Re: Now... Linear Phase First of all great job Bill! As for biquad via miniDSP, it's definitely doable with the "advanced" plug in. In fact my last miniDSP transfer function was 100% biquad values plugged in - no "built in filters" were used at all. This thread may be a good reference: http://www.avsforum.com/t/1285070/bi-quad-filter-programming As far as those coefficients discussed in Post 12, you may not need them anymore as I believe miniDSP has updated their software since then.
Here ya go, based on the original Lambda Unity I think (not mine!): http://sketchup.google.com/3dwarehouse/details?mid=309a1989ac64db96729851fa1fa5f1f0&prevstart=0 I'm another lurker coming out of the woodwork. I have a 2-way Paraline built that will feed a horn based on bwaslo's spreadsheet. The horn is also built just waiting on drivers. I'm digging the linear phase stuff as well!
Here's an attempt at playing back squarewaves: http://youtu.be/C5mEQVoaJFs ...or not?... "An error occurred, please try again later". Oh well, my first attempt to produce a YouTube video.. OK, now it's working. Bear in mind that to make something look like a sharp square wave, all frequency components from about f/10 to 10*f have to be reproduced in correct amplitude and phase. So I wouldn't expect this to do much below maybe 600Hz (which is about where it starts to resemble a square wave) or above 2kHz (where the high harmonics don't get recorded to the CD or picked up by OmniMic). There are a few places where it looks kind of strange, might be where the response variations cause problems or the phase is too far off. But it does look "square-ish" (squaryness?) over most of the sweep. I wish I had some commercial speakers here to try this with to compare.
How about a teaser? Crappy phone shot. The horn's made out of cardboard hardened with polyester resin. It still flexes pretty good, I'm trying to figure out how to stiffen it up without spending days on it. I'll build another to match so I can listen in stereo until next year when I can hopefully do them up in MDF. I may see about using my cardboard method to make a mold and lay them up in fiberglass. That would be a lot easier than cutting all those angles!
those angles aren't all THAT hard to make on a table saw, just make sure you make yourselve sled jigs to use and plan all your cuts and then it's pretty hard to mess up.
Re: Paraline plus SEOS What do you think of the wild and (hopefully not so) crazy idea of using a paraline to couple both a CD and mid(s) to the back of a SEOS? Its a lot easier to build; might be smoother; need a large SEOS to control directivity down low but even with a 12" should get a point source
I don't think that would be easy at all, and I'm not sure it would work. The mouth on a "typical" Paraline is about 10" tall x 1" wide. The throat of the waveguide would have to match up to this dimension, and on a SEOS it looks like you'd have to cut it nearly in half depth wise to get the Paraline on there. I'll be honest, I don't know much about horns, just enough to get in trouble. I can't imagine this altered geometry would work though. There's a post in the Paraline thread I think over on DIYA that explains this really well - I'll see if I can find it.
You have to modify the paraline design to do it. Recall Patrick Bateman's video of a paraline with a square exit instead of a slot. I would design the paraline to have a round exit to match up to the SEOS throat. The way to do this is to take the standard paraline design in the patent and change the exit in the front plate from a slot to a centered 1" circle and then change the eye shapes in the inner layers to circles of about the same size. (Would that make it a paracircle?) Then you can surround the CD on the backplate with 4 small mids or cover it with one large one. I think the trick there is finding the best way to couple the mids into the input chamber. If using small mids, then I think the game is going to be getting the holes as close to the CD entry as possible, just like with a standard Synergy. This is what Danley seems to be doing. If using a large mid that covers the CD, then you can try to couple the mid into the circular slot between layers with a number of small holes spread uniformly around the circle. There they would be at max distance from the CD so less likely to disrupt the highs but still retain constant distance from the mid to the exit hole for coherence. My inclination is to go with the large mid using a vented rear chamber to extend the low frequency response, like Bill did. Right now its just something to think about and perhaps try to model
JackNC - sounds interesting, guess I wasn't thinking outside of the box! I wonder how that would work with equal path lengths and equal energy around the diameter of your circles. Beam like a laser or wide enough to let the SEOS take over? I wonder how you would handle the mouth of the Paraline. A quick calc for the circular mouth of a Paraline with the same path length as mine (5"on the long ends) would equal a little over 3" diameter. I'd think you would need a large "phase" plug protruding well into the SEOS. Fun to think about anyways! mtg90 - Will the BA do a clean 300hz? Dig the youtube Synergy vids ;D you ought to know the benefits of the Synergy design
Re' BA BA driver directly on SEOS with mid below doesn't give point source behavior. When your mid is on a baffle, baffle step messes with the flatness of your power response. With mid and CD combined into a single point line source , then you have both controlled directivity, no baffle step to worry about, and (so I hear) better imaging. Recall what BWaslo said comparing his Synergy to his TD15M/SEOS - the Seos was smoother up top but the Synergy was more seductive. Marrying a paraline to a SEOS is attractive because it gets me into this game with the least cabinet building effort. But I wouldn't be surprised if I ended up with the line source version in order to get tighter vertical dispersion - the original point of this thread. Jack
edit end of previous post - not this thread but the New Center Channel for the Octagon Thread on AVS, which also discusses paralines
Well you sure are not going to get the paraline through the 1" throat on the SEOS without major compression in the lower midrange just not enough area for all that air to move. So my point was why even bother if you cannot run it lower then say the BA driver. I guess I should have eleborated that a bit more in my previous post. You just can't build something that will bolt onto the back of a 1" entrance SEOS and cover the full range. Unless I misunderstood what you were describing?
As described in Danley's patent, the reason that Synergy works for lower frequencies is because the midrange is ported into the horn at the point where the horn's expansion rate is right for the longer wavelength involved. You could use a BA driver and get out of the SEOS all it can do from that geometry (down around 600Hz or so with a SEOS18), but to go lower, you should enter the horn further out from the throat. Just cutting the horn and sticking in a Paraline in the resulting larger throat hole probably wouldn't be easy or do so well, I don't think. The 90 degree Synergies I made are only about 11" deep with driver, not much to save in depth from using a Paraline, I wouldn't think. Though, as you say, you could put woofers in the SEOS horn walls just to get the drivers closer together so they act as a point source. I looked into that, too. But the SEOS walls don't have a lot of flat surface on the outside, with the effort needed to change things to approach that, you could probably make a regular Synergy horn and get decent directivity down lower. I think I may have scared some diyers off with my finicky instructions, the Synergy horn part isn't really as hard to make as it might seem. It just takes time (and extra wood to accommodate learning curve screwups). My friend Tony in UK is making some Synergy horns from plywood, and came up with a decent way to hold things while assembling and while glue is drying. He made an assembly jig with a board and some small pieces of wood (or mdf) attached with hot melt glue to it where they hold the wide edges of the horn panels in place (while being assembled with throat end up). And he uses little dabs of hot melt to hold the corners of the horn together on the outside of the joint while gluing, the hot melt hardens almost instantly and can be easily chipped off after the glue has dried. I'll try that on my next horns!
Thanks for the reminder, I forgot about that thread over there already. Good stuff! My Paralines are 10" tall and the eye is sized hopefully to get about 40 deg vertical dispersion. I did some vertical measurements of the Paraline itself but got some really nasty comb filtering and I can't make anything of the data. I think it's due to the fact that the Paraline is a line source after all and I don't know how to properly measure one along it's length, at least not in my living room. Hopefully this winter is like last "winter" (that is almost nonexistent!) here in Minnesnowta......my mids won't be here for another month probably and I'd like to get these Paraline horns outside for some real measurements.
You are correct. I started worrying about that as soon as I tried to sleep last night. Squeezing back down to a 1" throat exit will no doubt kill it. The soundpath is supposed to be a continuous expansion just as in a 3D horn. Its not clear that you get exactly that with the line source paraline but its clear you are not even close with a 1" exit.
Missed those last two posts because my browser was already open on the last page when I sat down. Its the compression I worried about if thats not apparent. (why can't we edit posts?) Bwaslo: I've always been both fascinated and intimidated by the synergy concept but I'm getting more comfortable with it. To visualize the horn construction, I think of making a basic corner/wedge and then cutting it down to the pyramid shape with two passes of my chain saw and a third to square off the apex. I'd probably invest in a new chain before attempting this Hot melt glue is a great idea. I just used blue tape to hold the angled pieces of my oak SEOS tophats together while gluing and it worked but only just barely. Tape just doesn't hold as tight or a precisely in place as one would like. Hot melt glue on the inside edges of the angles where it wouldn't show would have been just the trick Jack
I get only $11.50 shipping from California to Georgia. The coupon does not now show. It was for 10% off. I signed up for their discounts newsletter so hopefully that will have another one in it.
Does this look like the correct shopping list for 3 speakers? Note that i upgraded the 4uf for the tweeter to Dayton PMPC instead of DMPC (hardly any price difference in the big picture anyways), and there wasnt a 6uf on PE so i choose a 4uf and a 2uf instead.
hard to compare the list and the schematic on a cellphone, but nothing looks out of place there. You coulx get away with a 0.47 or a 0.55mH inductor for the out of stock 0.5mH if you're in a hurry.