Hi Nemophyle,
Did you get my mail?
Or check this thread for more:
http://www.diyaudio.com/forums/showthread.php?s=&postid=759892#post759892
Did you get my mail?
Or check this thread for more:
http://www.diyaudio.com/forums/showthread.php?s=&postid=759892#post759892
thank you Ed, i'm looking forward to it 
has anyone constructed a big two way in the dr geddes fashion with BIG drivers that have decent excursion in the midband ?
Dr geddes uses a 15", i was thinking about using a 10" or 12" because of the size of my room.
i understood that in the dr geddes fashion, the crossover is at the frequency that equals the wavelength of both the bass driver and waveguide, in the case of dr geddes loudspeaker 15" = 900hz
if you are still around the corner zaph, i dont have a very scientifical approach about loudspeakers, not because i don't wan't to, but because at my point of knowledge i can't, but i try to be the most systematic that i can about loudspeakers.
I saw on your site that you use only small woofers, is this for convenience, because you don't see real advantages to using big ones, making three ways, etc ?
personnally i listen to lots of "soft" acoustic music as jazz, african music, ethnological recordings of traditionnal music, and at the same time i do electronic music on my computer, so i need a good low end extension, lots of power handling and a controlled directivity system for mastering purposes. So i thought a waveguide system with a 10" woofer crossed at 1350hz might do the trick pretty nicely.
now i would like to know what you (or other knowledgeable guys) think about such an idea . is it ridiculous ? with a 10" waveguide might one of the low resonnance tweeters around work with a 1350hz crossover ? it might need steep slopes, so is there any opposition to use steep slopes in a WG system (provided i would cross active and so equalize the gain of the waveguide) ?
i already have a good 10" and a good tweet for this application, the only question left is , am i gonna buy a good medium driver at the price of increased complexity, or go the WG way ?
if anyone has knowledgeable information on this subject, please feel free to comment .
has anyone constructed a big two way in the dr geddes fashion with BIG drivers that have decent excursion in the midband ?
Dr geddes uses a 15", i was thinking about using a 10" or 12" because of the size of my room.
i understood that in the dr geddes fashion, the crossover is at the frequency that equals the wavelength of both the bass driver and waveguide, in the case of dr geddes loudspeaker 15" = 900hz
if you are still around the corner zaph, i dont have a very scientifical approach about loudspeakers, not because i don't wan't to, but because at my point of knowledge i can't, but i try to be the most systematic that i can about loudspeakers.
I saw on your site that you use only small woofers, is this for convenience, because you don't see real advantages to using big ones, making three ways, etc ?
personnally i listen to lots of "soft" acoustic music as jazz, african music, ethnological recordings of traditionnal music, and at the same time i do electronic music on my computer, so i need a good low end extension, lots of power handling and a controlled directivity system for mastering purposes. So i thought a waveguide system with a 10" woofer crossed at 1350hz might do the trick pretty nicely.
now i would like to know what you (or other knowledgeable guys) think about such an idea . is it ridiculous ? with a 10" waveguide might one of the low resonnance tweeters around work with a 1350hz crossover ? it might need steep slopes, so is there any opposition to use steep slopes in a WG system (provided i would cross active and so equalize the gain of the waveguide) ?
i already have a good 10" and a good tweet for this application, the only question left is , am i gonna buy a good medium driver at the price of increased complexity, or go the WG way ?
if anyone has knowledgeable information on this subject, please feel free to comment .
Happy Holidays!
I'd say yes it's doable. But seeing as this is sort of new ground for DIY'ers, you're probably going to be on your own. I don't think I'll be building a larger waveguide myself, simply because I have too much else brewing at the moment. (like a 50" high, 6 driver 3.5 way)
Below we see the before and after of the dome with and without the waveguide. Black = flat baffle and blue = waveguide. We get a 6 dB boost in efficiency, centered on about 2.5 kHz. It's "only" a 6dB boost because the waveguide as I modified it is relatively short and shallow.
While the waveguide I used was 6.5", it's effectively a bit smaller than that because of the gradual radius of the lens. Now, if we use the Parts Express 8" lens, without the outside radius it's effectively a much larger lens. I would expect the boost to be much larger and centered much lower in frequency. I wouldn't be surprised if it's centered at 1kHz or lower, easily making a 1kHz crossover point possible. Remember, the increase in efficiency just means the tweeter will not have to work as hard to reach a certain level and that's why it can cross lower.
I also wouldn't be surprised if the boost was around 10-15dB rather than the 6dB I got with my shallow lens. With the bigger boost, don't expect a single cap to work like it did for me. The crossover will take a lot more work to get it flat.
Go for it and let us know how it turns out.
Regards,
john
Nemophyle said:
I'd say yes it's doable. But seeing as this is sort of new ground for DIY'ers, you're probably going to be on your own. I don't think I'll be building a larger waveguide myself, simply because I have too much else brewing at the moment. (like a 50" high, 6 driver 3.5 way)
Below we see the before and after of the dome with and without the waveguide. Black = flat baffle and blue = waveguide. We get a 6 dB boost in efficiency, centered on about 2.5 kHz. It's "only" a 6dB boost because the waveguide as I modified it is relatively short and shallow.
While the waveguide I used was 6.5", it's effectively a bit smaller than that because of the gradual radius of the lens. Now, if we use the Parts Express 8" lens, without the outside radius it's effectively a much larger lens. I would expect the boost to be much larger and centered much lower in frequency. I wouldn't be surprised if it's centered at 1kHz or lower, easily making a 1kHz crossover point possible. Remember, the increase in efficiency just means the tweeter will not have to work as hard to reach a certain level and that's why it can cross lower.
I also wouldn't be surprised if the boost was around 10-15dB rather than the 6dB I got with my shallow lens. With the bigger boost, don't expect a single cap to work like it did for me. The crossover will take a lot more work to get it flat.
Go for it and let us know how it turns out.
Regards,
john
Attachments
Just for grins, here's a 1/2" waveguide on a Vifa D26NC55. It doesn't really lower the potential crossover point at all, but it does give about 3dB of boost centered on 3.5kHz. Actually, this is usefull, because a waveguide this small will still exhibit diffraction ripple on a baffle, and there's usually a dip right in the 3-3.5kHz area. So, while it's small, it does serve a purpose. (and it makes the odd flange easier to mount from behind too)
Attachments
Crawling thru the pile of horns at Partsexpress, I see the B+C ME10, at 90x60 degrees, and the Sammi Sound SH-150 at 90x90 degrees as the only ones close to that dispersion. Though I wonder how good/bad a different dispersion might work in different rooms/different situations.
John, looking again at your graph, I see how the boost does taper off to 3db or less by the time you get to 1Khz. So, a larger one would be desired for crossing down low with 8"ers or more...
I'm wondering what happens to the angular dispersion curve in milling down the PE 75 deg horns.. a symmetrical horn curve wouldn't change, an exponential would? And the PE horns are "nonconventional"... whatever that means. Would be nice if they did widen their pattern as they were shortened. Best way is to dive in and start experimenting...
And a question on the '1/2" waveguide', 1/2" piece of MDF and roundover bit? ;-) Have to figure out the OD of a 1 1/2" roundover surrounding a 1 1/2" hole... probably only about the same as the 6" we already have.. and we probably need a much flatter curve than the radius. How big do massive panel trim bits get?
John, looking again at your graph, I see how the boost does taper off to 3db or less by the time you get to 1Khz. So, a larger one would be desired for crossing down low with 8"ers or more...
I'm wondering what happens to the angular dispersion curve in milling down the PE 75 deg horns.. a symmetrical horn curve wouldn't change, an exponential would? And the PE horns are "nonconventional"... whatever that means. Would be nice if they did widen their pattern as they were shortened. Best way is to dive in and start experimenting...
And a question on the '1/2" waveguide', 1/2" piece of MDF and roundover bit? ;-) Have to figure out the OD of a 1 1/2" roundover surrounding a 1 1/2" hole... probably only about the same as the 6" we already have.. and we probably need a much flatter curve than the radius. How big do massive panel trim bits get?
The ME10 looks good - a more manageable 5.1" x 5.1", and good loading down to 1.5 kHz.
This one has a 1" throat, so it looks like it's ready to go.
I'm not clear on just what happens when you put a dome tweeter on a horn meant for a compression driver.
Does the fact that there's no compression make it behave like John's design (better loading and efficiency at the lower end of the band but the civilized behavior of a dome at the top) or was that from shortening the waveguide?
Thanks
This one has a 1" throat, so it looks like it's ready to go.
I'm not clear on just what happens when you put a dome tweeter on a horn meant for a compression driver.
Does the fact that there's no compression make it behave like John's design (better loading and efficiency at the lower end of the band but the civilized behavior of a dome at the top) or was that from shortening the waveguide?
Thanks
AJinFLA said:
Agreed about the edge transition, but I don't see why 75deg coverage is a bad thing. From experimentation I've found that smaller/narrower rooms really need narrower directivity, while larger spaces generally work better with wider coverage. The name of the game seems to be "limit early midrange reflections as much as possible, while not screwing up everything else".
By "narrower" I mean 60-75deg and "wider" I mean ~90deg. Also, a narrower directivity waveguide will allow a higher crossover point, which may be a good thing (but may just as well be a bad thing).
Zaph, I was looking at the plans for the Waveguide TMM again last night, and I noticed that the spacing between the bottom of the waveguide and the top of the upper woofer is not the same as the spacing between the two woofers. It's not a huge issue, but I was wondering if it would cause a serious problem to move the lower woofer up by a half inch (for aesthetic reasons)?
Wodgy said:
Actually, there's a reason for the driver spacing I used. Each of the two woofers has a wall or a crossmember nearby. if I were to move the bottom woofer up, the top wall of the lower enclosure might become close enough to the wall to restrict airflow enough and cause distortion from uneven cone loading.
The Waveguide has no airflow behind it, and as such, it can overlap the crossmember right below it without problem. Moving the waveguide closer broadens the vertical lobe and allows more flexibility in listening height.
So yes, I planned it that way. I basically chose to deviate from driver mounting symmetry in the interests of sound quality and vertical off axis listening. I am overly picky and analytical though, and it probably would not hurt at all if you moved the lower woofer up. It may not even make a difference at all.
On a subject related to this project, I'm currently testing some larger waveguides. 8" and 10" from Parts Express.
After reading the theory section as well as reviews linked here
http://www.4sptech.com/version2/index.html
I'm really sold on waveguides.
I'm still not clear on some of the basics.
Are waveguides a subset of horns, following horn theory?
Why is it that waveguides are mostly concave, while horns are convex?
SP Tech crosses theirs at about 1 kHz, from which point the waveguide gives -6 dB at +/- 45 deg.
To match this directivity, the bass driver would need to be 15" dia.
Two 7" or 8" drivers next to each other would match this, and allow a simpler XO.
I would also prefer tighter control than 90 deg. Does a smaller angle mean the waveguide has to be deeper because it still needs to have the same mouth diameter?
Thanks
http://www.4sptech.com/version2/index.html
I'm really sold on waveguides.
I'm still not clear on some of the basics.
Are waveguides a subset of horns, following horn theory?
Why is it that waveguides are mostly concave, while horns are convex?
SP Tech crosses theirs at about 1 kHz, from which point the waveguide gives -6 dB at +/- 45 deg.
To match this directivity, the bass driver would need to be 15" dia.
Two 7" or 8" drivers next to each other would match this, and allow a simpler XO.
I would also prefer tighter control than 90 deg. Does a smaller angle mean the waveguide has to be deeper because it still needs to have the same mouth diameter?
Thanks
Greets!
WGs are primarily about controlling directivity while horns are more about acoustic gain. Good horn design is a combination of the two, which nets you a CD horn.
Yes, they follow basic conical horn expansion theory except with the caveats of needing a flared mouth and usually a transitional throat.
Close enough if you mean effective piston diameter, so a really 'fast' 18" would be required if a single DR is used.
Yep, over the decades several manufacturers have used this layout for various studio monitors for these reasons.
Yes, but not for this reason. For example, at 1 kHz the mouth width to control it at 90 deg would need to be at least ~ 10^6/(1000/90) = ~11.11" wide and ~16.67" wide/60 deg, ergo the horn must be much longer for a given throat area (St) since it's a narrower angle expanding to a larger mouth (Sm). Plus there's the termination, which increases with increasing mouth dims, just like a flared vent.
Another thing to consider is critical distance (Dc) in that the narrower the angle, the longer the horn's 'throw', so if you're sitting inside it the horn's rolling off slower than the DR, ergo the DR will need greater gain to get a good blend between them at the listening position. Since accurately calcing 'Dc' is at best time consuming/tedious, measuring them in-room to find how much gain or attenuation is required (if any) is what I did.
Last, but not least, DR tweeters can't handle much loading (at least none I tried it on), so I wouldn't waste any time designing a speaker until you've proven that the desired tweeter can handle the loading with low distortion at its rated power without rolling off its extreme HF.
GM
WGs are primarily about controlling directivity while horns are more about acoustic gain. Good horn design is a combination of the two, which nets you a CD horn.
Yes, they follow basic conical horn expansion theory except with the caveats of needing a flared mouth and usually a transitional throat.
Close enough if you mean effective piston diameter, so a really 'fast' 18" would be required if a single DR is used.
Yep, over the decades several manufacturers have used this layout for various studio monitors for these reasons.
Yes, but not for this reason. For example, at 1 kHz the mouth width to control it at 90 deg would need to be at least ~ 10^6/(1000/90) = ~11.11" wide and ~16.67" wide/60 deg, ergo the horn must be much longer for a given throat area (St) since it's a narrower angle expanding to a larger mouth (Sm). Plus there's the termination, which increases with increasing mouth dims, just like a flared vent.
Another thing to consider is critical distance (Dc) in that the narrower the angle, the longer the horn's 'throw', so if you're sitting inside it the horn's rolling off slower than the DR, ergo the DR will need greater gain to get a good blend between them at the listening position. Since accurately calcing 'Dc' is at best time consuming/tedious, measuring them in-room to find how much gain or attenuation is required (if any) is what I did.
Last, but not least, DR tweeters can't handle much loading (at least none I tried it on), so I wouldn't waste any time designing a speaker until you've proven that the desired tweeter can handle the loading with low distortion at its rated power without rolling off its extreme HF.
GM
Zaph said:
Very cool - I'm quite interested in how these perform. Do you have any idea whether they follow a defined profile (ie Oblate spheroidal), or are they simply 'smooth'?
These are cheap enough that it's probably worth picking up a few different sizes for experimenting. Too bad they don't do a 15" version - it'd be ideal for a quick n dirty Summa clone.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.