friend

    Friend
    Joined:
    Oct 4, 2015
    Likes Received:
    357
    Dislikes Received:
    0
    Trophy Points:
    53
    @Serious, have you considered doing an active crossover between the woofer(s) and Voxactiv drivers? This is how I do it with my Sequence speakers and allows for a lot more EQ fine tuning of the bass as needed. This way you can also use a low wattage tube amp for the Voxactivs and higher power solid state or tube for the bass.

    About H-frame woofers, they don't right to me either. An ideal dipole should have a perfect null +/-90 degrees and this seems impeded by the sides of the H. From my experience I prefer to keep the woofer baffle small and stiff and use active EQ to compensate for the dipole self cancellation.
     
    taisserroots, Xecuter and Serious like this.
  11. Audioviel

    Audioviel Rando

    Joined:
    Sep 18, 2016
    Likes Received:
    7
    Dislikes Received:
    0
    Trophy Points:
    3
    Location:
    Essen, Germany
    I don't care about the dispute Keller-Adler. My friend had/has a Voxativ Ampeggio (AC-3X) and a Ayon Dragon (AER BD3B). Both impressed me a lot. Can't compare the drivers because of the difference of the speakers. (I had more fun with the Ayon)
     
    Serious likes this.
  12. Priidik

    Priidik Friend

    Friend BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    1,338
    Dislikes Received:
    2
    Trophy Points:
    93
    Location:
    Estonia
    I'm trying to hunt down a woofer for OB that does well and loud down to 30 Hz and cross over from somewhere 150..250 Hz. Not so trivial task as it turns out, I have been looking around for a while already but can't decide.

    I want it to be able to do microdynamics and be effortless in presentation. Choices for amps are all open as for now. I can go stupid power class-D to SET to anything in between. Bi amping doesn't look terribly problematic to me. I haven't yet made up my mind on x-over either. It's probably going to be passive 2-th order linkwitz all around.
    So the actual eff is not that important on it's own, if indeed it doesn't strongly correlate to better sound.
    It looks like I need at least 2x12'' units per speaker or better yet an 18''. But which one to choose?
    I have no substantial experience on how T/S parameters will translate to sound.
    I suspect that the big paper drivers that I have liked have had low moving mass and high mech Q.
    So is this trading pistonic movement (tightness) in favor of responsiveness ?

    So far I'm thinking low moving mass, high mech Q, high tech motor design (shorting rings and such) as priorities.
    Downsides with such drivers usually are high-ish Fs, not extremely rigid cone and moderate x-max.
    This has lead me to these as potential choices:
    http://precision-devices.com/Product-Details/PD184C01---Bass-Driver
    https://hebasound.de/de/product_info.php?info=p13246_bms-18-s-450-l---18--subwoofer.html
    http://www.sbacoustics.com/index.php/products/woofers/nrx/12-sb34nrxl75-8/
    The Satori WO24-P looks great, but I'd need lots of them and I wonder if it is all worth it vs a single large diameter woofer.

    Am I on the right track here, or should I look at bulky high excursion IB woofer instead?
    Does anyone have exp in how squeezing better low end extension by adding resistance in series to make higher Q translates to actual sound quality?
     
    Last edited: May 6, 2017
    dark_energy and Serious like this.
  13. Donald North

    Donald North Friend

    Friend
    Joined:
    Oct 4, 2015
    Likes Received:
    357
    Dislikes Received:
    0
    Trophy Points:
    53
    This would make an excellent dipole woofer:
    https://www.madisoundspeakerstore.com/approx-18-subwoofers/aurasound-ns18-992-4a-18-subwoofer/

    I led the product development on these when I returned to Aurasound in 2000. It has a very long throw, high-ish Q, large X-max, and very low noise.

    If I were able to design a custom version for dipole, I would find a way to reduce moving mass while maintaining a stiff cone, and adjust BL as needed to keep the Q around 0.5-0.6.

    In my opinion you don't want cones that can flex, be they paper or plastic or metal or composite. You can see if there are signs of flexing in the bass range by looking at the impedance curve and notice if you see any anomalies.
     
    sfoclt, Priidik and Serious like this.
  14. murphythecat

    murphythecat Friend

    Friend
    Joined:
    Apr 4, 2016
    Likes Received:
    659
    Dislikes Received:
    17
    Trophy Points:
    93
    Serious, you are in the deep end of Fullrangers with Voxativ, AER, Supravox! congrats! Ive decided to explore that avenue myself. the distributor in montreal will make me audition the supravox field coil!

    have you auditioned the supravox field-coil?
     
  15. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,519
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
    @Priidik
    All of those drivers have low Qts around 0.3 so you would need EQ to get decent bass extension in the open baffle. It's hard to find a woofer that does it all.

    The super cheap Eminence Beta 15A has a specified moving mass that is lower than even the 12" Satori you linked (60g vs 90g). Maybe moving mass does correlate with "speed" in the bass, but I'm not sure on that. The Beta 15 sounds less tight than good headphones, but I find it comparable in terms of speed, at least in the open baffle. Good microdynamics, too. Still it does sound much more colored than the HD800, for example. In my room it's almost flat to 40Hz and has good extension to the low 30s. Distortion is pretty low down to 35Hz, but starts to rise below that.
    I ordered a 2nd pair of Beta 15s, which arrived today. I already have the baffles and plan to mount them tomorrow. Since I'm essentially mirroring the whole thing for now, it'll be a very tall speaker.

    There's always the compromise between strong magnets, high Qts and low Fs. I'd be interested to see if there's a straight upgrade to the Beta 15, but most of those woofers are much more expensive. I actually suspect that many of these expensive woofers are sidegrades if anything.

    @Donald North
    That one looks interesting. These spec sheets are obviously not lying about sensitivity, but <87db/W looks pretty low to me. Obviously a compromise to get good bass extension. That looks like some nasty breakup at 2kHz.

    I tried bi-amping my speakers with a Class D amp for the bass. I think I can get closer to what I want by designing a passive speaker with first order crossover. In my room I got around 94db/W in the bass in the H-frame. I'm pretty confident that a second woofer in parallel will be able to match the sensitvity of my widebander. That will obviously not get me as good extension and I'll have a nasty 3 Ohm impedance in the bass. We'll see how it turns out.

    @murphythecat
    No, I actually haven't heard the supravox stuff. There are lots of widebanders that I haven't heard that I'd be interested to hear. Ultimately the Voxativs did wow me enough to make me stop worrying about other drivers, but I'm still interested to hear them.
     
  16. Donald North

    Donald North Friend

    Friend
    Joined:
    Oct 4, 2015
    Likes Received:
    357
    Dislikes Received:
    0
    Trophy Points:
    53
    Sensitivity is a calculated midband value. In my opinion it is not very meaningful as a criteria for selecting a woofer to use for dipole. What you want it: large volumetric air displacement capability (big Sd and/or Xmax) with low distortion and audible noise, pistonic behavior of the cone in the bass frequencies, low Fs, and a Q around 0.5-0.6. Higher moving mass simply means more power to achieve the same SPL, so lower moving mass is beneficial but not at the expense of pistonic behavior from a flexy cone or a wimpy voice coil.
     
    Serious likes this.
  17. murphythecat

    murphythecat Friend

    Friend
    Joined:
    Apr 4, 2016
    Likes Received:
    659
    Dislikes Received:
    17
    Trophy Points:
    93
    Serious, I doubt any speakers will sound as transparent as hd800 in the bass in a untreated room; you need a lot of bass trapping in order to have similar decay response in rooms.

    As for active vs passive, I think youll have to try yourself: I built a couple of minidsp active speakers and none integrated seemlessly like good passive speakers: even when doing FAST systems with very low and single XO, let alone trying to integrate a 12 inch woofer with a midrange driver at 300hz (something I tried to do but gave up eventually).
     
    Last edited: May 6, 2017
  18. Priidik

    Priidik Friend

    Friend BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    1,338
    Dislikes Received:
    2
    Trophy Points:
    93
    Location:
    Estonia
    Based on modeling a 5 ohm resistor in series will flatten fr response down to 25 Hz or so for the PD184C01
    driver, but I'm afraid it will not be free lunch in terms of sonics.
    Will a low e-Q driver with res in series sound the same to high e-Q driver for same total Q, all other things being equal ?
    A good thing about the PD driver is the exceptional eff and power headroom, I can get away with dsp or EQ if it comes to that.

    Given that these have underhung voicecoil the efficiency is rather impressive. The merits of UH over the OH coil have not been unequivocally proven in practice as far as internet smarts go, in theory at least the linearity should be better. Although based on measurements of BL and inductancecurves better OH designes with polecaps or rings are just as good, if not even better. A great example of OH performance :http://www.audioxpress.com/article/test-bench-scan-speak-32w-4878t00-subwoofer
    I'm reluctant to consider these partly because of price, partly because of lack of
    availability here. These look gorgeous though, on specs these are god mode, only caveat is that I'm not sure I get the microdynamics out of these.

    These Betas are not that high tech, perhaps that's why they sound colored and soft. Are they less so at low SPL?

    Maybe tightness in bass is function of stiffness of cone and lowest total Q ?
    A quote from Troels Gravesen's OB construction: From a visitor hearing the prototype OBL-11s, I can report that these 15" Eminence DeltaLites beat the crap out of Eminence Alpha 15A, which is often referred to in OB connections as being an ideal driver for this application. This was my expectation and what I had hoped for by choosing a fairly low-Qt driver and applying the equalisation step.
    On top of that I read somewhere that same driver (some Precision Devices unit) that was updated for higher mech Q had more kick/slam.
    Another thing often mentioned is that the panel must be stiff and heavy. Some dudes make sandwich with sand and veneer, other go with several layers of heavy wood. The wood panel on the picture looks a bit wimpy.

    I don't quite get it how you can get away with first order low pass with the Betas given the humongous spike at 2 kHz.
     
    Last edited: May 7, 2017
  19. Serious

    Serious Inquisitive Frequency Response Plot

    Friend BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    1,519
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Munich
    Installed a 2nd Beta 15 above the widebander. This is what they look like now. As you can see they're absolutely massive. It's actually slightly taller than me.
    OB1.jpg
    I tried padding down the Voxativ widebander with a random 6.8 Ohm resistor I had and it sounded terrible. Grainy, losing a lot of transparency. There are more expensive resistors that would probably fair better, but in that case I actually far preferred the sound without the resistor and with a worse FR.

    I doubt it.

    The thing is right now I want to be make a passive speaker, which means that I absolutely need high efficiency, even if that is (obviously) at the cost of bass extension. The Beta 15s just barely make it.
    When it comes down to it, no matter how you look at it you need a low moving mass for higher sensitvity. 400g vs 60g moving mass costs you about 6db efficiency even at Fs, while a stronger magnet doesn't help much.
    Yup. OJ also didn't like the Alpha 15 in his OBs and replaced it with one of the AE woofers. Both the Beta 15 and the Deltalite have much lower Qts than the Alpha, but the Beta actually has a lower Fs and higher Qts.
    The panel I'm using is super cheap spruce wood, 18mm. It's definitely wimpy. I've thought about making a sandwich with two of those and a damping middle layer, but it's good enough for now. Making it heavier also makes it much more difficult to move, say for measurements outdoors. You can feel the panels flexing, but it's not super horrible.
    It's not quite as horrible, because 1) I am listening slightly off-axis and 2) because I can use the Zobel circuit to slightly vary the slope. Overall I don't really notice it, even with just the woofers playing from their crossover. It's probably not as bad as with metal cone woofers. I'll take measurements later, but it's really not that horrible.
    When it comes down to it, it actually wasn't an issue with the 500Hz crossover with the FE83En, but in that case it was probably masked by a 2kHz resonance the FE83En has.

    I didn't explicitly state this and it does seem like the Blumenstein Orca has its following, but I found the Beta 15 to be a better driver than the FE83En. My other speaker was held back much more by the tiny Fostex widebander than the woofer. The Beta 15 actually sounded faster, snappier and more textured than the Fostex in that case. Even with the (far superior) Voxativ widebander I don't feel the need to replace the woofers.

    In terms of measurements speakers won't match the HD800, but I do think that good speakers can be comparable or even better in a good room.


    Anyway, how does it sound now?
    I'm still using a similar crossover, so the woofers now roll off too early because of the halved impedance. Crossover now obviously needs different parts, which I don't have at the moment. Still, it's better than I expected. First up, the imaging now obviously makes more sense. At the high crossover frequency I used I could hear the image shift with decreasing frequency with just one woofer. Bass guitars are now at the same height as everything else. Bassdrum hits actually sound like they're coming from a defined point in space.
    From early measurements I get about 5db more bass, which is better than I expected given that the upper woofers don't have the floor as a baffle and the Ragnarok isn't the most stable amp. I don't have the H-frame sides on right now, which means I lose about 3db. I could maybe use a bit more bass. Extension into the mid 30s seems decent and I'm not too concerned about bass below 30Hz. I actually wonder how much bass extension you would gain with a woofer that goes lower - the back wall reflection probably limits bass extension anyway.
    I'm also surprised by the bass quality gains, not just quantity. Even at lower volume levels it now feels more solid than when bi-amping one woofer to 5db higher. Bi-amping, at least with the amps I used, also didn't achieve the coherence. Another benefit of using two woofers like this is that it avoids certain room modes. I had two major resonances between 100 and 200Hz (see measurements above) before that now seem to be almost gone. The Beta 15 didn't exactly sound soft, but it did sound colored. I actually feel that it sounds slightly less colored now. Probably less 3rd order distortion at the same level. It does sound much more impactful now. Still not the best around, but in that regard it's also limited by the Ragnarok.
     
    Last edited: May 7, 2017
  20. Donald North

    Donald North Friend

    Friend
    Joined:
    Oct 4, 2015
    Likes Received:
    357
    Dislikes Received:
    0
    Trophy Points:
    53
    From my experience, underhung does offer an advantage of allowing the designer to put more mass in the cone (for greater possible stiffness) for the same target Mms than an overhung design which has a heavier coil.

    With the development of the Klippel analyzer I know some people use its test results as the final word. I can only say from experience that what measures supposedly well on the Klippel doesn't always sound better in real use.

    Lastly regarding microdynamics, in my opinion you want less stored energy in cone bending modes which can't be corrected and removed through EQ. As long as the cone is acting as a piston, the amp will tell it when to start and stop.
     
    Serious and Priidik like this.

Share This Page