A friend of mine needed a center channel, so of course I recommended a waveguide speaker. He didn't want to spend a lot of cash (his other speakers were of middling quality) so I suggested he put two NS6's (the $9 each buyouts at Parts Express and Madisound) below a SEOS12/D350 in a 1.6cuft box and bring it over to my place for quick measurements and off-the-cuff crossover design. Matt brought it by last weekend and we scratched together a crossover from spare parts. As it turned out, I went too fast and forgot to enter in the Z-Axis offset in PCD... oops. With a few changes, we were able to get the speaker response back together, though. Sound was quite impressive, better bass than it should have had (rolloff at 72Hz), very sensitive (should be around 95dB@2.83V,1m). Really a pretty nice way to go if you are looking for a SEOS project on a budget that sounds like you went premium. Here is what it looks like: Matt plans to use it above his screen, with the horn below the woofers. I mentioned that it will look like a goggly-eyed face, so he told me that he's named the speaker "Daniel" and sent me another picture. (I should explain that Matt's last name is "Webster", and Daniel W. was known as a "great speaker". I had to google the quote to get the joke). I worked on the crossover a little after Matt left, starting from scratch accounting for the Z offset from the beginning (rather than a fix after the fact). All quite inexpensive parts, and a relatively simple circuit (for one of mine!). The letters in the schematic relate to a generic crossover board Erich will have available some time in the future, after I redo it to use the unnecessarily high powered resistors he wants to use. Check out the response (notice the phase, lower curve -- almost linear phase, this thing could almost do square waves!). A bunch of these would do for a budget surround sound system (all speakers set as "small", with an added subwoofer, or better, two of them). The only problem (besides less than stellar looks!) is that I'm told that the NS6es are starting to run out and are expected to be gone in the next several months. So if you are planning to build these (or my coSyne design, or other speakers that use NS6, you might want to stock up while you still can).
Wow, Terrific results. Sealed/Ported? Off-Axis Curves? Impedance? Better Off-axis with woofers closer together? I was wondering of four NS-6's in square formation together in series/parallel for an 8-Ohm impedance would be a tremendous 8-Ohm budget SEOS. JSS
Thanks Maxmercy, Box is sealed. Impedance is ~4 ohms min, with a peak at bass resonance point and a wide peak in the upper treble (that could be tamed with a few more parts, but a better approach is to just use a decent amp!). Yeah, I'd probably move the woofers closer together, though I didn't notice any off-axis oddities in listening (don't have the speaker here to test anymore). The NS6 has really nice midrange, I'm going to miss not having these dirt cheap highend drivers available for design, after they're gone. I'd like to make a NS6x4 Malcom, but the box would have to be huge.
That looks really good. I liked that you got a pretty flat phase out of it. I would have thought the large acoustic offset would have been trouble. On the 12" faital pro design I was working on I had a 2nd order crossover that gave a somewhat flat phase response. However I thought it was letting too much junk from woofer above the crossover point through so I bumped it up to a 3rd order which tossed the flat phase out the window. I have had family members tell me the dual 6.5"/waveguide speakers I built looked like a face as well. ;D
Sure, the D360, D350 and B&C DE250 are basically interchangeable (other than bolt sizes). D360 has better consistency than D350, but I doubt anyone could hear it. Parts Express's new 1" mylar dome driver would also work pretty well here.
BW, apologies in advance for a question that may not have a clear-cut answer, or may be retarded for some other reason, but here goes anyway: Is it possible to guess what the effects would be of adding an inch or so of vertical separation between the waveguide and the woofer array? I assume that would tend to reduce the height of the forward lobe, but do you think unworkably so? Other possible problems come to mind? The reason I ask is, I am looking for a center channel solution for my living room to match SEOS mains and I'm considering the feasibility of something like "Daniel" built into two discrete cabinets; the waveguide in its own box (just for looks of course) sitting on top of a typical home entertainment console table beneath the screen, and the woofers in their own cabinet sitting on the highest shelf of the same cabinet. The wood making up the top surface of the table is then sandwiched between the waveguide and the woofers, adding at least an additional 3/4" of separation. Thoughts?
Hi Bill, This design seems like a good candidate for my surrounds. PE's mylar CD is close enough to a drop in for the DE250? Have you tested with any of those new drivers? If so what do you think of them? Would the DE150 on a seos 12 work here with crossover workup? That would be an ultimate budget surround speaker. Nice work thanks!
I need to upgrade my center channel (followed later on with some SEOS mains). Just ordered a bunch of NS6s. I am very interested in building a NS6x4 Malcolm box. Just how big would the enclosure need to be? I have lots of width to work with (going under a FP screen, however - would like to keep it no more than 13"~14" tall). Any/All suggestions are welcome.
Ditto. I'd likely be interested in a couple of ceiling mounted 4xNS6 malcolm-likes that Bill was thinking of working on at some point, for surround use. He was once looking to bug EricH when the landscaping season died down, donno what became of that effort. Granted, with word of NS6's finally running out, not sure if its a great plan to work up too many designs based around those. :-/ For my own selfish needs, I'd like to see exactly how big a box Bill would end up needing so that I might get me some of them "Reese"'s pieces ( ::snort:: ) before they run out.
The effect would be a narrower vertical range before you run into response oddities. Probably workable, but in general it's better to keep the drivers closer, particularly since Daniel's crossover point is pretty high (about 1600Hz). An inch isn't a lot, though. Can I assume you'll be naming your next "William"? ;D I've tested the PE mylar version as it was being worked up. Not a bad driver, not quite as smooth as the others though but that's up where I can't hear anyway! I don't know about the DE150. You could probably get it to work, but I'm sure the crossover would need changes. Kind of low crossover for that, though. I reran the box sim on the NS6 drivers, it looks like the Daniel would work reasonably in a box volume as low as 0.8cuft, so a Malcolm3 using NS6es would need 1.6cuft. The "trick" to Malcom is having only the middle two woofers playing at the crossover point, then bringing the outer ones down lower (around 600 Hz or so) to maintain directivity to lower frequencies. To do that wire the middle two in parallel (getting them to 4 ohms), and then the outer two in parallel with each other. Put a big capacitor (maybe 150uF or so) across the outer two drivers so they get shunted out at higher frequencies. Then wire the outer pair (with their capacitor) in series with the inner pair, and use the combination for the "woofer". Probably using that with the Daniel crossover would work pretty close (you'd have to measure to be sure) but it shouldn't be far off. Call it "Big Mal". (Sorry to answer in batch mode like this, but I read these on my Android phone in the daytime, but answering with that is a pain. I have to spend more time finding what embarrassing word substitutions it has injected in there, than the time I spend poking in a response! So I have to wait till I'm at a laptop before I can reply).
Bill, Most of the people wanting the 4 driver "Daniel" appear to want to use the speaker very close to the boundary of the floor or ceiling. Won't the higher Q of the small enclosure along with the boundary reinforcement plus the low frequency boost due to the roll off of the outer drivers lead to a frequency imbalance? It seems to me that you would need to rework the entire crossover.
The higher Q boosts around 100Hz about a half dB..... in the noise for room effects. There is no boost from the outer drivers. 2 drivers at 4 ohms goes to 4 drivers at 8 ohms -- power into drivers cuts in half, but radiation efficiency doubles, so net is no change in level.... it's all just directivity. Floor or ceiling nearness might be a factor, but that's true with any speaker. If you have to place there, that's where Audyssey or other EQ comes in, not a crossover.
It appears that PE still has a decent number of the woofers available, but I guess they're selling pretty quick. I can't recall how many they originally had, but they still have a couple thousand going by the shopping cart on the PE site. I've got a decent number of the DNA-350's as well. It's kind of interesting that most people went with the 360's when both are so similar. I tested the D-250 from PE a couple weeks ago. They sounded alright, but there were some unusual blips in the response that might not be normal, so I'm going to pick up another one and check it out. The enclosure for four of the Aurasounds would seem to be pretty darn big for a center channel, but if anyone has ideas, I could get some made up in the next couple weeks. I'm working on subwoofers and the 1 - 2.5 cuft SEOS enclosures right now.
For a 'Big Mal" speaker, I'd go with inside dimensions of about 33x11x9 deep. Would work above most LCDs. Yeah, I hear now there are still about 2500 or more left, so my warning was maybe premature. Around spring there were about 6500.
Thanks for the feedback BW, and most of all for the great work you've done designing all these speakers for us.
Pretty sure I am understanding the wiring description above. However; could I be so bold as to beg for a "making-sure" sketch (crossover, wiring and added cap.) This would be for a "Big Mal", using 4-NS6 drivers, and a sealed enclosure with internal dimensions of 33w x 11h x 9deep. Pretty sure I have captured the bigger details here - please correct me as needed THANK YOU!!
Ok, here you go - Again, not modeled or tested with the Malcom woofer arrangement, but should be close if not right on.
Dr. Waslo, THANK YOU, that is terrific!! My "Big Mal" drivers & waveguide shipments should start showing up next week. With this drawing I can now order the crossover parts - thanks again! As for the cabinet, I'm thinking: -3/4" MDF for the baffle, -either 1/2" or 3/4" BB for the box? (and it will be cross-braced)
Yes, and they have a discount on 30 or more! I plan to use these for both the front three and surrounds. For the Mains and Center I plan make a "Super-Malcolm" with 6 drivers and a ~4.5 cu ft enclosure. Unfortunately, I believe will need to do a completely new crossover using an inductor to roll off the outer 4 drivers instead of the capacitor shunt. I think impedance at crossover will be ~ 16 ohms so Bill's crossover will not work for this design. If anyone knows how to adapt Bill's crossover to 6 drivers I would love to here suggestions. Bill, could you disclose what software do you you use to model your crossover designs? I have a complete set of measurement tools for sound measurements.
If anyone is interested in a large freestanding design with 6 Aura midwoofers, below is a modification to Bills design that incorporates BSC. It should not effect the impedance at crossover(<0.01ohm), but will reduce the impedance above and below resonance to ~6 ohms. I based it on a baffle width if 14 inches which I calculated to require a 325Hz knee point. I may use an L-pad to adjust the amount of boost at least for prototyping. Of course the cabinet will have to be enlarged. This has not been modeled or tested yet.
rdrewmc, sorry for late reply. I had read your question but on my android (on which answers are hard to type!) and so I forgot and the board marked it as "read" so it went off my radar.... anyway, I use PCD these days. I did the original Malcom on an RF modeling program called "RFSim", which isn't an easy way to go. But it can do networks that PCD can't (such as the Malcom arrangement). Bill