BAM said:
Someone else would be better to answer this one but what about a Daline (Decoupled Anti-Resonant Line)? With this design you can just build a box and add a line to it.
Steve
An example of a Daline
They are under the Pioneer buyout, 4" extended range drivers #269-568 and #269-570. The frequency response is from 75 to 15,000 Hz. The difference between the #269-568 and thr #269-570 is that one has a light gray cone and the other has a black cone, but all of the T/S parameters have been verified to be identical. So it just becomes a choice between gray and black. I personally prefer the gray. The drivers themselves are super cheap. I got my four for $1.25 each, but I think there may have been a price drop. They have been described as having an "airy" midrange. Some of you may remember the "69-cent wonders" that Parts Express used to sell (4" full-range drivers originally designed to go into televisions). These are as close as Parts Excpress ahs come to topping that. You can also order these by the case. 32 drivers are $.90 each, for a total of about $32. They are extremely versatile. Their only limitation is their low power handling. 5 watts RMS, 10 watts max. I am building a pair to replace some boom box speakers that I loaned to a friend a long time ago to be powered by his stereo reciever. The reciever promptly blew them. So I am building some decent speakers to replace the plastic-box things that originally came with them. The 4" NSB (No Stinkin' Badges) drivers are capable of filling a large room at 5 watts, with only a hint of IM distortion at their loudest levels. The drivers are also video-shielded.
Where can I find information about Dalines? That is, box size, line length, etc.
I am interested in this type of transmission line, because it does not need to have the area of the drivers, and so you can do cool things like putting the line itself into the speakers' stand.
Also, will the System IV ever see production? It would be a shame if it doesn't. I know it's probably been replaced more or less by the Nonesuch 4, but I'd imagine the sound is really different between the two models given that one is a Daline and the other is something else (probably a transmisison line from what I gather.)
Also, another question has really been bugging me about designs such as the System IV and the Nonesuch 4. With the four drivers like that, don't you get sonic oddities from comb filtering? How is this avoided?
And yeah, 84 drivers for just one bandor...$100 per driver...8 drivers in all...that's $800 in drivers. But comparing the Bandors to the NSBs are like comparing apples and oranges. The Bandors are no-holds-barred uber-quality specialty units, and the NSBs are just general little 4" full-rangers for consumer electronics devices, so I'm sure there's no comparison. There is also a Pioneer 2" driver that I've got a couple of. I've been thinking a transmission line would be a fun thing to do with those. They actually manage to do everything the Bose cube drivers SHOULD be doing, but don't. (200-20,000 Hz)
I am interested in this type of transmission line, because it does not need to have the area of the drivers, and so you can do cool things like putting the line itself into the speakers' stand.
Also, will the System IV ever see production? It would be a shame if it doesn't. I know it's probably been replaced more or less by the Nonesuch 4, but I'd imagine the sound is really different between the two models given that one is a Daline and the other is something else (probably a transmisison line from what I gather.)
Also, another question has really been bugging me about designs such as the System IV and the Nonesuch 4. With the four drivers like that, don't you get sonic oddities from comb filtering? How is this avoided?
And yeah, 84 drivers for just one bandor...$100 per driver...8 drivers in all...that's $800 in drivers. But comparing the Bandors to the NSBs are like comparing apples and oranges. The Bandors are no-holds-barred uber-quality specialty units, and the NSBs are just general little 4" full-rangers for consumer electronics devices, so I'm sure there's no comparison. There is also a Pioneer 2" driver that I've got a couple of. I've been thinking a transmission line would be a fun thing to do with those. They actually manage to do everything the Bose cube drivers SHOULD be doing, but don't. (200-20,000 Hz)
rambling on...
I was perusing Planet10's website (http://www.t-linespeakers.org) and I came upon the design for the BD-Pipes. If my NSBs were only slightly more TL-friendly, I could probably build BD-pipes. What I'm wondering, though, is what the effect on the sound would be of having the second woofer mounted where it is on the BD-Pipes. Would it be dumb like Bose's Direct/Reflecting stuff or would it just provide a nice bipole effect? I might consider that (mounting the rear driver at an angle to add ambience) if it helps improve the sound.
I was perusing Planet10's website (http://www.t-linespeakers.org) and I came upon the design for the BD-Pipes. If my NSBs were only slightly more TL-friendly, I could probably build BD-pipes. What I'm wondering, though, is what the effect on the sound would be of having the second woofer mounted where it is on the BD-Pipes. Would it be dumb like Bose's Direct/Reflecting stuff or would it just provide a nice bipole effect? I might consider that (mounting the rear driver at an angle to add ambience) if it helps improve the sound.
Just type "daline" into Google. You'll find plenty of stuff, including Dave's excellent TL site.BAM said:
I built two experimental prototypes when I designed the System IVs. I could vary the box volume, port size, line length, area and shape (taper or not). Incredible things - looked like crosses between Daleks and petrol pumps (sci fi monsters/gas pumps for those over the pond). Should have kept them.
After experimenting, I ended up with the box the same size as the equivalent sealed box plus the line on the end. The original Daline article talked about a reflex enclosure decoupling the line from the box. In fact, I found that you just made the port the same area as the line. I don't think it has anything to do with ported box theory at all.
I'll never say "never" but I have no current plans to revive the System IV. I just love my new babies.
Actually, the Nonsuch 4 is a sealed box speaker. It is designed to be used with the separate bass enclosure(s). On its own, the System IV goes deeper and has a better bass. However, the Nonsuch 4 can play louder as it doesn't have as much cone displacement.
Also, the System IV is a pig to build and the cabinet construction for Nonsuch 4 is more advanced and less coloured (not that the System IV was particularly bad in this respect).
In practice it's not a problem provided I don't use more than four drive-units and I keep them as close together as possible. However, both the Nonsuch 4 and the System IV are "concert speakers". You have to sit down to listen. Stand up and you'll lose the frequencies over 10kHz. People listening never comment on the high frequencies except to say how well integrated they are with the mid-range and bass. Just check this review. No mention of comb effects at all. The guy who wrote it is a Martin Logan electrostatic owner. He heard the speakers that I'd left to demo at a dealer and was so impressed that he wrote the review. I was well chuffed.
Steve
Okay, after having mulled it over for a while in my head, I think I have the proper balance of concepts here. This is what I've come up with:
I don't know if to classify this as a one-way speaker or a two-way speaker, and you'll see why. The cabinet has a pentagonal shape. (I know of one other project using cheap 4" drivers that had a pentagonal cabinet, but I think mine's much less misguided than that particular one.) In the front, there is a vertical baffle, with drivers mounted on a horizontal axis. There will be one of the 4" NSBs, along with a mylar semi-dome tweeter element from Parts Express's Onkyo buyout, rear-mounted into a waveguide of sorts which looks sort of like a tiny round horn but much too short to achieve any meaningful loading. The NSB in the front (acting as a woofer) will cross over to the tweeter (more of a supertweeter)at about 10,000 Hz with a first-order high-pass on the tweeter and a first-order low-pass on the woofer. The sides of the enclosure will be vertical, and the top and bottom of the enclosure will be horizontal. In the back, the rear baffle will be vertical except for that the very top portion will be slanted back at 45 degrees. There I will mount another NSB (angled up and to the rear at a 45-degree angle in the way of the BD-Pipes). The rear NSB will be allowed to run through its full range of output (75 to 15,000 Hz.) I predict that this design decision will add some nice bipolar ambient effects, and make the speakers "look" bigger to one's ear than they actually are. The enclosure itself will be small, with the line built into the floor stand. I am hoping to be able to get a line such that its opening into the enclosure is about 3"x2". I have not yet determined the length of the line, and am still looking for information on how to determine this, or if a line with such small dimensions is even attainable. I am not sure whether to classify this as a one-way or two-way speaker because it has one set of drivers in a two-way configuration, and a second NSB driver running through its full range.
7V Steve: So, to get the required enclosure volume, I should model a ported enclosure, with a port that tunes the enclosure to a certain frequency, but instead of the reflex port I should substitute the line, correct? And perhaps the line should be the length of a quarter of the wavelength of what would otherwise be the enclosure's tuning frequency. Also correct?
The thing I am trying most to avoid here is having to build multiple test enclosures, if it may be avoided at all.
I don't know if to classify this as a one-way speaker or a two-way speaker, and you'll see why. The cabinet has a pentagonal shape. (I know of one other project using cheap 4" drivers that had a pentagonal cabinet, but I think mine's much less misguided than that particular one.) In the front, there is a vertical baffle, with drivers mounted on a horizontal axis. There will be one of the 4" NSBs, along with a mylar semi-dome tweeter element from Parts Express's Onkyo buyout, rear-mounted into a waveguide of sorts which looks sort of like a tiny round horn but much too short to achieve any meaningful loading. The NSB in the front (acting as a woofer) will cross over to the tweeter (more of a supertweeter)at about 10,000 Hz with a first-order high-pass on the tweeter and a first-order low-pass on the woofer. The sides of the enclosure will be vertical, and the top and bottom of the enclosure will be horizontal. In the back, the rear baffle will be vertical except for that the very top portion will be slanted back at 45 degrees. There I will mount another NSB (angled up and to the rear at a 45-degree angle in the way of the BD-Pipes). The rear NSB will be allowed to run through its full range of output (75 to 15,000 Hz.) I predict that this design decision will add some nice bipolar ambient effects, and make the speakers "look" bigger to one's ear than they actually are. The enclosure itself will be small, with the line built into the floor stand. I am hoping to be able to get a line such that its opening into the enclosure is about 3"x2". I have not yet determined the length of the line, and am still looking for information on how to determine this, or if a line with such small dimensions is even attainable. I am not sure whether to classify this as a one-way or two-way speaker because it has one set of drivers in a two-way configuration, and a second NSB driver running through its full range.
7V Steve: So, to get the required enclosure volume, I should model a ported enclosure, with a port that tunes the enclosure to a certain frequency, but instead of the reflex port I should substitute the line, correct? And perhaps the line should be the length of a quarter of the wavelength of what would otherwise be the enclosure's tuning frequency. Also correct?
The thing I am trying most to avoid here is having to build multiple test enclosures, if it may be avoided at all.
BAM:
Any chance of a quickie sketch in MSPaint or some equivalent? I am having a hard time following your description.
Okay, but watch out for that crossover. Generally, it is recommended that the centers of the speakers being crossed over be one wavelength or less of the crossover frequency.
1 Hz = 13,500 inches. Therefore, the wavelength of 10,000 Hz is 1.35 inches, meaning that you will not be able to be anywhere near one wavelength at the crossover frequency. Adding to the difficulty are the first order crossovers, which have substantial output from both speakers through three or four octaves. Major lobing concerns.
Perhaps, instead of having the 4" and tweeter abutting each other, the answer is to move the tweeter so it is 2, 3 or 4 wavelengths away from the woofer. I don't know. You might want to start a separate thread just for this crossover problem-it appears complex.
Any chance of a quickie sketch in MSPaint or some equivalent? I am having a hard time following your description.
BAM said:
Okay, but watch out for that crossover. Generally, it is recommended that the centers of the speakers being crossed over be one wavelength or less of the crossover frequency.
1 Hz = 13,500 inches. Therefore, the wavelength of 10,000 Hz is 1.35 inches, meaning that you will not be able to be anywhere near one wavelength at the crossover frequency. Adding to the difficulty are the first order crossovers, which have substantial output from both speakers through three or four octaves. Major lobing concerns.
Perhaps, instead of having the 4" and tweeter abutting each other, the answer is to move the tweeter so it is 2, 3 or 4 wavelengths away from the woofer. I don't know. You might want to start a separate thread just for this crossover problem-it appears complex.
BAM said:
Looking at the specs for the speakers you mentioned, we get the following:
Specifications for #269-570:
Power Handling: 5 watts RMS/10 watts max
VCdia: 3/4"
Le: .60 mH
Znom: 8 ohms
Re: 7.50 ohms
Frequency range: 75-15,000 Hz
Fs: 105 Hz
SPL: 86 dB 2.83V/1m
Vas: .08 cu. ft.
Qms: 5.0
Qes: .90
Qts: .76
Xmax: 1 mm
I don't know how well the published specs agree with the actual specs, but your Qts does not seem to be as out of line as you think.
Also, I have seen Qts be raised by a series resistor. I'll give you the link if you want-it works. I have not tried it, but it seems logical that Qts should therefore be able to be lowered by a parallel resistor. However, until I check it out,I cannot say for sure. If your Qts is indeed just a hair over the line, that parallel resistor probably will not materially lower sensitivity.
PS: If you put these speakers in a ported box, with a Qts of 0.79, you'll get distortion to the max.
However, if indeed you can lower the Qts by means of a parallel resistor, then you can take the Qts down to 0.5 and get a perfectly workable ported box with a box size larger than Vas and an F3 well below Fs. Of course, the more you lower the Qts, the less the sensitivity will be. But then, using multiple woofers can result in a sensitivity increase, so you have a balancing act.
Might work. Might be fun. Just a suggestion for your investigation.
Here is a post by myself where I scan a page out of the Great sound Stereo Speaker Manual by the excellent audio author David Weems. The formula he gives is for a series resistor, therefore useful only for raising Qts. Whether or not the parallel resistor will lower Qts remains to be seen, but I guess that it probably would.
http://www.diyaudio.com/forums/showthread.php?postid=93570#post93570
However, if indeed you can lower the Qts by means of a parallel resistor, then you can take the Qts down to 0.5 and get a perfectly workable ported box with a box size larger than Vas and an F3 well below Fs. Of course, the more you lower the Qts, the less the sensitivity will be. But then, using multiple woofers can result in a sensitivity increase, so you have a balancing act.
Might work. Might be fun. Just a suggestion for your investigation.
Here is a post by myself where I scan a page out of the Great sound Stereo Speaker Manual by the excellent audio author David Weems. The formula he gives is for a series resistor, therefore useful only for raising Qts. Whether or not the parallel resistor will lower Qts remains to be seen, but I guess that it probably would.
http://www.diyaudio.com/forums/showthread.php?postid=93570#post93570
KW: I don't think a parallel resistor will have any effect on Qts. Think about the limiting case of a driver being driven directly (hence shunted) by an amplifier of near-zero source impedance.
Rather than using a series resistor to increase Qts, a little passive or active eq before the power amp will not be as wasteful. In box enclosures, it's pretty easy to adjust the tuning to accomodate the particular driver. Weems like the simple things, and the resistor has that virtue.
Rather than using a series resistor to increase Qts, a little passive or active eq before the power amp will not be as wasteful. In box enclosures, it's pretty easy to adjust the tuning to accomodate the particular driver. Weems like the simple things, and the resistor has that virtue.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.