Here's my latest Accuton project. I've posted this over on HTguide.com and PE Techtalk, but I think the "crowd" over here is different enough to warrant cross posting. Apologies in advance to those who feel differently.
This has been an interesting build for me. After trying out the Accuton C-79 accuton® Carefully selected loudspeaker drivers. and C-12http://www.accuton.de/drivers/detail.php?driver=1&matID=4&appID=1 in an older build, I knew these drivers were special. However, despite hearing some magical qualities in each driver, I struggled for months to get the system tonal balance right. Unfortunately, I never got to a place where I found them satisfying in the long term. I decided to shelve the drivers for a while and build another Accuton based project as interim speakers: accuton 3 way 2013 - a set on Flickr After mulling over exactly how I was going to attempt another shot at the C79/C12 combo for years, I finally dipped my foot into the deep end by approaching Dave Pellegrene about coming up with a waveguide for the C12 after seeing just how smooth the off-axis response can be if properly designed. (I was under the misconception that all waveguides had poor off axis response.) Knowing that the C12's bottom end is not it's strong suite, and finding out that most waveguide loaded tweeters need little more than a VERY small value capacitor for a crossover, I was SOLD! Next was the midrange. I wanted to try something different out here as well: aperiodic loading. My experience with open back mids was ultimately frustrating, so I thought a compromise between open back and sealed box might be the way to go. Instead of going with something like a variovent, I just drilled a bunch of 1/2" holes in the back of the midrange chamber knowing that the rear mounted crossover board was going to limit (to a degree) some of the back wave. I'd then be able to cover and/or expose the holes with damping materials (I used primarily F13 wool felt) to fine tune the output to the rear. The lower midrange/woofer is handled by ACCUTON C²95-T6 . I'm running it in an unusually large cabinet (about 30L) and tuned WAY below standard tuning since the entire system is electrically high passed at 80Hz, and past experience has demonstrated that this driver does NOT like to be run in a sealed box regardless of the passband used (I've run it as a midrange driver in a sealed box in the past.) Just ask Jeff @ Sonic Craft about running one in a sealed box, then hold the phone AWAY from your ear while he rants. Raw driver ground plane measurements were taken in box outdoors. (This becomes important later....)
Tweeter to mid crossover turned out to be in the exact same spot as past attempts: 2.7kHz. I thought that the waveguide loading would influence the crossover point, but apparently not. That seems to be where these two drivers like to hand off. The crossover from the mid to woofers is going to raise eyebrows: 700Hz. I know, I know.... Crossing the C79 down in the more typical 200-300Hz range never worked for me in the past: the resultant sound always lacked body and punch despite measuring well. When run in a sealed box I have had no problem running them down to 400-500Hz, and that was what I had in mind until I started messing with crossover sims: I couldn't get the crossover point lower without a lot of arm twisting and brow beating, so instead I ended up using a simple 1st order electrical high pass. This is one reason I went with the C95 for lower mids/woofer duty: it's super easy to work with, no huge peaks at the top of its passband and works just fine as a mid-woofer on it's own. That (and more importantly) I had a pair sitting on the shelf. ;-) When I got the speakers assembled and installed in the listening room is when the fun began. A lot of things were really amazing about the sound: had a seamless quality like a Martin Logan CLS. Unfortunately the tonal balance was even WORSE than a CLS: lacked body. Once I got the speaker positioning nailed the soundstage was ENORMOUS, yet images were tightly focused. I'd never heard anything like it in my room, even with designs utilizing a rear mounted ambience tweeter. (original Von Schweikert VR4 and original VR7) Keep in mind that I've never attempted a true dipole speaker in my system because of the unusual shape of my listening room: the speakers are right in front of the wall behind them. Without extensive acoustic treatments (not gonna happen with those huge paintings hanging there....) a dipole is not likely to work well in that situation. Well, since I was getting such good imaging I figured there wasn't that much sound coming from the midrange out the back of the enclosure. WRONG!!! Put my ear behind there and was shocked by how loud it was despite all the damping material in place. Hmmmm....This might explain why I was having a difficult time getting any body to the sound. And that would explain why my sims were so far removed from the sound I was getting: the measurements were taken outdoors, so the rear wave off the mids don't come into play. Time to play with the damping of the midrange enclosure. I initially went too far with the damping (mind you, I'm just using F13 felt inside the enclosure. No polyfill.) This is one of those subjective tunings: too much damping shuts the sound down. I backed off the damping until it stopped sounding constipated, but before I could hear sound coming out the back of the cabinet. Much, MUCH better! Now the sound was more in line with what the sim had indicated. I did lose that ENORMOUS soundstage that I had before, but good tonal balance is priority numero uno for me, so no complaints on that front. Furthermore, detail retrieval (such as delineating between a sound source and the ambient information of the acoustic surrounding that source) went way up. At this point it was just the somewhat tedious task of finding the right padding for the mids and tweeter. That and playing with the box tuning of the C95 (I've got a couple of fist sized rocks [!!!!] in each woofer enclosure to reduce the internal volume.) There was also some toying with the ports: stuffing them (makes hand off to the sub easier, but mucks up the midrange), hosing them (rubber banding panty hose over the ends to damp the port....no go.) I still can't believe just how much output there is from the ports (they're currently tuned to around 25 -30 Hz...I forget which) despite there being a 80Hz high pass on the system. <shrug> I still have to do something to eliminate port noise. I know if I didn't listen to some of the crazy electronica that I frequently enjoy, I'd never even notice it.
I have to say, the mids and tweet are not for the faint of heart. If a lifelike tonal balance is important to you, you might find these drivers frustrating. I've certainly seen plenty of DIY builds comment on this very thing. After days of messing with fine tuning of the crossover, I thought I'd never get enough "meat on the bones" to the sound without inserting tubes into the system. (Tube pre-amp anybody?) I think being a professional orchestral musician has made me a little overly sensitive to tonal balance issues. I've also found that getting a speaker to sound good on classical music as well as electronic/pop/rock can be a tricky balancing act: if it sounds appropriately rich on orchestral recordings, it can sound thick and/or dull on pop/rock. Get the presence region right for pop vocals and it can sound bright/lean on orchestral music. I can safely say that this speaker system has gotten closer to that ideal balancing act than anything else I've had. Is it the perfect speaker? Hardly. For instance, I know that there are other similarly sized speakers out there with greater dynamic range. But overall, I am more than pleased with the sound.
I wish I had more measurements/graphs to share, but about a week ago my computer was almost eaten by a virus. While I was having it fixed I had them do a fresh install of windows 7. In the process they managed to lose all my LspCad files, so all I've got are some of the screen shots that I took before the virus hit. I'll try to get some in room measurements just to satisfy my curiosity. Until then....
This has been an interesting build for me. After trying out the Accuton C-79 accuton® Carefully selected loudspeaker drivers. and C-12http://www.accuton.de/drivers/detail.php?driver=1&matID=4&appID=1 in an older build, I knew these drivers were special. However, despite hearing some magical qualities in each driver, I struggled for months to get the system tonal balance right. Unfortunately, I never got to a place where I found them satisfying in the long term. I decided to shelve the drivers for a while and build another Accuton based project as interim speakers: accuton 3 way 2013 - a set on Flickr After mulling over exactly how I was going to attempt another shot at the C79/C12 combo for years, I finally dipped my foot into the deep end by approaching Dave Pellegrene about coming up with a waveguide for the C12 after seeing just how smooth the off-axis response can be if properly designed. (I was under the misconception that all waveguides had poor off axis response.) Knowing that the C12's bottom end is not it's strong suite, and finding out that most waveguide loaded tweeters need little more than a VERY small value capacitor for a crossover, I was SOLD! Next was the midrange. I wanted to try something different out here as well: aperiodic loading. My experience with open back mids was ultimately frustrating, so I thought a compromise between open back and sealed box might be the way to go. Instead of going with something like a variovent, I just drilled a bunch of 1/2" holes in the back of the midrange chamber knowing that the rear mounted crossover board was going to limit (to a degree) some of the back wave. I'd then be able to cover and/or expose the holes with damping materials (I used primarily F13 wool felt) to fine tune the output to the rear. The lower midrange/woofer is handled by ACCUTON C²95-T6 . I'm running it in an unusually large cabinet (about 30L) and tuned WAY below standard tuning since the entire system is electrically high passed at 80Hz, and past experience has demonstrated that this driver does NOT like to be run in a sealed box regardless of the passband used (I've run it as a midrange driver in a sealed box in the past.) Just ask Jeff @ Sonic Craft about running one in a sealed box, then hold the phone AWAY from your ear while he rants. Raw driver ground plane measurements were taken in box outdoors. (This becomes important later....)
Tweeter to mid crossover turned out to be in the exact same spot as past attempts: 2.7kHz. I thought that the waveguide loading would influence the crossover point, but apparently not. That seems to be where these two drivers like to hand off. The crossover from the mid to woofers is going to raise eyebrows: 700Hz. I know, I know.... Crossing the C79 down in the more typical 200-300Hz range never worked for me in the past: the resultant sound always lacked body and punch despite measuring well. When run in a sealed box I have had no problem running them down to 400-500Hz, and that was what I had in mind until I started messing with crossover sims: I couldn't get the crossover point lower without a lot of arm twisting and brow beating, so instead I ended up using a simple 1st order electrical high pass. This is one reason I went with the C95 for lower mids/woofer duty: it's super easy to work with, no huge peaks at the top of its passband and works just fine as a mid-woofer on it's own. That (and more importantly) I had a pair sitting on the shelf. ;-) When I got the speakers assembled and installed in the listening room is when the fun began. A lot of things were really amazing about the sound: had a seamless quality like a Martin Logan CLS. Unfortunately the tonal balance was even WORSE than a CLS: lacked body. Once I got the speaker positioning nailed the soundstage was ENORMOUS, yet images were tightly focused. I'd never heard anything like it in my room, even with designs utilizing a rear mounted ambience tweeter. (original Von Schweikert VR4 and original VR7) Keep in mind that I've never attempted a true dipole speaker in my system because of the unusual shape of my listening room: the speakers are right in front of the wall behind them. Without extensive acoustic treatments (not gonna happen with those huge paintings hanging there....) a dipole is not likely to work well in that situation. Well, since I was getting such good imaging I figured there wasn't that much sound coming from the midrange out the back of the enclosure. WRONG!!! Put my ear behind there and was shocked by how loud it was despite all the damping material in place. Hmmmm....This might explain why I was having a difficult time getting any body to the sound. And that would explain why my sims were so far removed from the sound I was getting: the measurements were taken outdoors, so the rear wave off the mids don't come into play. Time to play with the damping of the midrange enclosure. I initially went too far with the damping (mind you, I'm just using F13 felt inside the enclosure. No polyfill.) This is one of those subjective tunings: too much damping shuts the sound down. I backed off the damping until it stopped sounding constipated, but before I could hear sound coming out the back of the cabinet. Much, MUCH better! Now the sound was more in line with what the sim had indicated. I did lose that ENORMOUS soundstage that I had before, but good tonal balance is priority numero uno for me, so no complaints on that front. Furthermore, detail retrieval (such as delineating between a sound source and the ambient information of the acoustic surrounding that source) went way up. At this point it was just the somewhat tedious task of finding the right padding for the mids and tweeter. That and playing with the box tuning of the C95 (I've got a couple of fist sized rocks [!!!!] in each woofer enclosure to reduce the internal volume.) There was also some toying with the ports: stuffing them (makes hand off to the sub easier, but mucks up the midrange), hosing them (rubber banding panty hose over the ends to damp the port....no go.) I still can't believe just how much output there is from the ports (they're currently tuned to around 25 -30 Hz...I forget which) despite there being a 80Hz high pass on the system. <shrug> I still have to do something to eliminate port noise. I know if I didn't listen to some of the crazy electronica that I frequently enjoy, I'd never even notice it.
I have to say, the mids and tweet are not for the faint of heart. If a lifelike tonal balance is important to you, you might find these drivers frustrating. I've certainly seen plenty of DIY builds comment on this very thing. After days of messing with fine tuning of the crossover, I thought I'd never get enough "meat on the bones" to the sound without inserting tubes into the system. (Tube pre-amp anybody?) I think being a professional orchestral musician has made me a little overly sensitive to tonal balance issues. I've also found that getting a speaker to sound good on classical music as well as electronic/pop/rock can be a tricky balancing act: if it sounds appropriately rich on orchestral recordings, it can sound thick and/or dull on pop/rock. Get the presence region right for pop vocals and it can sound bright/lean on orchestral music. I can safely say that this speaker system has gotten closer to that ideal balancing act than anything else I've had. Is it the perfect speaker? Hardly. For instance, I know that there are other similarly sized speakers out there with greater dynamic range. But overall, I am more than pleased with the sound.
I wish I had more measurements/graphs to share, but about a week ago my computer was almost eaten by a virus. While I was having it fixed I had them do a fresh install of windows 7. In the process they managed to lose all my LspCad files, so all I've got are some of the screen shots that I took before the virus hit. I'll try to get some in room measurements just to satisfy my curiosity. Until then....