So my journey in diy has come full circle. Two years ago I bought an Onix Rocket HT package. I really like them but felt the high frequencies could use some improvement. After doing a search I ran across a a post in this forum by Jon Marsh about how the Scanspeak 9800 is nearly a drop in replacement for the Vifa XT25. Well I bought the SS9800's but before I could install them all the research had given me the diy bug. I decided that I was going to replace my Rocket package with something I designed. After talking many baby steps that progressed farther and farther I ended up building the little bookshelf speaker using Peerless HDS drivers. The sound quality was good enough that I knew I could make something better than the Rockets. This is the culmination of that journey, I'm calling them the Petra's because, well, they rock! Get it?
So a brief rundown. The mains and center are 3 ways using the Peerless 810921 tweeter, Peerless Nomex mids (832873 mains, 830872 center) and Dayton Reference woofers (RS270 mains, RS150-4 center). The rear surround will be 2 way speakers using the 810921 and 832873. Why these drivers? Well I just love the 810921. Hard to work with but just a great tweeter. The 832873 were compared against the RS150's. I have no real preference at this point, but I liked the how easy to work with the Peerless was. What sealed the deal was the 830872 being a superior driver. Easy to work with response and better distortion figures than the RS125 or W4-1337. And it just barely fit the baffle of the PE 1.0cf box with it's truncated frame. So Peerless was selected for mid duty. For woofers it's tough to beat the Dayton Reference. The RS270 is an excellent driver for this project. For the center channel I knew I was going to use a 4 ohm speaker in series so I could keep the impedance reasonable. Also it was perfect for matching the mid's sensitivity with little BSC built in. This was tough choice between the RS180 and RS150. Both model somewhat similar in the PE 1.0cf box. Being a center channel I wanted to make sure their was no chance of chestiness, and both of these drivers are just barely happy with 1.0cf. I figured the RS150 would be just a bit more comfortable here, but what really helped me decide was the fact that the RS150's modeled a higher sensitivity and I already had a pair, so...
Crossovers will be fairly conventional. The mid/tweet will be 1750-2500hz. The woofer/mid will be 200-350hz depending on the baffle step frequency, floor bounce calculations, driver layout etc. The one departure from conventional will be using a Behringer CX2310 for the RS270 to TM crossover. I figure if I can get rid of a big old (and expensive!) inductor, and have some flexiblity with BSC and voicing why not?
I'm using PE enclosure except for the RS270 which will be bass bin style. I optimized woofer placement according to box resonances predicted by boxnotes, and crossover frequency. Floor bounce calculations indicated that I'll be fine up to about 450hz with the woofer, so no worries there. Actually the mid in the mains will be the limiting factor, with a floor bounce of 375hz. I like to cluster the drivers, so elected to raise the woofer up near the mid/tweeter instead of being near the floor.
The RS270 enclosure is 69L net, tuned to 24hz. This provided the best trade off of decent bass when running full range, while still crossing properly to a sub. I planned on using a 12" TC2K for the sub and crossing at 50hz. Here is the interesting part: an RS270 in a sealed .7 qtc alignment has an F3 of 52hz. In other words a perfect match to my sub. Now sealed is bit lean for my taste so it was time to get creative. I wanted to find a ported alignment that would follow the sealed rolloff with the 50hz crossover in place (from the HT receiver). With the above alignment and a 60hz BW2 crossover this is what I get (sealed black, ported red):
Pretty cool huh? I love SE for being able to apply active filters in the enclosure designer! Now this of course requires an HT receiver that can apply different XO's to different channels, instead of globally. Fortunately my Rotel can do this. So what about the others?
The RS150-4's were placed in a 26L net box tuned to 45hz. Here is the RS150-4's with 50hz BW2 crossover compared to the reference RS270 sealed alignment:
Not bad. Yes the RS180-4 was marginally better. The 832873 in the rear surround is in a 10L net box tuned to 52hz. Here is the 832873 with 50hz BW2 crossover:
Again not bad. The 832873 in the mains will of course be sealed, being crossed to the RS270. So all in all I'm pretty happy with the results. Here is the raw modeled response without crossover (RS270 black, RS150 red, 832873 green):
Here are some pics of the speakers. The RS270 bass bin is just a rough mockup for measurements, the actual one will have some sort of hardwood inlaid on the front baffle.
More to come when I get the crossovers modeled.
So a brief rundown. The mains and center are 3 ways using the Peerless 810921 tweeter, Peerless Nomex mids (832873 mains, 830872 center) and Dayton Reference woofers (RS270 mains, RS150-4 center). The rear surround will be 2 way speakers using the 810921 and 832873. Why these drivers? Well I just love the 810921. Hard to work with but just a great tweeter. The 832873 were compared against the RS150's. I have no real preference at this point, but I liked the how easy to work with the Peerless was. What sealed the deal was the 830872 being a superior driver. Easy to work with response and better distortion figures than the RS125 or W4-1337. And it just barely fit the baffle of the PE 1.0cf box with it's truncated frame. So Peerless was selected for mid duty. For woofers it's tough to beat the Dayton Reference. The RS270 is an excellent driver for this project. For the center channel I knew I was going to use a 4 ohm speaker in series so I could keep the impedance reasonable. Also it was perfect for matching the mid's sensitivity with little BSC built in. This was tough choice between the RS180 and RS150. Both model somewhat similar in the PE 1.0cf box. Being a center channel I wanted to make sure their was no chance of chestiness, and both of these drivers are just barely happy with 1.0cf. I figured the RS150 would be just a bit more comfortable here, but what really helped me decide was the fact that the RS150's modeled a higher sensitivity and I already had a pair, so...
Crossovers will be fairly conventional. The mid/tweet will be 1750-2500hz. The woofer/mid will be 200-350hz depending on the baffle step frequency, floor bounce calculations, driver layout etc. The one departure from conventional will be using a Behringer CX2310 for the RS270 to TM crossover. I figure if I can get rid of a big old (and expensive!) inductor, and have some flexiblity with BSC and voicing why not?
I'm using PE enclosure except for the RS270 which will be bass bin style. I optimized woofer placement according to box resonances predicted by boxnotes, and crossover frequency. Floor bounce calculations indicated that I'll be fine up to about 450hz with the woofer, so no worries there. Actually the mid in the mains will be the limiting factor, with a floor bounce of 375hz. I like to cluster the drivers, so elected to raise the woofer up near the mid/tweeter instead of being near the floor.
The RS270 enclosure is 69L net, tuned to 24hz. This provided the best trade off of decent bass when running full range, while still crossing properly to a sub. I planned on using a 12" TC2K for the sub and crossing at 50hz. Here is the interesting part: an RS270 in a sealed .7 qtc alignment has an F3 of 52hz. In other words a perfect match to my sub. Now sealed is bit lean for my taste so it was time to get creative. I wanted to find a ported alignment that would follow the sealed rolloff with the 50hz crossover in place (from the HT receiver). With the above alignment and a 60hz BW2 crossover this is what I get (sealed black, ported red):
Pretty cool huh? I love SE for being able to apply active filters in the enclosure designer! Now this of course requires an HT receiver that can apply different XO's to different channels, instead of globally. Fortunately my Rotel can do this. So what about the others?
The RS150-4's were placed in a 26L net box tuned to 45hz. Here is the RS150-4's with 50hz BW2 crossover compared to the reference RS270 sealed alignment:
Not bad. Yes the RS180-4 was marginally better. The 832873 in the rear surround is in a 10L net box tuned to 52hz. Here is the 832873 with 50hz BW2 crossover:
Again not bad. The 832873 in the mains will of course be sealed, being crossed to the RS270. So all in all I'm pretty happy with the results. Here is the raw modeled response without crossover (RS270 black, RS150 red, 832873 green):
Here are some pics of the speakers. The RS270 bass bin is just a rough mockup for measurements, the actual one will have some sort of hardwood inlaid on the front baffle.
More to come when I get the crossovers modeled.
Comment