Here's what the lens looked like a week ago. Lots of sharp edges, and sharp edges cause diffraction. So does symmetry. A circle with a sharp edge creates loads of diffraction... and that's what my lens is (at the moment.)
Here's some things I did to reduce diffraction. First, I filled in the gaps behind the lens. Those gaps will "collect" sound and create dips and peaks in the response. The next thing to do is to eliminate the sharp edges in the lens. I did that by creating a seamless transition from the lens to the baffle. I did this with duct tape, which I then fiberglassed. You could make it look better if you used pantyhose. But I didn't have any around, and was too lazy to go to the supermarket to buy some.
Once the glass had dried I put some wood filler on it and I filed it down.
I'll probably put another layer of glass on it.
Here's a measurement from last week. This is a polar measurement. Red is on-axis, green is 45 degrees off axis, blue is 90 degrees off axis. IE, it's the listening axis.
Here'a measurement from yesterday. This is NOT a polar measurement. All of the curves are measured 90 degrees off axis. (IE, they're measured where I'll be listening, which is ninety degrees off axis.) In this measurement the red line is the frequency response with the changes I made to the lens, the ones that reduced the sharp edges. (As illustrated in my post from earlier today.) The blue line is the same lens at the same angle, but with pressed felt applied near the lens. Similar to what Dunlavy did with their tweeters. The green curve is with the lens firing THROUGH a whole inch of closed cell foam. I basically cut out a 1" thick hunk of foam and covered the entire bottom platter with it. Similar to what Geddes does with his speakers, but with closed cell foam instead of reticulated foam. We can see that the foam really smooths out the response, so this would indicate that there are both HOMs and diffraction in the Beolab lens. IE, if the foam didn't have an effect, that would indicate that there were no real diffraction issues, IE it wasn't fixing anything. But it DOES smooth the response, which would seem to confirm diffraction and HOMs. I am not going to use closed cell foam, it's too dense, but I *will* cut out some reticulated foam.
Here's some things I notice:
- enlarging the midrange taps has had the desired effect; it's filled in the hole at 1500hz. That hole is there because the midrange can't play high enough, and the compression driver can't go much lower.
- The things I did to reduce diffraction are working. The high frequency response is starting to smooth out.
- On the down side, we're seeing diffraction effects on the midrange. For instance, note the dips at 280 and 600hz.
The ideal solution to the diffraction problem would be to eliminate all sharp angles entirely.
An externally hosted image should be here but it was not working when we last tested it.
That's why the Beolabs look like this.
One thing I noticed subjectively, is that the speakers sound *significantly* more neutral, and image better, when you listen above the lens. The sound almost seems to "float" above the speaker, which is a very cool illusion.
I did some sims in Hornresp, and it appears that the shape of the lens causes the lobe to be aimed up about 30 degrees. So I may start measuring the speaker *above* the lens. The lens may work better there. (And that might explain some of the peaks and dips, this may be comb filtering artifacts, similar to what you'd get if you measured a conventional speaker 45 degrees off axis.)
Last edited:
By the way, this thing sounds f'ing fantastic. It throws a huge sphere of sound that seems to stretch from the left and to the right. And the authority and dynamics of listening to a B&C woofer and a BMS tweeter at a distance of less than a meter are remarkable. When I heard the "real" Beolab speakers, I found their imaging to be very diffuse. But this thing is diffuse *and* pinpoint; the ambience is huge, but when there's an instrument in the mix I can practically point it out in the soundstage. I believe this is happening because the ratio of direct to reflected sound is very high, due to the use of acoustic lenses, and listening at such a close distance (less than one meter from the lens.)
Basically don't let the blips in the frequency response scare you away; this thing works.
Basically don't let the blips in the frequency response scare you away; this thing works.
Soldermizer's shameless knock-off of Bateman's...
Have assembled a working prototype now. Instead of frying the driver with the NU3000 I am using a Kenwood KR-V7070 receiver a neighbor donated to me (and it works, after some power on crackles.) Here is a photo of my current test rig. The husk of the former Frankenclosure sub serves as a table, awaiting a bonfire opportunity. The source is an Android tablet. Patrick, thanks for the ambiophonics links. I've found an Android app called Neutron (music player) that has built-in Ambiophonics. 5-day free trial.
With the set-up as pictured, there is an audible difference with the Ambiophonics engaged. Right now I am listening to just plain stereo and the sound is quite spacious. Listening to older tunes (Jazz, 1960s) helps...more separation in the 4 and 8 track recorder days, I believe.
It's amazing what can be done with crappy auto speakers (3-way Scosche 5.25" from the local Sprawl-Mart) and trash! As noted above (?) the enclosures are a pair of kitty litter boxes, stuffed with rags. I don't know if it makes any improvement, but I've improvised some very low-effort "lenses" out of respectively, the cap of a super glue container (tweeter) and plastic bottle lid found in drawer (mid), each cut in half.
Now, the only problem is how to camouflage this appalling creation 🙂
Have assembled a working prototype now. Instead of frying the driver with the NU3000 I am using a Kenwood KR-V7070 receiver a neighbor donated to me (and it works, after some power on crackles.) Here is a photo of my current test rig. The husk of the former Frankenclosure sub serves as a table, awaiting a bonfire opportunity. The source is an Android tablet. Patrick, thanks for the ambiophonics links. I've found an Android app called Neutron (music player) that has built-in Ambiophonics. 5-day free trial.
With the set-up as pictured, there is an audible difference with the Ambiophonics engaged. Right now I am listening to just plain stereo and the sound is quite spacious. Listening to older tunes (Jazz, 1960s) helps...more separation in the 4 and 8 track recorder days, I believe.
It's amazing what can be done with crappy auto speakers (3-way Scosche 5.25" from the local Sprawl-Mart) and trash! As noted above (?) the enclosures are a pair of kitty litter boxes, stuffed with rags. I don't know if it makes any improvement, but I've improvised some very low-effort "lenses" out of respectively, the cap of a super glue container (tweeter) and plastic bottle lid found in drawer (mid), each cut in half.
Now, the only problem is how to camouflage this appalling creation 🙂
Attachments
As usual, I'm totally ADHD, so I am taking a second crack at this, but with Synergy horns this time.
Because my Synergy horns only play down to 360hz, I need something to fill in the two octaves from 90hz to 360hz.
The midbass is designed to fit into an Ikea Expedit.
I'm actually sitting on a subwoofer, so this box goes to my left and to my right and it's designed to do midbass duty between my sub and my Synergy horns.
Here's the design. It's a clone of the Danley BDEAP, but shrunk to fit into an Ikea Expedit. The drivers are a Dayton ND91. The horn is tuned to 180hz. The sealed part of the enclosure is 4" PVC that's been sliced in half. The wood is 1/2". You can use MDF or plywood.
I got started by using Eazy Horn. These are the params it gave me.
Here's the hornresp parameters
Here's a comparison of the output of the horn with two woofers, versus a bandpass box with four woofers. The bandpass box plays lower and is smoother, but the horn is louder. The horn sim shows some nasty peaks. In my experience, I've found that hornresp exaggerates peaks. So hopefully it's not that bad IRL. Also, I would like to use at least two per side, for midbass impact, and arraying the horns will smooth the response. The BDEAP is designed to be arrayed.
Because my Synergy horns only play down to 360hz, I need something to fill in the two octaves from 90hz to 360hz.
The midbass is designed to fit into an Ikea Expedit.
I'm actually sitting on a subwoofer, so this box goes to my left and to my right and it's designed to do midbass duty between my sub and my Synergy horns.
Here's the design. It's a clone of the Danley BDEAP, but shrunk to fit into an Ikea Expedit. The drivers are a Dayton ND91. The horn is tuned to 180hz. The sealed part of the enclosure is 4" PVC that's been sliced in half. The wood is 1/2". You can use MDF or plywood.
I got started by using Eazy Horn. These are the params it gave me.
Here's the hornresp parameters
Here's a comparison of the output of the horn with two woofers, versus a bandpass box with four woofers. The bandpass box plays lower and is smoother, but the horn is louder. The horn sim shows some nasty peaks. In my experience, I've found that hornresp exaggerates peaks. So hopefully it's not that bad IRL. Also, I would like to use at least two per side, for midbass impact, and arraying the horns will smooth the response. The BDEAP is designed to be arrayed.
Attention Span Salad
My kitty litter speaker is now painted blue. Meanwhile, the Kenwood receiver after working ok for an hour or so (both pre and post my cleaning) decided to pitch a fit (noise in one channel) so it was set out with the trash this morning 🙂
I had to look up Expedit. Are you using a specific cabinet? it looks like primarily shelving. How did you get an enclosure of this??? I know there is an old thread about building speakers with Ikea cabinets (not the Bateman one!) but using real wood or whatever they make the kitchen out of. What is your current box made of?
Why can't IKEA just make and sell a series of subwoofer kits? We know they'd be flat-packs. And Baltic Birch would be no problem, for crying out loud, Sweden is right around the Baltic Sea or whatever.
Meanwhile, re ADHD, something I have been diagnosed with, here is a cat's point of view:
My kitty litter speaker is now painted blue. Meanwhile, the Kenwood receiver after working ok for an hour or so (both pre and post my cleaning) decided to pitch a fit (noise in one channel) so it was set out with the trash this morning 🙂
I had to look up Expedit. Are you using a specific cabinet? it looks like primarily shelving. How did you get an enclosure of this??? I know there is an old thread about building speakers with Ikea cabinets (not the Bateman one!) but using real wood or whatever they make the kitchen out of. What is your current box made of?
Why can't IKEA just make and sell a series of subwoofer kits? We know they'd be flat-packs. And Baltic Birch would be no problem, for crying out loud, Sweden is right around the Baltic Sea or whatever.
Meanwhile, re ADHD, something I have been diagnosed with, here is a cat's point of view:
Attachments
I tried Ambiophonics with my Gedlee Summas but it didn't work as well. My hypothesis is that this is due to the center-to-center spacing. Basically crosstalk cancellation relies on timing differences that are under a single millisecond. So my ambiopole isn't as smooth as my Summas, but it's better suited to ambiophonics because the midrange and the tweeter are so tightly spaced.
John
Your experience is the exact opposite of the Princeton Professor who was using Geddes speakers in his cross-talk cancelling experiments because he claimed that the high directivity made the effect work better than the other speakers that he had tried. And remember that with about an 800 Hz crossover there is no effect in the critical range from the displaced drivers. One more thing- why is there any timing differences when one is in the median plane?
Maybe a little Beranek effect going on here.
Your experience is the exact opposite of the Princeton Professor who was using Geddes speakers in his cross-talk cancelling experiments because he claimed that the high directivity made the effect work better than the other speakers that he had tried. And remember that with about an 800 Hz crossover there is no effect in the critical range from the displaced drivers. One more thing- why is there any timing differences when one is in the median plane?
Maybe a little Beranek effect going on here.
Last edited:
Could it be that the near field listening position (assuming a simple L,R stereo signal w/o any processing) is "merely" giving a much stronger signal to the correct ear (e.g. L speakers to L ear, etc.) which is another way of saying the crosstalk signal to the other ear is for whatever reasons rejected better by the brain as compared to a more conventional stereo set-up? If nothing else, with my extra-cheap set-up I hear good separation and spaciousness. Based on where the box was set up, each ear should be getting a strong direct signal from its proper speaker and much attenuated from the farther one. Furthermore, any secondary reflections from walls and floor are feet instead of inches away.
P.S. I have upon a (sort of) recommendation here watched "American Psycho" so I have a glimpse into our Patrick's fictional namesake. Now, I must be going, I have to return some video tapes...
P.S. I have upon a (sort of) recommendation here watched "American Psycho" so I have a glimpse into our Patrick's fictional namesake. Now, I must be going, I have to return some video tapes...
This thread was getting lonely...
I recently bought my first class-D amp (board), a TDA7492 for a whopping $12 or so. Hooked it up using a spare PC power supply (thanks to this forum for advising I could do so). I have spares so why not (ab-)use one of them? Amazingly i was able to hook stuff up and soldermize connections and it works out of the box. I am using the "kitty litter" speakers currently.
I recently bought my first class-D amp (board), a TDA7492 for a whopping $12 or so. Hooked it up using a spare PC power supply (thanks to this forum for advising I could do so). I have spares so why not (ab-)use one of them? Amazingly i was able to hook stuff up and soldermize connections and it works out of the box. I am using the "kitty litter" speakers currently.
I've been keeping a fairly low profile on the forums lately. If anyone's curious why, here's what happened:
Back in 2006, I started working from home. I've been doing that for eleven years now. For the first few years, I traveled a LOT. But I switched jobs (still WFH) and the amount of travel basically ground to a halt in 2008. I had a good run there - nearly nine years of not going into an office. But I got laid off last May. The new gig has me traveling about 3-4X as much as before.
So I need some portable tunes.
I generally fly Southwest Airlines. The maximum carry on is 10"x16"x24". That should be plenty big enough for a Boombox Bateman Style. (Basically I'll put my suitcase in the cargo hold, and carry on my Boombox.)
The amp I bought is this Sabaj A2. It's $68 on Amazon. I originally intended to build an amp, using a Parts Express kit. But this Sabaj amp costs about the same, and it's assembled. This amp has some features I like for a boombox:
1) it has APTX Bluetooth, for the best possible bluetooth sound
2) it's small
3) it's relatively loud - 50w x 2
4) I can plug it straight into a car, with no DC-DC conversion
5) I could potentially run the whole thing off batteries.
6) It uses the TPA3116, which is reputedly quite good : First Timer. Anything better than TPA3116?
As a bonus, it also has a remote, though I likely won't use it.
Back in 2006, I started working from home. I've been doing that for eleven years now. For the first few years, I traveled a LOT. But I switched jobs (still WFH) and the amount of travel basically ground to a halt in 2008. I had a good run there - nearly nine years of not going into an office. But I got laid off last May. The new gig has me traveling about 3-4X as much as before.
So I need some portable tunes.
I generally fly Southwest Airlines. The maximum carry on is 10"x16"x24". That should be plenty big enough for a Boombox Bateman Style. (Basically I'll put my suitcase in the cargo hold, and carry on my Boombox.)
The amp I bought is this Sabaj A2. It's $68 on Amazon. I originally intended to build an amp, using a Parts Express kit. But this Sabaj amp costs about the same, and it's assembled. This amp has some features I like for a boombox:
1) it has APTX Bluetooth, for the best possible bluetooth sound
2) it's small
3) it's relatively loud - 50w x 2
4) I can plug it straight into a car, with no DC-DC conversion
5) I could potentially run the whole thing off batteries.
6) It uses the TPA3116, which is reputedly quite good : First Timer. Anything better than TPA3116?
As a bonus, it also has a remote, though I likely won't use it.
Patrick,
Wishing you all the best with your new gig - and hope with the hard yards that you're putting in now, it leaves you some time to focus on more DIY projects in the near future.
You got a portable synergy horn yet? 😀
Wishing you all the best with your new gig - and hope with the hard yards that you're putting in now, it leaves you some time to focus on more DIY projects in the near future.
You got a portable synergy horn yet? 😀
I tried the Beolab style acoustic lens probably 10 years ago. The response was almost perfectly matched within the 180deg range. The diffraction was what caused me to stop, plus the Beolabs sounded terrible. But now with DSP of Bodzio Software “Ultimate Equalizer” it seems to worth another look.
I'm going to use an Adire Audio Extremis for the woofer section. I consider quite a few drivers. An MCM 55-2421 works nicely, but it weighs too much. A B&C 8NDL51 weighs about the same, goes louder, but doesn't play low enough for a woofer. I considered using a pair of Dayton ND91s, but all of mine seem to be disintegrating. (This is an issue I've been having with all of my woofers with rubber surrounds.)
Here's the published specs versus what I measured. Dan Wiggins sold me my Extremis at one of his BBQs over a decade ago, and the parameters have probably drifted quite a bit due to age.
published / measured
fs = 30Hz / 45Hz
re = 7 / 7.6
vas = 33.26 / ?
qes = .383 / 0.73
qms = 2.54 / 1.8
qts = .332 / 0.52
Here's the published specs versus what I measured. Dan Wiggins sold me my Extremis at one of his BBQs over a decade ago, and the parameters have probably drifted quite a bit due to age.
published / measured
fs = 30Hz / 45Hz
re = 7 / 7.6
vas = 33.26 / ?
qes = .383 / 0.73
qms = 2.54 / 1.8
qts = .332 / 0.52
I'm really REALLY getting sick of AuraSound drivers and their Dayton clones. I probably have around 30-40 of them. Literally 75% of the ones with rubber sounds have cracked. You would think they could figure out how to make a woofer that doesn't spontaneously die, while sitting on a shelf.
My Whispers are fine, but they use a foam surround. All the rest of them, it's rubber. This is super annoying.
I'd originally intended to use a pair of AuraSound NS4s for this project. (I opted for that instead of the Adire Extremis.)
The idea is to extend the directivity control. The way that this works is that when you have two drivers and you sloooooowly spread them apart by more than 1/4WL, you get an interference pattern. This interference pattern narrows the beamwidth. You can see this effect using Basta's XDIR or John K's ARPE software.
So sticking a couple of woofers on the edge of the waveguide can yield a pattern that's narrower than what you would get with a Woofer-Tweeter array like the Lambda Unity Horn or the Yorkville Unity Horn. Basically narrow directivity in a relatively compact package.
A pair of NS4s would get me a cutoff of about 42Hz, along with narrow directivity, in a package that's small enough to fit in an overhead bin on an airline.
Madisound Speaker Components
I considered a bunch of options instead of the AuraSound. Everything from the Adire Extremis (I only have one) to the 8NDL51 (there's no room for two.) In the end, I settled on a pair of SB Acoustics 5" woofers. They're nicely suited for vented boxes. They're not neodymium, but the frame is plastic, which keeps the weight down.
In the end, the SB Acoustics has flatter response, about the same F3, and a little bit more output due to higher efficiency. On the downside, they're a bit heavier.
- Status
- Not open for further replies.