The following is the ranting of someone who has just spent time at the library trying to teach himself basic acoustics. All information is likely to be total bunk. Feel free to laugh.
Recently, as a result of the "T-amp revolution" and computer speaker mass production, fullrange speakers have come back into style. Better quality control and clever design have made speakers with nearly flat responses and great quality, like the CSS FR125s, reasonably affordable.
However, all fullrange speakers have a problem: If only due to low mass and area (I think...feel free to correct me), most FR speakers make a tradeoff in bass for their treble. Even those that can produce bass sound better if they're crossed over to a sub at 100hz or so.
So, why not just make some 2-ways?
In this era of the Gainclone and 41hz kits, active speakers are getting to the point where they're a better value than speakers with traditional crossovers - a good set of caps and inductors can run up to more than the cost of a Gainclone matched with some itsy-bitsy high-quality inductors and caps in a Linkwitz-style op-amp based active crossover/filter.
By using two amplifiers set to a correct proportion of gain (to compensate for the difference in efficiency), a pair of inexpensive 2-ways can easily be built using inexpensive fullrange drivers and some subs. Open-baffle is no longer a big issue if you're crossing over the speaker at 100hz, and bass can be easily provided by a 40$ driver like the 8" Silver Flute down to 30hz - and pretty flat, too.
Furthermore, a 1st-order crossover could be used without much trouble. By crossing over a standard FR driver (with an Fs of about 100hz) with said Silver Flute woofer (flat up to about 1khz) with a 1st order crossover around 225hz, the sound will smoothly move from FR to woofer - they can produce sound without noticeable distortion until they're over 16db down in power.
Am I loopy, or does this make more sense than a traditional 3-way? This would be especially advantageous for HT speakers - a sub would be much less important.
Recently, as a result of the "T-amp revolution" and computer speaker mass production, fullrange speakers have come back into style. Better quality control and clever design have made speakers with nearly flat responses and great quality, like the CSS FR125s, reasonably affordable.
However, all fullrange speakers have a problem: If only due to low mass and area (I think...feel free to correct me), most FR speakers make a tradeoff in bass for their treble. Even those that can produce bass sound better if they're crossed over to a sub at 100hz or so.
So, why not just make some 2-ways?
In this era of the Gainclone and 41hz kits, active speakers are getting to the point where they're a better value than speakers with traditional crossovers - a good set of caps and inductors can run up to more than the cost of a Gainclone matched with some itsy-bitsy high-quality inductors and caps in a Linkwitz-style op-amp based active crossover/filter.
By using two amplifiers set to a correct proportion of gain (to compensate for the difference in efficiency), a pair of inexpensive 2-ways can easily be built using inexpensive fullrange drivers and some subs. Open-baffle is no longer a big issue if you're crossing over the speaker at 100hz, and bass can be easily provided by a 40$ driver like the 8" Silver Flute down to 30hz - and pretty flat, too.
Furthermore, a 1st-order crossover could be used without much trouble. By crossing over a standard FR driver (with an Fs of about 100hz) with said Silver Flute woofer (flat up to about 1khz) with a 1st order crossover around 225hz, the sound will smoothly move from FR to woofer - they can produce sound without noticeable distortion until they're over 16db down in power.
Am I loopy, or does this make more sense than a traditional 3-way? This would be especially advantageous for HT speakers - a sub would be much less important.
leadbelly said:Er, so you think fullrange drivers have recently come back into style, and you yourself have discovered the potential to use a fullrange driver in a multi-driver speaker? What's next, patenting the bicycle?
Well, no.
This is more along the issue of "If we're trying to carry 300 lbs. on a bicycle...why not just use a tricycle?"
That, and I've yet to see any. For example, while John Krutke seems to like the B3S drivers, he hates the lack of bass. Why not just use two open-baffle with a good woofer and be done with it?
These fullrange drivers generally need a tweeter as well. I'm using the Aurasound NS3-193-8a's as rear speakers in surround sound, and when used for normal music you can definitely tell that the treble just isn't up to snuff.
In my experience, treble from full range drivers is just not as good as treble from tweeters. It beams, it's not as clear, and it just doesn't give me the feeling that I'm hearing the real thing. On the other hand, the midrange (female vocals that are mostly vowels) is to die for.
There's no doubt that these "full range" drivers have a place- but I think that place is above a woofer and below a tweeter.
Linkwitz has published all of the spreadsheets you need to get a good estimate of how much excursion (or how many drivers) you need for your application for a given SPL. Dipole or sealed, my simulations have generally shown that at least two of these "full range" drivers are needed to provide enough low-end to cross to a subwoofer and play loudly. If you cross over higher, you usually run into low sensitivity and low power handling. If I was going to use the any three inch driver, I'd probably use four of them in an MMTMM configuration.
Just my two cents.
In my experience, treble from full range drivers is just not as good as treble from tweeters. It beams, it's not as clear, and it just doesn't give me the feeling that I'm hearing the real thing. On the other hand, the midrange (female vocals that are mostly vowels) is to die for.
There's no doubt that these "full range" drivers have a place- but I think that place is above a woofer and below a tweeter.
Linkwitz has published all of the spreadsheets you need to get a good estimate of how much excursion (or how many drivers) you need for your application for a given SPL. Dipole or sealed, my simulations have generally shown that at least two of these "full range" drivers are needed to provide enough low-end to cross to a subwoofer and play loudly. If you cross over higher, you usually run into low sensitivity and low power handling. If I was going to use the any three inch driver, I'd probably use four of them in an MMTMM configuration.
Just my two cents.
joe carrow said:
Who said a 3" driver is mandatory? Or that a proper "subwoofer" driver must be used?
Many have had excellent luck with Fostex' FE126 and FE107 fullrange drivers as high as 18khz. They can go below 100hz with ease.
Another option is the inexpensive Hi-Vi B3S. It's 10$, and although abysmally inefficient (about 81db/w), two in parallel (still only 4 ohms) will result in an efficiency of about 88db/w (providing I did'nt flub my math: 2 running at 1/2 watt should produce 78db, and multiplying x2 would be +10db?)
Although limited in power and requring a notch filter (as per Krutke's reccomendations), they have excellent high-frequency response - and look awesome.
http://zaphaudio.com/audio-speaker18-channelmatching.gif
A match would be the technically non-subwoofer Silver Flute drivers. The W20RC38 8" drivers are pretty flat down to about 30hz from about 3 khz according to Madisound's specs, and are none too expensive, either.
http://www.madisound.com/cgi-bin/index.cgi?cart_id=2431946.9632&pid=1081
Perhaps the Hi-Vi's could be run open-baffle (attractive, simple, and supposed to work pretty well) with the woofer in a standard bass-reflex enclosure?
A LM4780 (half for the HI-VIs, half for the woofers) per side would deliver more than adequate power, and don't cost too much to use.
Hmmm.........
If you study acoustics and loudspeeaker design you'll learn that
there are umpteen ways to skin a cat, unpteen people who think
they know the right way to do it, and very few people who are
actually any good at it.
If you want to build a speaker of the type you suggest, fine
ask questions. But don't simply ask " why aren't all speakers
like this, because it seems the right way to do it to me ".
If the silver flutes sensitivity is for real (94dB/W), then it could be
used with a 3" fostex (88-89dB/W) open baffle crossed over with
a series 1st order somewhat higher than 225Hz, with possibly a
supertweeter (Apex Jr ribbon ?) added to taste.
/sreten.
If you study acoustics and loudspeeaker design you'll learn that
there are umpteen ways to skin a cat, unpteen people who think
they know the right way to do it, and very few people who are
actually any good at it.
If you want to build a speaker of the type you suggest, fine
ask questions. But don't simply ask " why aren't all speakers
like this, because it seems the right way to do it to me ".
If the silver flutes sensitivity is for real (94dB/W), then it could be
used with a 3" fostex (88-89dB/W) open baffle crossed over with
a series 1st order somewhat higher than 225Hz, with possibly a
supertweeter (Apex Jr ribbon ?) added to taste.
/sreten.sreten said:Hmmm.........
If you study acoustics and loudspeeaker design you'll learn that
there are umpteen ways to skin a cat, unpteen people who think
they know the right way to do it, and very few people who are
actually any good at it.
If you want to build a speaker of the type you suggest, fine
ask questions. But don't simply ask " why aren't all speakers
like this, because it seems the right way to do it to me ".
If the silver flutes sensitivity is for real (94dB/W), then it could be
used with a 3" fostex (88-89dB/W) open baffle crossed over with
a series 1st order somewhat higher than 225Hz, with possibly a
supertweeter (Apex Jr ribbon ?) added to taste.
There's about a million reasons why speakers would'nt be built like this. The only good reason why they would is cost, and convienience - open-baffle speakers are easy to make.
I was just wondering why I have not seen any designs like this, and why no one else has done it. I was hoping to try it myself, and as a n00bie, I wanted to know if I had a flawed idea.
Spasticteapot said:
Why do you keep saying this, it is just driving me up the wall! I have seen so many examples of this type of speaker over the years, right on this forum. SEARCH
Here's just one, and I only mention this one because it is in my memory:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=9706&highlight=
Most fullrange designs are intended to go as low as possible, then cross over to a seperate subwoofer. They're essentially seperate speakers.
Instead, I was curious as to the viability of using a very slow rolloff on the crossover around 250hz (say, -6db/octave) for a proper 2-way instead of just having a FR speaker and a subwoofer.
In other words, the FR drivers would be used in place of traditional mid-tweeters, and the woofers would be used not just for low-end reinforcement, but as normal woofers.
Instead, I was curious as to the viability of using a very slow rolloff on the crossover around 250hz (say, -6db/octave) for a proper 2-way instead of just having a FR speaker and a subwoofer.
In other words, the FR drivers would be used in place of traditional mid-tweeters, and the woofers would be used not just for low-end reinforcement, but as normal woofers.
Have a look at Basszilla - it's an 8" full range driver with a 15" pro woofer. The crossover is around 150-200hz, but I think it's passive. you could obviously save money on components by making such a low crossover active.
The higher-end versions of basszilla use various tweeters for the highest frequencies.
The higher-end versions of basszilla use various tweeters for the highest frequencies.
This is really what DIY is all about. Understanding what others are doing, seeing possibilities and trying them.
I designed my first speaker with a pair of NSB 4" full-range drivers in an MTM arrangement, with an inexpensive NEO tweeter and crossed to a pair of 12" woofers.
Then I found the crossover points that sounded best to me.
Try it, and report back.
Doug
Gentlemen,
I think something is being overlooked here. A full range driver CAN be made to work over the full audio spectrum. It's not done commercially due to the cost of magnets, and the buying public has been convinced they need to see a baffle covered in cones.
If a FR driver needs a tweeter, chances are the magnet isn't big enough. If it needs a sub, then chances are the Fr is set too high.
I choose to use a sub to assist 30 - 60 Hz. So if that constitutes a 2 way system, my 10" tweeters have magnets 7" in diameter on an alloy basket, and weigh 11lb. (If I switch off the "tweeters", I can barely hear the sub)
If DIY's like us can appreciate the benifits of a FR driver, the manufacturers will see the market, and continue to make the drivers.
Regardless of whether you choose to go FR or multiple drivers, much can be gained from exploiting the full capability of a given driver, even $10.00 ones.
Let's not shake up each other, shake up the manufacturers instead.
I think something is being overlooked here. A full range driver CAN be made to work over the full audio spectrum. It's not done commercially due to the cost of magnets, and the buying public has been convinced they need to see a baffle covered in cones.
If a FR driver needs a tweeter, chances are the magnet isn't big enough. If it needs a sub, then chances are the Fr is set too high.
I choose to use a sub to assist 30 - 60 Hz. So if that constitutes a 2 way system, my 10" tweeters have magnets 7" in diameter on an alloy basket, and weigh 11lb. (If I switch off the "tweeters", I can barely hear the sub)
If DIY's like us can appreciate the benifits of a FR driver, the manufacturers will see the market, and continue to make the drivers.
Regardless of whether you choose to go FR or multiple drivers, much can be gained from exploiting the full capability of a given driver, even $10.00 ones.
Let's not shake up each other, shake up the manufacturers instead.
Geoff H said:
That's exactly what I was reccomending.
However, instead of a very steep roloff on the low-pass for the woofer, a very shallow roll-off could be used instead. This would allow less expensive drivers (like the NSB) to be better integrated with the woofers.
The idea here is to think of a FR + woofer speaker as a standard 2-way, except that the "tweeter", as you said, can go down to 100hz. Any speaker builder will tell you that a proper crossover (instead of a generic low-pass filter) will give vastly better sound quality, so why not treat the subwoofer as part of the speaker instead of a seperate entity?
I'm trading for some NSBs, and if I can get some decent low-frequency woofers, I hope to try exactly this.
In my case, the sub is temp. New enclosures for the wide range will fix that. Currently infinite baffle, 19 litres. I wasn't sure how low I could drive these. They were modified to x-over at 500. However the bass response hasn't suffered much. A QWTL will take care of that.
So the sub is not part of the system. More of a bass boost, as the system evolves.
Your idea has merit. Explore the pssibilities of the NSBs, and work on a xo freq. Let the WR driver roll of naturely, and use the supplement driver to fill the gap, with a low pass filter to match the roll off of the WR driver.
And the size of the supplemental driver is not critical. The SPL at the required range is. My sub is small compared to the WRs. Different, but it only has to deal with an octave, not 9.
So the sub is not part of the system. More of a bass boost, as the system evolves.
Your idea has merit. Explore the pssibilities of the NSBs, and work on a xo freq. Let the WR driver roll of naturely, and use the supplement driver to fill the gap, with a low pass filter to match the roll off of the WR driver.
And the size of the supplemental driver is not critical. The SPL at the required range is. My sub is small compared to the WRs. Different, but it only has to deal with an octave, not 9.
That works for me, for now. If you havn't any yet, try to get some Sheffield Lab recordings. These are live recordings, such that the output from custom built mics are fed straight to the master. No multi track or tape. They have fantastic phase linearity and transient attack.
You'll get more info from that than specs or sine waves.
You'll get more info from that than specs or sine waves.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.