After doing some pro audio in college for a while, I've casually been picking up speaker building as a hobby. I've never done a 3-way, but I'm interested in making a set. Here are the requirements:
Requirements:
- (relatively) small, light, and portable with the though being to take them to various small outdoor activities (~25 people max)
- f3 of at least 38Hz outdoors
- 'flat' response up to ~90dB for a small patio party/private outdoor movie setting
- Would like to keep this under $500/pair, if I can, but slightly flexible
Currently, I'm planning to do a WWMT setup and am focusing on the woofers to start with. I considered going with a single 10 or 12, but that just seems too.....mainstream I guess? Anyone can go buy a floor 10 or 12 that hits 38Hz, but it's much more awesomer hitting 38Hz out of a 'small' tower with small drivers, so I've generally settled on dual 6.5 drivers even though a 10 or 12 would be cheaper. I liked the sensitivity, power handling, and look of the Dayton DCS165's, but I just couldn't model an f3 below 42Hz out of them, so they're out. Next in line is the Tang Band W6-1139SIF's. I ran a pair of the 5.25 TB's as my computer sub for a while and was quite pleased with their performance for their size. Logic has it that more cone area, xmax, and double the drivers could only be significantly better. Modeling them up in WinISD, I can stuff 2 of them in a 0.8 ft^3 box with 2x 2"ID ports at 15.76" long, which produces a tuning of 35Hz and keeps the port velocities down to 24m/s without running out of excursion. F3 is modeled at 35Hz with SPL at 105dB before the roll off. Throw in a second speaker and that'll boost output by a couple dB, so sims suggest ~108-110dB to roll off, and ~106dB at 35Hz, without running out of excursion (sims at 1m).
I haven't really settled on the mids/tops much as I realize my woofers will be the key component in this build, but I've thought to toss in an AMT just for fun. I realized there was a speaker that already uses the TB 6.5's and a Dayton AMT up top. I may be generally basing my MT setup, and crossover, on Scott's Zirconium project, but that has yet to be decided.
Just wanted to toss my thoughts up here before diving into it in case something looks not quite right. I'm not interested in nitpicking and super high details at this point, but am mostly just checking to see if my goals for my first 3-way seem reasonable, or if there's a big hole in there somewhere before I dive into it too much.
Requirements:
- (relatively) small, light, and portable with the though being to take them to various small outdoor activities (~25 people max)
- f3 of at least 38Hz outdoors
- 'flat' response up to ~90dB for a small patio party/private outdoor movie setting
- Would like to keep this under $500/pair, if I can, but slightly flexible
Currently, I'm planning to do a WWMT setup and am focusing on the woofers to start with. I considered going with a single 10 or 12, but that just seems too.....mainstream I guess? Anyone can go buy a floor 10 or 12 that hits 38Hz, but it's much more awesomer hitting 38Hz out of a 'small' tower with small drivers, so I've generally settled on dual 6.5 drivers even though a 10 or 12 would be cheaper. I liked the sensitivity, power handling, and look of the Dayton DCS165's, but I just couldn't model an f3 below 42Hz out of them, so they're out. Next in line is the Tang Band W6-1139SIF's. I ran a pair of the 5.25 TB's as my computer sub for a while and was quite pleased with their performance for their size. Logic has it that more cone area, xmax, and double the drivers could only be significantly better. Modeling them up in WinISD, I can stuff 2 of them in a 0.8 ft^3 box with 2x 2"ID ports at 15.76" long, which produces a tuning of 35Hz and keeps the port velocities down to 24m/s without running out of excursion. F3 is modeled at 35Hz with SPL at 105dB before the roll off. Throw in a second speaker and that'll boost output by a couple dB, so sims suggest ~108-110dB to roll off, and ~106dB at 35Hz, without running out of excursion (sims at 1m).
I haven't really settled on the mids/tops much as I realize my woofers will be the key component in this build, but I've thought to toss in an AMT just for fun. I realized there was a speaker that already uses the TB 6.5's and a Dayton AMT up top. I may be generally basing my MT setup, and crossover, on Scott's Zirconium project, but that has yet to be decided.
Just wanted to toss my thoughts up here before diving into it in case something looks not quite right. I'm not interested in nitpicking and super high details at this point, but am mostly just checking to see if my goals for my first 3-way seem reasonable, or if there's a big hole in there somewhere before I dive into it too much.
Keep in mind F3 is only meaningful (Toole) in the filter sense, since you aren’t XOing to anything lower F6 or F10 are much more meaningful.
If it is a tower, box dimension ratios will mostly likely make it an ML-TL so your bass reflex sims are not telling you what is happening. You should be able to go lower with a properly tuned ML-TL.
dave
Your ProSound work should help you accept an EconoWave design is best for the indoor/outdoor mix.
--Waveguide to control directivity to the small group.
--Light weight NdFeB magnet on Eminence 12" woofer
--Two good drivers fit your $500/pair budget
diysoundgroup has proven designs with CNC cut flat-pack kits.
Fusion-12 DIY Sound Group
HT-12 DIY Sound Group
-----------
Dayton DS315 econowave
SEOS12 with Designer12 woofer | Page 4 | HiFiCircuit
Parts Express has DS315, a low cost 1" Dayton compression driver and $10 Dayton SEOS12 or PVR waveguide.
Parts Express also has a flat pack plywood cabinet for a 12" woofer + room for SEOS12.
--Waveguide to control directivity to the small group.
--Light weight NdFeB magnet on Eminence 12" woofer
--Two good drivers fit your $500/pair budget
diysoundgroup has proven designs with CNC cut flat-pack kits.
Fusion-12 DIY Sound Group
HT-12 DIY Sound Group
-----------
Dayton DS315 econowave
SEOS12 with Designer12 woofer | Page 4 | HiFiCircuit
Parts Express has DS315, a low cost 1" Dayton compression driver and $10 Dayton SEOS12 or PVR waveguide.
Parts Express also has a flat pack plywood cabinet for a 12" woofer + room for SEOS12.
Last edited:
I built a 10" sub out of a sheet of 3/4" MDF and man, that sucker is HEAVY! It's birch plywood for me these days if I can help it (might even rebuild the box if I'm feeling spunky). That and I'm not really interested in doing another kit speaker either. I'm much more interested in building a good sounding speaker that nobody else has made and being able to point to it and say "I designed and built that". That and it's fun building stuff after flying a desk all day.
I do have a CNC machine, so precision and quality control is as good as the monkey running it.
The more I think about it, the more it's making sense to go with a TL. I'll have to do some research and maths on it in the next few days. I just figured I'd do a port since a lot of the big box places did ported. That why I posted up here though, is for pro tips like that before I dove into it.
I do have a CNC machine, so precision and quality control is as good as the monkey running it.
The more I think about it, the more it's making sense to go with a TL. I'll have to do some research and maths on it in the next few days. I just figured I'd do a port since a lot of the big box places did ported. That why I posted up here though, is for pro tips like that before I dove into it.
Last edited:
Field tested TL ( actually, MLTLs - "mass loaded") or similar topology designs abound hereabouts for which all that pesky math as already been done. The costs of entry and learning curve for much of the measuring and modeling systems now available to the enthusiast aren't as much of an obstacle as in the past, but there's still no substitute for the experience of many years of builds that didn't quite work out exactly as anticipated - sometimes surprisingly better so.
Yes, a CNC can be a wonderful tool - very effective and repetitive - but the design needs to be right first.
Yes, a CNC can be a wonderful tool - very effective and repetitive - but the design needs to be right first.
No, there's a lot of daylight or "new math" separating "classic" / old school TL theory and the more recent designs thanks to the very generous work of folks like Martin King.
Spending a few days down this rabbit-hole http://www.quarter-wave.com is highly recommended reading. My own personal take away was - let the real math geeks (Scott Lindgren, Bob Brines, Dave D etc) go to town with this - just give me a set of accurately dimensioned drawings.
Spending a few days down this rabbit-hole http://www.quarter-wave.com is highly recommended reading. My own personal take away was - let the real math geeks (Scott Lindgren, Bob Brines, Dave D etc) go to town with this - just give me a set of accurately dimensioned drawings.
King & Augspurger’s TL modelers (perhaps more accurately called quarter-wave modelers) showed that the TL space was much larger than we imagined. And that traditional classic TL design was woefully inadequate. A traditional TL is essentially a long line that gets smaller as it approaches the terminus. A TL that expands has in the past been known as a Voigt (or TQWT/TQWP tapered quarter wave tube/pipe). In between are straight pipes with no taper. Mass loading a line means adding a restricted terminus — often in the form of a vent. Turns out tapered TLs are usually best if they have a significant taper ratio, and the Voigts really need to be mass-loaded. A straight pipe can be much shorter if it is mass loaded. Such an ML-TL often looks like a tall skinny bass reflex but isn’t — the way the air works inside is quite different.
This is a ML-Voigt but it gets the idea across.
Horns also turn out to be in TL-space but are further out along the axis as Voigts with more exotic taper and other details.
Something else that modern TL modelers showed us is the advantages of having the driver offset from the end of the pipe.
dave
Ok, cool. It's starting to make sense a little bit after that explanation, and some research. I was picturing a slight modification on a "regular" T line with all of the folds to calculated by the target wavelength...length, and stuff like that. Sounds like the general air behaves similar, but the construction is significantly different with no folds inside (presuming the box size ratios are correct anyways).
It looks like the AviaTrix speakers use an MLTL design. This is essentially what an MLTL's innards would look like, I presume.
It looks like the AviaTrix speakers use an MLTL design. This is essentially what an MLTL's innards would look like, I presume.
Yes, you can see more examples here: Frugal-phile | Box Library / Mark Audio, near the bottom there is a Pensil section, all ML-TLs, and below that a section with random ML-TLs.
dave
dave
Some thoughts...
- 38Hz and loud outdoors is difficult. This doesn't need to be loud, so I'm confident we can get something good.
- Econowaves won't go nearly low enough without some kind of EBS alignment, which will end up huge for PA midbass drivers.
- The Tang Band mini-subs need an absolute rocket up them to get going. At demo levels, I had the -6dB light on an NU6000 lit pretty often. That's over 500w peaks for each driver. Their free-space (outdoors on a stand) sensitivity is below 80dB@1w.
If it was mine, I'd consider simplifying and doing an 8" 2-way standmount design. If you only want 90dB, it'll do that just fine, even outdoors. If you wanted an easy life, you could pick up a pair of Behringer B2031A (check the reviews) and be done with it. They'll hit 40Hz and have enough adjustments built-in to make them sound good outdoors.
If you don't fancy just buying some speakers, I'd still recommend the form-factor. That sort of thing is much easier to move around than a tower, and will still do everything you want it to do.
Chris
- 38Hz and loud outdoors is difficult. This doesn't need to be loud, so I'm confident we can get something good.
- Econowaves won't go nearly low enough without some kind of EBS alignment, which will end up huge for PA midbass drivers.
- The Tang Band mini-subs need an absolute rocket up them to get going. At demo levels, I had the -6dB light on an NU6000 lit pretty often. That's over 500w peaks for each driver. Their free-space (outdoors on a stand) sensitivity is below 80dB@1w.
If it was mine, I'd consider simplifying and doing an 8" 2-way standmount design. If you only want 90dB, it'll do that just fine, even outdoors. If you wanted an easy life, you could pick up a pair of Behringer B2031A (check the reviews) and be done with it. They'll hit 40Hz and have enough adjustments built-in to make them sound good outdoors.
If you don't fancy just buying some speakers, I'd still recommend the form-factor. That sort of thing is much easier to move around than a tower, and will still do everything you want it to do.
Chris
Well, the designs posted that Dave referenced have had all that icky math done already, and you could build probably at least one pair of speakers in the time you tinker around with getting the simulators to "work".
While I generally prefer the sonics of a floorstanding / tower system for home use, I'd have to agree with Chris on the advantages of smaller for portability. But don't forget our old friend mr Hoffman
pick any two:
- efficiency
- deep bass
- compact size
While I generally prefer the sonics of a floorstanding / tower system for home use, I'd have to agree with Chris on the advantages of smaller for portability. But don't forget our old friend mr Hoffman
pick any two:
- efficiency
- deep bass
- compact size
Yeah, I'm working on digesting some of the info there, but it'll take some time for sure. It seems that, like a BR box (or most any other type of box, really), a properly designed MLTL box revolves around the specific driver params and I can't realistically drop in different drivers than the box was designed for and expect good results. Therefore, if I want to use a specific driver, I have to do all the maths to match that driver or driver combo.
Yeah, gotta love Mr Hoffman. I've gone for numbers 2 and leaning more towards 3 with my initial thoughts of the TB 6.5's only being 83dB.
Yeah, gotta love Mr Hoffman. I've gone for numbers 2 and leaning more towards 3 with my initial thoughts of the TB 6.5's only being 83dB.
So, with my newfound knowledge of MLTL physics, I'm getting nervous that another build I've lazily been building won't sound like I modeled it up. I modeled it as a BR box MTM style and have the cabinet mostly built, but not complete.
Looks like this with 2x 6.5's and a tweeter:
Looks like this with 2x 6.5's and a tweeter:
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.