Where to begin?

Discussion in 'DIY Speakers and Subwoofers' started by Aryn Ravenlocke, Sep 3, 2014.

  1. So, I just recently finished getting a new home media server installed in my home. By the end of next week, all 32 TB of media info should be done loading onto the server and it will be servicing my house. That means it is time to move on to the next portion of my home theatre - the sound stage. Unlike computers though, where I have just enough knowledge to be dangerous, I have absolutely no idea what I am doing when it comes to building speakers as I am pretty sure my forays into building subwoofers with coffee filters and nail polish in the late 80s does not count as truly useful experience. Nonetheless, the idea is to actually build out my entire home theatre myself (with help from others), not buy $250K+ of factory/store-bought equipment.


    So, where to begin? I do know what I want for reference level/minimum performance standards. I absolutely can get this standard in my price range by buying from a dealer. So in theory, I should be able to blow past that level if I build my own.


    I want a matched set for sound consistency. The pieces I finally settled on through almost 4 weeks of listening to various families of speakers were from the Klipsch family.


    R/L Fronts: RF-82 II Reference Series Floorstanding
    Center: RC-52 II Reference Series
    R/L Surround: RB-51 II Reference Series Bookshelf
    Sub: SW 110 Reference Series 10" 200 Watt


    These are awfully nice speakers, not top of the line by any means, but certainly a premium series of loudspeakers for a pretty average guy building a home theatre. Eventually I will expand from 5.1 to 7.2. But for starters (and so I don't wind up divorced) 5.1 is where I will be starting. Although I have found both a Denon and a Yamaha AVR I really like for powering the system, I will be holding off on making a final AVR selection until after HDMI 2.0 specs are finalized and some models start shipping with the 2.0 connections. I understand that there is no such thing as future-proof, but I'd prefer not to be kicking myself over something like a standard connector 2 months after I make my purchase.


    I'm looking at a budget of ~$3000 + the cost of acquiring some tools as needed.


    So, I guess my first question is, am I fooling myself thinking that I am going to be able to meet the sound quality of those Klipsch speakers going the the DIY route with my current budget? I'm not looking to necessarily entirely re-create those specific speakers, just that level of performance in a HT setting.


    Second, if I can indeed go the DIY speaker route for that budget, where is the best place for me to begin? Building the cabinets? Building the actual speaker? And how do I know what sort of performance levels I can expect from the various parts I acquire? If speakers are anything like the rest of the HT world, manufacturer specs are all but worthless (not that I can necessarily tell good specs from bad outside of some very obvious cases).
     
  2. I think you came to the right spot. This forum can be a bit slow, but a lot of the folk that drop in here are very knowledgeable. Most of us hang out probably a bit more often at multiple places like Partsexpress Techtalk, AVS, DIYAudio, and other audio forums, but we still come here now and then because we're big fans of pretty much everything Erich does for the community. So even if this isn't the fastest moving forum, you'll get a lot of great info here if you're patient.


    I think one thing you have to realize is that the cheaper Klipsch lines are usually voiced to have "Boom n Tizz". They probably still fall under a +/- 3db window, but audibly tend to have thick bass and a treble emphasis. The subsequent midrange shelving tends not to appeal to a lot of experienced listeners, so I need you to understand that if you go the DIY route, you probably won't find many existing designs that follow that design philosophy. In fact even high-end Klipsch stuff doesn't necessarily follow that philosophy. In television terms, it's kind of like selling you something permanently set on so-called "torch mode". It sells units with instant appeal, but isn't the most accurate or long-term pleasant sound.

    I think that's a great choice. FWIW, I'm partial to Marantz gear gear. While it's a risk/reward thing, I've had a good experience with refurb Marantz gear off of Accessories4less.com

    That's a fine budget. I recommend looking at the Plastic-SEOS based kits or other established designs right here at DIYSG. The SEOS-10 kits will do great as surrounds, and the larger stuff will be exceptional not only for home theater but also for music listening. The hard part is picking one, because they all do different things great. The quality of drivers we're talking about, for what it's worth, can be comparable to commercial offerings ranging as high as 5 to 10k per pair for the pricier stuff. The SEOS is really an exceptional waveguide and is pretty much the core of what most of us come here for :)

    Of course, there's also Bill Waslo's CoSyne design. It's not a kit, but all the info to construct it is here, and it's based on a DIY take of the extremely unique and exceptional Danley Synergy Horn concept.

    That's speakers of course. Subs are a whole nother monster and you should probably expect as much as half your DIY budget to go into them. I highly recommend multiple subwoofers (minimum two, preferably three or four) and you'll need to buy amplification for them. Each alignment has its pros and cons and will probably be dictated by price and available floorspace, but generally I don't consider anything less than a 15" long throw driver to be a subwoofer. One of my favorite drivers is the Dayton RSS18HO:

    http://www.parts-express.com/dayton-audio-rss460ho-4-18-reference-ho-subwoofer-4-ohm--295-472

    A pair of these powered off of a single Crown XLS2500 and you're probably set.

    You can by-far exceed that sound quality with your budget. The hard part is replicating that sound character with a kit, because it's just not something skilled designers normally shoot for.

    Building the cabinets (to spec) + assembling the crossover (internal circuit) are both very do-able. What's probably outside your capability is actually designing the loudspeaker.

    A few basic tips:

    - if your HT is wired for 20a circuits, that's awesome. Though it's certainly not the end-all-be all. More importantly, if you're going to be doing anything with the HT wiring, make sure you have good proper common grounding.
    - most designs will call for flush mounting, so your router will be your best friend.
    - although MDF is great, it can be heavy. Some of us prefer low-void ply simply for the weight savings. 3/4" is usually sufficient.
    - PL Premium is great. I've had less positive experiences with other glue (Gorilla / Carpenter's Glue / PL 9000) personally
    - Dado everything when woodworking. Otherwise you'll learn as I did and others do that butt joints can be hellish to deal with.
    - try to get a brace covering every five inches of diameter.
    - Even if you don't feel comfortable soldering, these days quick disconnects are a very good option.- 14 awg electrical wire is usually overkill, and it's pretty cheap. Don't spend money on overpriced "speaker cable" unless the techflex appeals to you. For a very high power application (i.e. subwoofer with 2000 watts going into it) I do use 12 awg however.








    Well, it's pretty complicated, and I could write you a short essay. In general, (anechoic) frequency response is the single most dominant measurement of loudspeaker design. You have to be able to see a graph, and it needs sufficient resolution to highlight issues. A spec sheet will rarely have this information. For an idea of how this information SHOULD be presented, please go check Bamberg Engineering Sound Labs' website and download the PDF versions of their spec sheets. THAT is a good "spec sheet".


    Other elements are also significant - their off-axis response (and the nature of it can vary significantly), efficiency, impedance, and power compression are the other ones I generally analyze. I don't mind elaborating at all, but perhaps first I can point you to this:

    http://www.audioholics.com/loudspeaker-design/loudspeaker-measurement-standard

    Gene worked with people like Dr. Floyd Toole in establishing their standard, and the article is pretty detailed.

    Sorry if my responses were a bit brief. I'll be here to expand upon stuff but I didn't want to start getting into pages of prose probably far beyond your interest / current understanding.
     
  3. We know how much budget you have, but the real kicker will be how loud you plan on listening, and how low you want to replicate the bass found in films. That will make all the difference.


    Is your $3k for everything (receiver, speakers, subs, amp for subs)? If so , choices will get narrrower quickly.


    JSS
     
  4. Since HDMI 2.0 and HDCP 2.2 are still in the finalizing stages and not readily available, I'll be waiting until probably the middle of next year to upgrade my AVR. I may even wait until the holiday season next year if the rumors of 4K Blu-ray are true, just to make sure the receiver I wind up using is able to handle the newest standards. I don't foresee myself investing in more than the smallest handful of titles, but I would rather have the option to if I so decided.


    So, that means $3K for the speakers. And, as others here and at AVS have pointed out, since subs will be a large portion of that, and possibly even be some of the biggest difference makers, I think I am going to start with building the subs first. I was going to start with just a 5.1 system. But, since the ultimate goal is a 7.2 system, and enough people that know way more about audio than I do have all recommended a minimum of 2 subs, I'm going to start with that. Then, once I have the 2 subs built, I'll know what my budget is for the other 5 speakers.


    Other concerns brought up, yes the HT will be wired for 20 amp circuits.
    I learned my lesson with speaker cable 20+ years ago. Monoprice and I are now close friends. I was expecting to use 12 or 14 AWG for this project, though that's more of an expense to go with the AVR than the initial speaker builds.


    I guess that means it's time for me to start learning about subwoofers, as I expect it would help to know a bit more than merely the difference between ported and sealed.
     
  5. Something you should note is that two subwoofers does not necessarily mean ".2" - we usually sum the bass into mono so that the subs behave as one sub, except spread throughout the room so that room interactions randomly average into a more clean response. Equalization is often a necessary part of this equation too.


    Ask away!

    FWIW, the RSS460HO I mentioned earlier in this thread was a thought-out suggestion. It has attributes that allow a single high-power amp to drive it to its impressive limits, and those limits ought to be pretty satisfactory unless you're dying for infrasonics. Even if you wanna do the infrasonic thing, LLTs are a great option dollar for dollar, even though they're unpopular these days.
     
  6. Compared to the "normal" person, I am exceptionally sensitive to low frequency sounds. I'm one of those people, that while watching television at a fair volume, still is aware of cars pulling into the house next door or driving along the street, and that's through several walls. I don't hear down into the super-low frequencies, but I do hear much lower than most of my friends and family. So for me, lows are more important than volume. Getting down towards 12 Hz would be my preference. If getting down into single digits where the sound is felt rather than heard is possible, then that would be the cherry on top of a particularly tasty sundae. I'm not looking to break the bank just to get down there though. If I could actually get down to 12 Hz, I'd be quite happy for until such time that the entire theatre is done and I get the itch for an upgrade somewhere.


    If the sound is still quite clear and defined, even at the lower end, I might even consider going up a bit from that point, but not by too much.
     
  7. Okay, 12hz is really low. It's nearly an octave below our hearing.

    There are two ways to get that low:


    A) A massive enclosure (Large And Low Tuned (LLT), Tapped Horn, Front Loaded Horn, Tapered Transmission Line)
    B) A massive budget with lots of displacement and lots of watts, but (relatively) small enclosures. You probably need four long-throw eighteens, and doubling that wouldn't be out of the question.


    The drawback of (A) is that it really be hard to place and hard to handle. It can also be a diffraction issue for your main speakers if they're too close to it.

    Now Large and Low Tuned is essentially an Extended Bass Shelf ported alignment. The difference is that you tune the box low enough that tones below its vent-tuning frequency will be so low in magnitude that they cannot bottom the driver - and thus only true large subwoofers with the right box and right parameters qualify as LLTs - any driver can be an EBS otoh. To get adequate Low Tune while ensuring the vent resonance frequency is well above (preferably an ~octave above) the crossover, it needs to of course be.. Large.

    Using the aformentioned RSS460HO as an example, you could do 10 cu ft box (+ ~1 cu ft for the 48" long vent itself) for a 13hz tuning (this will get you to 12hz as your anechoic -10db point (an almost picture perfect rolloff for a home theater room that will probably end up very close to flat IRL), although below that frequency xmax may be exceeded (though it will depend on source material). However, assuming this driver has an xmech at about 30 mm, there IS this ever-so-slight chance of an 8hz LFE tone bottoming the driver, but it's important to remember most electronics are rolling off at around this frequency as well. You could feed such a speaker about 300 watts @ 4 ohm in the deep bass safely, which gives you a max output of around 114db per speaker @ 60hz, and about 108db @ 20hz. With two, you can add roughly three to six db to that number. This is... really damn powerful to say the least. Can you visualize two of such a box in your HT area?


    Now the sealed box for this driver is smaller - 3.5 cu ft which will net you a qts = 0.78 but power handling is much higher - you could easily give it 550 watts and realistically have no problem giving it its full rated 900 watt input. You won't get 12hz infrasonics out of this box without needing to buy many more drivers and EQing them all, but you'll still get deep bass extension (F10 = 23hz) and very high maximum output (as much as 119db @ 60hz PER driver!). See the sim below,


    Sealed with 900 watts input is in Magenta. Values below 40hz exceed xmax and should not be considered as the driver motor will not produce these numbers
    Sealed with 550 watts input is in orange. This is a more realistic estimate for values below 40hz
    LLT with 300 watts input is in light blue. Output is accurate down to about 11hz.


    That's just one particular driver of course (one with a versatile qts, high efficiency, low Fs), but I hope that puts things into perspective with respect to infrasonics VS box size VS maximum output in the midbass - and by extension sealed vs ported. Personally, I think the Dual LLT would be a very cool build that will shake the foundation of your home at relatively low cost, but practical, it is not.


    [​IMG]
     
  8. So 12 is probably not going to happen for me. Although I could potentially see 2 units of the LLT size you mentioned, the placement options would be very limited to say the least. 11 cu. ft. sounds very manageable and not terribly "large" until I keep reading and see a 48" long vent needs to also be included somehow. That changes the math in a hurry. The theatre is still undergoing draft revisions. SWMBO would not be thrilled with a pair of subs that size residing in the common room until such time as the theatre is done.


    I'll put together some new numbers and post my questions in the morning after I give some thought to putting together something a bit more "reasonable".
     
  9. Just to clarify, the ported box I designed was as follows


    10 Cubic Feet box volume
    +
    48" long x 11" x 3" slot port

    It was just a rough model I threw together to get some perspective in for ya.

    Now you can easily wrap a 48" long port around the box. This should give you an idea of what we're talking about


    [​IMG]

    Get it?
     
  10. Yeah, I get it now. For some reason my mind was stuck in a very linear head space for a bit. Your diagram makes much more sense than what I was sketching in my mind. ::)
     
  11. After careful consideration, and also after consulting others that live in the house about what will be acceptable, it looks like LLT might not be a terrible solution after all. Sure, it's going to cut down on placement a bit, but since any sub I have is going to either be going behind the seating or near the walls, the only limitation is really building something that physically fits in the designated space. Even these sorts of LLTs would fit that requirement.


    It also seems to me, and I could be wrong, that LLTs allow for slightly better performance for the price, so long as one is willing to live with the fact that the units will be massive, and quite heavy. However, these units will likely only be moved twice. The first move will be from the shop area to the common room. The second move will be from the common room to the home theatre. That second move will be done by professional movers, strapping young college lads I am paying to throw out their backs for me. So I'm honestly not too worried about that part.


    I guess that means it is time to design a pair of subs. Then, after those are done, I'll know how much I have left over for the 5 speakers.
     
  12. [SIZE=small]


    Alright. Here's some drivers I like in LLT:

    Dayton RSS460HO (The one I've already mentioned in this thread. xmax isn't as high as the next couple offerings, but it's got a fine motor design and very nice parameters and sensitivity in my opinion. Also happens to be the least expensive driver I'm willing to use as a sub :p )

    Mach V IXL-18 (at more cost, will get you a bit more rated excursion, although I can't comment on whether I think it's worth it over the Dayton)
    [/SIZE]

    Dayton Ultimax 18 (a bit more excursion and power handling than the RSS460, although like the IXL, I'm not sure the benefits are worthwhile. Plus, qts is kind of high so I feel it's a slightly better candidate for a tapered TL instead of a mass-loaded port - still falls under the "LLT" umbrella :p)

    JBL W15GTI (Like the Ultimax, qts is a bit highish, so a tapered TL is going to model nicer. Very graceful overload behavior and thermal power handling means it's a great option if you don't want to accidentally destroy a driver while still getting top-tier sound quality due to its low induction.)

    Mach V UXL-18 (Sweet sub driver a great motor. To me, the first real logical step up from the RSS460HO. Inductance is high and qts is low, so you will want to EQ the natural response a bit to get things flat though...
    Mach V FTW-LLE (brand new driver, we don't have parameters out for it so I don't actually know if it's well-suited. Waiting on more details)

    Mach V FTW-21 (A ****ing monster. The aformentioned UXL-18's motor is probably too powerful for an 18 (hence its low qts) so they just put a bigger, heavier cone to compensate. That means much more swept volume. As a 21" driver though, we're getting into BIGGER box size territory, not smaller!)

    Mach V Phallus (the "dream" driver I want to buy like, six of :D - pretty much the finest bass driver I can think of)

    TC Sounds LMS-5400 (also one of the finest bass drivers out you can get)
     
  13. As much as it would be nice to build some cabinets for a pair of Mach V Phallus drivers, those drivers would blow my entire budget so far out of the water that I'd be looking for money to finish just the subwoofer portion of the project before I was done. I'm currently pricing things out, and I might be able to squeeze into Mach V UXL-18s, but I'm pretty sure the drivers will wind up being RSS460HOs.


    I'm looking at Behringer NU1000 or NU3000 for the amplifiers.


    I've bee looking at different approaches/plans for a tapped horn subwoofer. I haven't settled on an internal design yet, but as this is my first go, I'm likely to lean towards simpler approaches. I'd rather get things right the first time and then simply upgrade later because I "feel like it" rather than get complicated from the beginning and find I have to start over because I messed up along the way.
     
  14. Although the UXL can do more due to its higher xmax, you will need to feed it a lot more power to actually do so. That 300 watts might become 2000 quickly. On the flip-side, careful eith too much power into the RSS460HO as you can over-drive it in these large boxes.


    I know it's more expensive, but fwiw i would get the Crown XLS1000 over the behringer unit. If for no other reason than warranty (1 year VS 3 years).
     
  15. Well, I did some more in-depth price-planning, and it looks like I would be best off sticking with the RSS460HO. It's quite possible that I cold go closer to $1000 each on the subs and honestly never have to worry about building another sub again, but if I did that, then I have no idea how long it would be before I was able to move on from my HK SAT-TS1s from circa 2000 that I currently have as my mains and sides.


    So, I'm sticking with the RSS460HO, and now I get to figure out the size. As you pointed out, I do not want to over-drive these things. At the same time, I want this large box to still be able to fit the aesthetics of the room. That means they are likely going to be on the large(ish) side. However, that would put them well beyond the 10 cu.ft. mark. So, now I'm measuring to see what external sizes will work based on different locations in the room and trying to find a size that best-accommodates as many placements as possible.


    So far, it is looking like a cube with external dimensions of 30" x 30" x 30" will work best. But then that changes the math significantly.


    Another possibility, one that is sort of a long-shot and I'm not sure we can make work, but it is still a thought, is to go even larger, but to put both subs in one cabinet - something like 36" x 36" x 72".


    This is turning out to be far more complicated (but also far more fun) than I had originally anticipated.
     
  16. We don't want both drivers in the same box. That defeats the purpose of multiple subs - which is to spread the bass sources out so that they excite room modes differently!

    I can definitely sim you a bigger box. 10 cu ft was my "small practical box" and honestly actually just misses true LLT status. As I mentioned earlier, it was more of a "for instance" example than a "this is the best option" design.

    One thing I need to make clear is that as box size goes up, you really really can't get away with too much power. So even though the driver is rated for 900 watts, I would suggest between 200 and 350 watts @ 8 ohm.


    The box I end up with is 16 cu ft plus a 38" long vent with a cross-sectional area of 33 square-inches. 3" x 11" would net you that cross section, as would a 6.5" round vent.

    I would prefer to design such a large vented box as a mass-loaded transmission line, because you do start getting quarter-wave cancellations etc with such dimensions. Give me some rough dimensions of something 16 cu ft that will work for you - and I'll model them properly for you. One suggestion I've got is to factor in "what width of box will squeeze through my door?"
     
  17. I'll get you some dimensions in the morning.


    Thankfully, the question of fitting through the door is not one I have to address really. The work space has dual sliding doors that lead into the house/viewing area. As long as I don't go over 7' across in all directions, I'll be fine.
     
  18. Actually, after reading Josh Ricci's review of the RSS460HO, I realized this driver has more usable throw than I thought (he suggests as much as 20mm usable). I don't think 18 cu ft, a 7" diameter x 43" vent, and 400 watts is out of the question at all if you want to maximize the system potential. With that, excursion still won't exceed 18mm until below 10hz, and vent air velocity stays below 17m/s until below 12hz. Then the kind of output we're talking for a single sub is something like:


    107db @ 12hz
    110db @ 24hz
    > 114db @ 48hz and higher


    and with two, you get the room-smoothing effect.
     
  19. I have been taking what you said under advisement. We did some more math. We did some more planning. We are still going to be going with the RSS460HO. Also, despite ZERO plans to move, it did make sense to at least consider the logistics of what would happen if the sub did indeed have to be moved for some reason. As a result, we have settled on a pair of subs that are between 30"- 32" in width. These are large(ish) but still do get through most any household door, and certainly through any door we might possibly ever be asked to move through.


    While the idea for external aesthetic purposes was to create these as cubes (which would later closely resemble Borg cubes in the finished room) there is not a clear imperative that they must be cubes. If I am choosing whereon that 30-32 scale I would rather fall, I have no issues against going to the 32 side of the scale if a better internal space can be created with that.


    By the way, I appreciate all the help you've been tossing my way. I've been learning a lot from reading up on the things you have mentioned along the way.
     
  20. You're already getting good help here, but just wanted to mention I listened to a pair of these on the weekend at a local store. Pretty sure they were them. About $800 each iirc. That was on sale? I can't remember the details. Anyways, the SQ they provided will be easily beat by DIY speakers. So ya, get 'er done. :)
     
  21. I'm actually starting to get very stoked for this project. The one part intimidating me a bit is learning to use a router to the best effect, but I'll get it down eventually. Well, actually, the router and properly assembling the crossovers. The difference between the two for me though is that I have no help on the first front, while I happen to know a few blokes that either were, or have been involved in building electronics, and I can call them at any moment to get a hand or to at least make sure I am going about things in the right manner.


    As GranteedEV pointed out, my biggest stumbling block looks like it is going to be properly designing my subs to get the right sort of performance out of them. These aren't like those "cheap" sub builds for the back of a car where the only consideration is just finding a box with the right sized cutout on the front and then dropping a 15" driver into the hole and pulling the leads out the back.
     
  22. I just received some interesting news, and it seems that there now exists the small possibility that I might be able to upgrade the quality of driver I go with. Will this change the actual build of the sub? Or will this simply provide for a greater range and cleaner sound? It's still a bit unclear, but I may be able to go upwards to around $750 for each of the 2 drivers now. If that is going to significantly improve my build, then I am all for it. But just because I can go that high, does not mean I necessarily want to go that high if I don't have to, or if the trade-off is not going to really be there.
     
  23. I think you'd be better off spending the extra money on more subs rather than presumably better ones. Make it four instead of two.

    Or just redirect the extra funds to the main speakers.
     
  24. I do like those RSS460HOs. I asked the question because I was considering those FTW-21s you posted earlier given the maximum allowed size of the cabinet. Sounds like maybe I'll build two to begin with to get a better idea of my exact finished cost in both dollars and labour and then go from there. Then, if I can swing it with the extra funds, I'll consider moving up to 4 subs. Although 4 subs in 32" cabinets will make for some interesting placement. Of course, they don't have to be 32" I suppose, but I since the idea is to get optimum performance out of the driver, I'm not anticipating going too much smaller.
     

Share This Page