From 80hz to 18+khz I would want a 8" Beryllium-or-equivalent Coaxial, with the tweeter-to-cone transition based on the excellent Kef Q900 midrange. Cibstructed with larger roundovers to minimize diffraction.
DEQX / Linear Phase on the crossover. Hypex NC400s on the drivers.
Toss in a Pair of 13.5" sealed subs to get that response down to 30hz and below range if possible.
Finally some adequately thick (6"?) denim insulation between the speaker and the desk to minimize the desk reflection.
Equalized flat below shroeder frequency.
Well, most of the above was a bit excessive. The coaxial however would be the focal point of a desktop setup in my opinion.
The midwoofer cone acts as a waveguide for the high frequency radiation and it's movement is not helpful in this respect. The higher quality KEF speakers use a coaxial mid/tweeter and crossover quite high to a separate woofer, as does the Genelec. Perhaps a 4"-5" coaxial mid/tweeter and a 4"-5" woofer would have higher sound quality? Or even a 4"-5" coaxial mid/tweeter placed in front of an 8" woofer in the manner of an ME Geithain?
Setting aside the beryllium and the 8", I would agree that the 5.25" coaxial midwoofer/tweeter from the Q series crossed to a subwoofer could form the basis of a good quality sanely priced system.
And just how much is an 8" driver about to move, two feet away from you face with a -6db point at 100hz......?
A fair bit more than a 5" mid crossed at 500 Hz. Would your 8" driver be designed for small movement and a surround to minimise diffraction like the KEF R series driver or as a midwoofer like the KEF Q series driver? And if you are going big wouldn't you want to crossover a bit lower to the subwoofer? I guess it's a bit academic given the practical choice is between a 5" mid/tweeter or an 8" midwoofer/tweeter if using KEF drivers.
For a nearfield speaker I would want a true point source. I didn't do the math but I suspect you'd have imperfections in the off-axis phase tracking, which is undesirable because of the likelyhood of shifting away from the theoretical sweet spot. The whole point of the coax is to keep one radiating source with a wide listening window, and I'm not sure 500hz is a low enough frequency for a 5" mid + woofer. I'd have to do the math so don't mind me if I'm wrong. It would be fine in farfield where the angle at IE 8 feet barely changes, but that's a different case.
With that said I'm not sure your statement is quite true. Factoring in the SPL gains from desk placement (IE a desktop speaker would have minimal baffle step) I decided that the typical speaker would only need to produce 98db SPL @ 1m for a whopping 105db peak (THX Reference Level, and probably 5db louder than any of us ever listen!) at the typical desktop LP.
Now factor in the -6db point of a Linkwitz Riley crossover. So you only really need 92db SPL @ 1m !!
For a 13cm midrange at 500hz, you need 0.076 mm of excursion
For a 22cm midrange at 100hz, you need 0.718 mm of excursion
Basically. both would stay under 1mm of excursion. Generally for a coax you're fine as long as you're under around 3mm of excursion. I don't think a difference of about .6mm of excursion is enough to modulate the sound. I prefer the simplicity of the two-way coax for this application. Maybe not for a farfield speaker, though.
I would not need much excursion, so i would use the R style surround. That said - minimizing diffraction is a pretty silly goal for a desktop setup. You've got a ton of diffraction going on from every single object there - rather unlike a KEF Blade sitting 6 feet into a dedicated listening space.
I would try to limit the upper midrange / lower treble reflection off the desk, so the larger 8" midwoofer makes more sense to me than the 5" midwoofer.
Why?
The large woofer is to control the pattern. I'd even go 10" if the size wasn't too much for a desktop. I'm not interested in the mains producing output below 80hz at all.
Last edited:
I understand the thinking of a coaxial speaker for the desktop but feel that as long as the center to center distance isn't to great and the radiation patterns match at crossover that shouldn't be much of an issue. As for the lower cutoff frequency I wouldn't consider anything above 150hz as practical without causing a noticeable discontinuity in the sound. Below about 150hz we wouldn't have any directional information to speak of. I am not looking to use any production drivers, they are all new designs and dedicated to this specific usage. A beryllium dome tweeter is one of the devices but I may be forced to also have a soft dome version just because of the increased cost of the beryllium diaphragm itself. I am attempting to keep the baffle width as narrow as possible as this would obviously be a limitation on most desktops or in many smaller monitor applications. So don't think of existing devices, that is not something to limit the discussion to.
Personally I do think it's an issue. Now I use a pretty typical 4" two-way for my desktop. For what I use it for, it's perfectly adequate but far from ultimate.
It has some obvious problems. dispersion is so wide that the frequency response across the band goes over the place, even a gated measurement. The only way to minimize that would be to raise the directivity index, but then you need a waveguide. how do you deal with the center to center distance on a waveguide? You either don't, or you use a coax.
The modern KEF, Tannoy, B&C, Genelec, and Pioneer coaxial drivers are just too good and "uncompromised" for me to discount them. If I were making a desktop speaker I would be looking to develop something on par or better.
And I say this as a guy whose main system is extremely different - using a tall, narrow vertical planar midrange and a ribbon tweeter.
I know soft domes can sound pleasant but they still have a character of their own. Ceramic Aluminum, Doped Magnesium, Beryllium-Copper, Woven Kevlar, Boron-Carbide, etc domes just have more transparency if they are designed correctly. Beryllium is of course my "ultimate" choice but materials wise there are a lot of choices for domes. A lot more problematic are the materials for larger cones, where that aformentioned KEF Q900 would fall short (It wants a crossover around 1.5khz or so, but it uses an 8" aluminum cone). In that case I would be looking at a Kevlar weave that can control the breakup peak. B&W's FST midrange comes to mind. Also coming to mind are the Accuton Ceramic with rubber damping.
Then consider going with an elliptical tweeter firing into an ellipsoidal coaxial dimensions something like 4" wide and 7" tall. Don't ask me how one would manufacture that. If I could I would
Last edited:
I pretty much built my ideal desktop system.
4" FR, 6.5" TB subs, crossed at ~500Hz.
When nearfield, plenty of dynamics and LF extension, while the beaming of the FR driver allows the treble to be tuned to taste.
Currently they're sat on some shelves in my uni dorm room. Less than ideal, but still sounding nice. Also far better (for uni, read: louder) than any iPod dock someone plugs in.
Chris
4" FR, 6.5" TB subs, crossed at ~500Hz.
When nearfield, plenty of dynamics and LF extension, while the beaming of the FR driver allows the treble to be tuned to taste.
Currently they're sat on some shelves in my uni dorm room. Less than ideal, but still sounding nice. Also far better (for uni, read: louder) than any iPod dock someone plugs in.
Chris
what follows is merely my opinion. An 8 inch coax would be mental! Good, but crazy. I have a large monitor though, there simple isnt the room for a 10-20 litre box each side. Im planning to go the sub n sat route. Using 3inch fullrangers, or an array of 2 inch drivers as i have a few to play with. Maybe a sub using a small woofer, considered 6th order BP, but sealed is the best way i reckon. Bass EQ and small 3 channel amp. TDA2050 monobloc or similar.
Last edited:
RockLeeEV,
The dome tweeter will be a true beryllium dome, not an alloy that is only a small percentage beryllium. The alloy beryllium/aluminum material is only a very small amount of beryllium is more of a marketing tool than any serious change compared to pure aluminum. That is what Brush-Wellman will tell you themselves and they are the producers of both materials. You are also reading my mind on a few other aspects I will keep to myself right now.
The dome tweeter will be a true beryllium dome, not an alloy that is only a small percentage beryllium. The alloy beryllium/aluminum material is only a very small amount of beryllium is more of a marketing tool than any serious change compared to pure aluminum. That is what Brush-Wellman will tell you themselves and they are the producers of both materials. You are also reading my mind on a few other aspects I will keep to myself right now.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.