I posted a while ago asking for input on building a mini tower employing a TMM double-chamber reflex configuration. The DCR idea was quickly pointed out to have no practical advantage, and to be honest I hadn't really thought out why I wanted the TMM configuration... So after some more consideration and research I've come up with another design. Hopefully some of you reading can help me iron out some of the details.
My goals are great sound, pleasing appearance, and low cost. I also want the project to be fun and challenging.
My knowledge of speaker design is basic, but I think I'm learning. My experience so far is re-capping my BeoVox S45.2s, building stands for them, and helping my son build a pair of Frugel-Horn Mk 3s (Fostex FF125wk). I'm a pretty skilled woodworker with a fairly well-equipped shop so construction shouldn't be a problem.
The basic layout for the speakers will be a small 2-way with a separate subwoofer in the same enclosure. Separate (external) amps will drive the 2-way and the sub.
The woofer for the 2-way will be the Silver Flute WRC1425-04. The tweeter will be the NHT soft dome I won in the BA2015 raffle, crossed over at 2.2kHz. I'll use the .35cf (9.91L) vented box recommended by Madisound for an F3 of 60Hz.
For the sub I'm thinking the Tang Band W6-1139SIF, a little 6.5" paper cone sub that appears to offer a lot of bang for the buck. For this driver, the vented box calculator at mh-audio.nl recommends a volume of .45cf (12.86L) for an F3 of 33.65Hz. However, my enclosure design has enough space for a subwoofer chamber up to .93cf (26.33L) so if anyone has any other ideas for the use of this space I'm open to suggestions. Maybe getting a little more bass extension, or perhaps a bandpass chamber to eliminate the need for an active crossover? For aesthetic reasons, I'd like the subwoofer to be rear-firing.
Any suggestions for improving this speaker concept would be appreciated.
Anyway, click on the thumbnails below for renderings of my proposed speaker.
My goals are great sound, pleasing appearance, and low cost. I also want the project to be fun and challenging.
My knowledge of speaker design is basic, but I think I'm learning. My experience so far is re-capping my BeoVox S45.2s, building stands for them, and helping my son build a pair of Frugel-Horn Mk 3s (Fostex FF125wk). I'm a pretty skilled woodworker with a fairly well-equipped shop so construction shouldn't be a problem.
The basic layout for the speakers will be a small 2-way with a separate subwoofer in the same enclosure. Separate (external) amps will drive the 2-way and the sub.
The woofer for the 2-way will be the Silver Flute WRC1425-04. The tweeter will be the NHT soft dome I won in the BA2015 raffle, crossed over at 2.2kHz. I'll use the .35cf (9.91L) vented box recommended by Madisound for an F3 of 60Hz.
For the sub I'm thinking the Tang Band W6-1139SIF, a little 6.5" paper cone sub that appears to offer a lot of bang for the buck. For this driver, the vented box calculator at mh-audio.nl recommends a volume of .45cf (12.86L) for an F3 of 33.65Hz. However, my enclosure design has enough space for a subwoofer chamber up to .93cf (26.33L) so if anyone has any other ideas for the use of this space I'm open to suggestions. Maybe getting a little more bass extension, or perhaps a bandpass chamber to eliminate the need for an active crossover? For aesthetic reasons, I'd like the subwoofer to be rear-firing.
Any suggestions for improving this speaker concept would be appreciated.
Anyway, click on the thumbnails below for renderings of my proposed speaker.
Attachments
Please read here: http://www.diyaudio.com/forums/multi-way/189847-introduction-designing-crossovers-without-measurement.html
I find particular useful the post #20 and the linked methodology. If you want to design on your own and don't have measurement capabilities, that is the only method you can use, forget about crossover calculator as they assume perfect drivers (FR and impedance), which don't exist.
However you need to have manufacturer measurements for all the drivers you plan to use, or at least some reliable ones. Is this the case for the tweeter?
Another tricky aspect on the proposed drivers is the so called subwoofer. How will you drive it? How will you implement the crossover between the sub and the mid-woofer? The sub is fairly inefficient, and needs probably a separate plate amp and digital crossover.
Without the willingness to spend money in measurement equipment and/or several crossover parts for tests, you are far better in choosing an already designed speaker and concentrating only in the execution part.
Ralf
I find particular useful the post #20 and the linked methodology. If you want to design on your own and don't have measurement capabilities, that is the only method you can use, forget about crossover calculator as they assume perfect drivers (FR and impedance), which don't exist.
However you need to have manufacturer measurements for all the drivers you plan to use, or at least some reliable ones. Is this the case for the tweeter?
Another tricky aspect on the proposed drivers is the so called subwoofer. How will you drive it? How will you implement the crossover between the sub and the mid-woofer? The sub is fairly inefficient, and needs probably a separate plate amp and digital crossover.
Without the willingness to spend money in measurement equipment and/or several crossover parts for tests, you are far better in choosing an already designed speaker and concentrating only in the execution part.
Ralf
I think nobody wants a high cost, ugly and awfully sounding speaker. But low cost means nothing as it is fairly subjective.
ChuckORWC, or Chuck if I may, since it rolls off the tongue easier... 
You've got it all wrong right from the start. You've selected some cheapish drivers that are unsuitable at 4 ohms, unless you like blowing amplifiers up. I would hate to design a 3-way with 4 ohm drivers.
You start with a concept you like, THEN choose your drivers. Well, I'm a bit more experienced than you, so trust me. THIS one we can do something with:
https://www.madisoundspeakerstore.com/approx-5-woofers/silver-flute-w14rc25-08-ohm-5-1/2-wool-cone/
How about something like THIS:
Vifa PL14WJ-
The crossover is actually very generic, and will work with most 5" paper cones and almost any old tweeter above 88dB efficiency. My one with or without the 15uF capacitor does much the same thing. I don't want to get into 2.5 ways or 3 ways right now. OK?
You've got it all wrong right from the start. You've selected some cheapish drivers that are unsuitable at 4 ohms, unless you like blowing amplifiers up. I would hate to design a 3-way with 4 ohm drivers.
You start with a concept you like, THEN choose your drivers. Well, I'm a bit more experienced than you, so trust me. THIS one we can do something with:
https://www.madisoundspeakerstore.com/approx-5-woofers/silver-flute-w14rc25-08-ohm-5-1/2-wool-cone/
How about something like THIS:
Vifa PL14WJ-
The crossover is actually very generic, and will work with most 5" paper cones and almost any old tweeter above 88dB efficiency. My one with or without the 15uF capacitor does much the same thing. I don't want to get into 2.5 ways or 3 ways right now. OK?
Attachments
Thanks for the responses guys.
Ralf: Post #20 in the crossover design thread (which I've read, if not fully absorbed) is just a link to crossover design tools, so I don't see how it's particularly helpful. You're right, I don't have the tweeter specs. I do have the Audio Tools app and a cheap measuring mic, and once I figure out how to use them I should be able to get numbers close enough to be somewhat useful. As far as plotting the tweeter's impedance, I'm not sure if I'm equipped to do that... I suspect I'll have to make some assumptions and do some tweaking. Anyway, crossovers can be re-designed and the tweeters can be replaced if push comes to shove; at this point the concern is getting the box tuning correct as that will be harder to change after the enclosures are built.
With regards to the subwoofer (don't know why it would only be a "so-called" subwoofer), as I said in the original post it would be bi-amped with a separate amplifier. I haven't quite figured out how I'll handle the crossover between the 2-way and the sub yet, but I'm open to suggestions.
Of course low cost is subjective. I love DIY for its own sake, but I also do it partly out of necessity. One kid in college, one starting college in a year-and-a-half, two mortgages, etc., all add up. I enjoy the challenge of seeing how much I can achieve on a very limited budget.
Steve: Chuck's fine; it's my name, after all. I think it was already taken as a user name. I haven't heard that 4 ohm drivers blow amps up. When someone says such-and-such an amp will do X watts into 8 ohms and 2X watts into 4 ohms, they should actually say that 4 ohm speakers will destroy the amp? If it makes a difference, the plan is for these speakers to be bi-amped with an Amp Camp Amp driving the 2-way and a souped-up ACA driving the sub. Since the ACA is a relatively low powered amp, I thought it would make sense to use a low impedance, relatively high efficiency woofer, hence my choice of the 4 ohm Silver Flute.
Ralf: Post #20 in the crossover design thread (which I've read, if not fully absorbed) is just a link to crossover design tools, so I don't see how it's particularly helpful. You're right, I don't have the tweeter specs. I do have the Audio Tools app and a cheap measuring mic, and once I figure out how to use them I should be able to get numbers close enough to be somewhat useful. As far as plotting the tweeter's impedance, I'm not sure if I'm equipped to do that... I suspect I'll have to make some assumptions and do some tweaking. Anyway, crossovers can be re-designed and the tweeters can be replaced if push comes to shove; at this point the concern is getting the box tuning correct as that will be harder to change after the enclosures are built.
With regards to the subwoofer (don't know why it would only be a "so-called" subwoofer), as I said in the original post it would be bi-amped with a separate amplifier. I haven't quite figured out how I'll handle the crossover between the 2-way and the sub yet, but I'm open to suggestions.
Of course low cost is subjective. I love DIY for its own sake, but I also do it partly out of necessity. One kid in college, one starting college in a year-and-a-half, two mortgages, etc., all add up. I enjoy the challenge of seeing how much I can achieve on a very limited budget.
Steve: Chuck's fine; it's my name, after all. I think it was already taken as a user name
. I haven't heard that 4 ohm drivers blow amps up. When someone says such-and-such an amp will do X watts into 8 ohms and 2X watts into 4 ohms, they should actually say that 4 ohm speakers will destroy the amp? If it makes a difference, the plan is for these speakers to be bi-amped with an Amp Camp Amp driving the 2-way and a souped-up ACA driving the sub. Since the ACA is a relatively low powered amp, I thought it would make sense to use a low impedance, relatively high efficiency woofer, hence my choice of the 4 ohm Silver Flute.
TBH, most speakers aren't designed with specific amplification in mind. They're designed to work with your regular old 4-8 ohm solid state. This is mainly to avoid lawsuits.
Most 6 channel home-theatre amps aren't actually very good at 4 ohm load. It's at the extreme of their operating range.
Now you can do this sort of difficult carpentry with a rear or sideways facing bass. It's usually called a FAST speaker, and it crosses low, or it just doesn't image well.
So many variations. Maybe this is as good as it gets in three ways right now, state of the art IMO:
ScanSpeak-3W-Discovery
But Troels Gravesen has hundreds of projects:
DIY-Loudspeakers
If you like doing it all for yourself, I can recommend a good sim program:
Downloads.
And a few projects to load into the projekte folder and get running and modding in minutes flat:
boxsim-db.de | Boxsim Projektdatenbank
Whether you end up with a good and easy to drive speaker is down to your skill. And TBH, skill takes YEARS to accumulate. It's why the old Football Pros kinda laugh at the 1st. year Rookies.
And IMO it's worth remembering somebody or other's wise words: "Between the concept and the reality, lies the ABYSS."
Play it safe! Baby steps at first.
Most 6 channel home-theatre amps aren't actually very good at 4 ohm load. It's at the extreme of their operating range.
Now you can do this sort of difficult carpentry with a rear or sideways facing bass. It's usually called a FAST speaker, and it crosses low, or it just doesn't image well.
So many variations. Maybe this is as good as it gets in three ways right now, state of the art IMO:
ScanSpeak-3W-Discovery
But Troels Gravesen has hundreds of projects:
DIY-Loudspeakers
If you like doing it all for yourself, I can recommend a good sim program:
Downloads.
And a few projects to load into the projekte folder and get running and modding in minutes flat:
boxsim-db.de | Boxsim Projektdatenbank
Whether you end up with a good and easy to drive speaker is down to your skill. And TBH, skill takes YEARS to accumulate. It's why the old Football Pros kinda laugh at the 1st. year Rookies.
And IMO it's worth remembering somebody or other's wise words: "Between the concept and the reality, lies the ABYSS."
Play it safe! Baby steps at first.
Last edited:
It seems you didn't read carefully that post. It indicates that there's a nice guide on how to design a speaker without measurement at the bottom of the linked page. The guide is found here: http://audio.claub.net/Simple%20Loudspeaker%20Design.pdf
The crossover design tools are needed for simulations, useful even if you have measurement equipment.
No mention whatsoever for bi-amping in your post, so the question. The easiest way to cross the sub to the mid would be to use a digital crossover, but I'm not an expert on that aspect, and there's a subwoofers forum. But IMHO a 6" driver is hardly a sub-woofer (or is very inefficient or doesn't go low enough).
In general I think you are grossly underestimating the effort needed in designing this 3-way, and I think you are better in finding a proven design (*), and make it beautiful.
Ralf
(*) A proven design should have at least simulated and measured FR. Zaph Audio designs for example are in that league, as they provide also explanation of the whole design.
Chuck,
Our experience with the W14 has been good. Althou we use it as a woofer, not a mid-bass.
Given its use with a woofer, do you really need a vented enclosure? Sealed with Q = 0.707 (butter worth) is 3 litre with an F3 at 90 Hz.
We put it into a 5 litre vented alignment with good results (actually 2 in 10 litre). We are also working on a push-push pr in an ML-TL.
4 ohms won't help with the ACA, you'll just run out of current capability sooner, and althou this Class A amp will work with 4 ohms it will be happier with 8.
dave
Our experience with the W14 has been good. Althou we use it as a woofer, not a mid-bass.
Given its use with a woofer, do you really need a vented enclosure? Sealed with Q = 0.707 (butter worth) is 3 litre with an F3 at 90 Hz.
We put it into a 5 litre vented alignment with good results (actually 2 in 10 litre). We are also working on a push-push pr in an ML-TL.
4 ohms won't help with the ACA, you'll just run out of current capability sooner, and althou this Class A amp will work with 4 ohms it will be happier with 8.
dave
Dave, that is a 10L thing of beauty. Onken, of course.
But you're a bit short on detail. Fostex FF85 fullranger of course. 4 or 8 ohm Silver Flute? Series or parallel wired?
What's the crossover? Come on tell us, though I could work it out for myself, of course. And probably will. But I'm already guessing 800Hz second order.
You can keep the secret of the magic paint dots to yourself.
But you're a bit short on detail. Fostex FF85 fullranger of course. 4 or 8 ohm Silver Flute? Series or parallel wired?
What's the crossover? Come on tell us, though I could work it out for myself, of course. And probably will. But I'm already guessing 800Hz second order.
You can keep the secret of the magic paint dots to yourself.
FF85wKeN XOed to 2 SF W14 4 ohm in series. 5 litre per W14, a mid TL for the FF85wk. Works a treat with PLLXO or the digital XO in Chris' Onkyo HT receiver. Passive XO (1st order series) is a work in progress, but the 1st iteration got (unofficial) best of show at diyFEST 2 years ago. XO somewhere between 300-400 Hz.
dave
dave
so far the XO is passive line level = active system, and no, it's a good octave lower than 800Hz,
but for the life of me, I can't remember whether 4 or 8 - after a couple of hundred pairs, it's hard to recall them all, and I'm horrible at taking notes or construction photos
The floorstanding 2-way that Dave will likely be passive high level, once we replenish our supplies - much harder to come by than one would wish
OK, so he beat me to it, and answered most of your questions, I think
but for the life of me, I can't remember whether 4 or 8 - after a couple of hundred pairs, it's hard to recall them all, and I'm horrible at taking notes or construction photos
The floorstanding 2-way that Dave will likely be passive high level, once we replenish our supplies - much harder to come by than one would wish
OK, so he beat me to it
, and answered most of your questions, I think
There is nothing wrong with using 4 ohm drivers in a 3-way. You just have to watch your impedance more carefully.
The 4-8 ohm spec is due to current capacity and how amps operate, as well as thinner traces or wires, and smaller heatsinks. They are more easy to manufacture. Most pro-amps do 2 ohms stereo. Lots of car amps go much lower.
Later,
Wolf
The 4-8 ohm spec is due to current capacity and how amps operate, as well as thinner traces or wires, and smaller heatsinks. They are more easy to manufacture. Most pro-amps do 2 ohms stereo. Lots of car amps go much lower.
Later,
Wolf
Dave did say that the 4 ohm SFs were wired in series - which I'd forgot, and something we've done more than once before.
There are indeed some excellent sounding and fairly robust car amps - some rated as low as 1 ohm (scary thought that). I'd think the issue with them in a home system would be a convenient and compact power supply capable of delivering the dozens of amps at 12V that would be required to deliver their rated power.
There are indeed some excellent sounding and fairly robust car amps - some rated as low as 1 ohm (scary thought that). I'd think the issue with them in a home system would be a convenient and compact power supply capable of delivering the dozens of amps at 12V that would be required to deliver their rated power.
Chris and Dave, I thought I'd tell you something interesting about your splendid Onken.
I just ran it up the flagpole with Cousin Billy's novel circuit, if you've been following the thread. The red 12 ohm is the magic mixing resistor.
I used any old 4 ohm paper woofers and some typical fullranger in a closed box about the right size. Ignore the bump around 10kHz. it's just what fullrangers do.
That looks good to me! Time-aligned it gets rather LR2, but flat baffle is a rather splendid BW3 type 90 degrees. Never below 6 ohms, either, not that wolf_teeth would care, but I do.
I just ran it up the flagpole with Cousin Billy's novel circuit, if you've been following the thread. The red 12 ohm is the magic mixing resistor.
I used any old 4 ohm paper woofers and some typical fullranger in a closed box about the right size. Ignore the bump around 10kHz. it's just what fullrangers do.
That looks good to me! Time-aligned it gets rather LR2, but flat baffle is a rather splendid BW3 type 90 degrees. Never below 6 ohms, either, not that wolf_teeth would care, but I do.
Attachments
Had a bit more of a play with your Onken MTM, Dave and Chris.
15cm wide, 60 cm high. Reflex neither here nor there in the modelling IMO. The MTTM version is almost unchanged. It's a 1dB louder at the top without adjustment.
I think a smaller bass coil is better. The bigger 15R mix resistor takes the midrange at 1.5kHz down a tidge. But this ought to be worth a try if you're feeling adventurous. Incredibly unfussy circuit.
15cm wide, 60 cm high. Reflex neither here nor there in the modelling IMO. The MTTM version is almost unchanged. It's a 1dB louder at the top without adjustment.
I think a smaller bass coil is better. The bigger 15R mix resistor takes the midrange at 1.5kHz down a tidge. But this ought to be worth a try if you're feeling adventurous. Incredibly unfussy circuit.
Attachments
No point in crossing lower than 2kHz on a 5" woofer, surely? Less excursion for the 3" tweeter too. So less distortion.
But don't let me stop you.
No point in crossing lower than 2kHz on a 5" woofer, surely?
Less excursion for the 3" tweeter too. So less distortion.
But don't let me stop you.
Hi,
Its not a tweeter, its a 3" FR. Its a FAST.
X/o'd low in the middle of the baffle step.
rgds, sreten.
A CR across the bass inductor is still
a good idea, tuned to the 8KHz peak.
No point in crossing lower than 2kHz on a 5" woofer, surely? Less excursion for the 3" tweeter too. So less distortion.
Talking about distortion is irrelevant.
We like XOing low to take in the goodness of the FF85wKeN, no phase issues, and an XO that is below the ¼ wl of the C-C.
With these we push the XO down even lower… we tried 160 Hz, but 240 Hz gives a bit more midbass punch.
The 4" woofers stretch up to 5-10k by themselves.
dave
Almost missed your post there, Dave. You must do what you like. You have the speaker in front of you. TBH, I'm more interested in easy filters and just side-step any potential issues if I can. I like stuff that works. And I hate distortion and lack of dynamics. Now a genuine FAST with a side-firing woofer raises the issue of the crossover point much more, but we were talking about an MTM, I thought. Established territory.
So since a 5" won't struggle for dispersion at 2kHz, why not do that?
Well, maybe combing. But I can never hear it really.
I tried your 7kHz 10R and 1uF notch. It flattened out the woofer slope to near 6dB/octave and aligned the phase LR2 style. Took the HF impedance down too. But, TBH, it just seemed like fiddling with a design that basically works. I know what tank notches do, but amps don't seem to like driving them. And FWIW, Troels Gravesen only uses them with metal drivers, so maybe he hears the same things I do.
Well, I know what it is, but Lynn Olson's notion that power doubles with every octave you go down still applies even with a Fostex FF85wk. A puny 5W rms power handling. It's just a cone tweeter with a soft suspension in a bigger enclosure to my mind.
So since a 5" won't struggle for dispersion at 2kHz, why not do that?
Well, maybe combing. But I can never hear it really.
I tried your 7kHz 10R and 1uF notch. It flattened out the woofer slope to near 6dB/octave and aligned the phase LR2 style. Took the HF impedance down too. But, TBH, it just seemed like fiddling with a design that basically works. I know what tank notches do, but amps don't seem to like driving them. And FWIW, Troels Gravesen only uses them with metal drivers, so maybe he hears the same things I do.
Last edited:
FF85wKeN has no issues. And it sounds way better than the W14 as you go up. Power ratings are a who cares. And an XO that high gets the XO into the range where they are not nearly as benign, and out of the ¼ wl criteria (very important in my books)
These FASTs are quite seemless.
A comment about the uFonkenSET+woofT in the last pic i posted: http://www.diyaudio.com/forums/full-range/280225-fast-fr-sealed-all-way.html#post4462849
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.