Hi Everyone,
I've spent the last few months researching drivers, alignments, bass loading and several other things, and I'm finally far enough along to start this thread.
This is where I'll document the design process, the construction process, and hopefully soon, the finished result.
I think a little background is in order first...
My current set of speakers is a very nice pair of Lynn Olsen's Ariels. These took a few years to build, and several more years to tweak to their current status. I love them, but I've been longing for the much larger and more dynamic sound that only a larger pair of speakers can offer.
The Ariels are reasonably efficient, at about 92dB but I'd like that figure to go up. They're also light on bass, and for some reason, they're quite difficult to integrate with a subwoofer. My last few attempts have ultimately ended in me not using a subwoofer at all.
The desire for "big" speakers started several years ago when I was lucky enough to come across three Altec "Voice of the Theater" speakers. I used two of them in my main system for a few months, and then sold all three because they were so bloody big. During those months though, I became completely obsessed with the incredible dynamics and "huge" sound that they were able to produce. The bass and midbass, although not incredibly low, was so powerful that it just captivated the listener, and demanded your utmost attention. Unfortunately, the rather mediocre horns turned out to be a let-down, and ultimately resulted in a rather fatiguing speaker.
What I hope to achieve with these new speakers is the best of both worlds. I want the detail and easy listening qualities of the Ariels, along with the dynamics and presence of the VOtT's.
So far, here's what I've got:
- NeoPro5i Ribbons.
- Either AE Lambda TD10M or B&C 10NW64 midbass.
- MTM arrangement.
- Independent transmission lines for each 10" midbass.
- 1600Hz - 2200Hz crossover point.
I want very high efficiency, so the NeoPro5i is pretty much set in stone. I've listened to and measured three different compression drivers in a variety of horns and waveguides, and I didn't like any of them. The NeoPro ribbons both measure and sound better than anything I can find in the 100+ dB range.
The midbass drivers are still a toss up. I've ordered 4 of both, and I plan on auditioning both to see which of the two is best. The B&C drivers fit in a smaller TL than the TD10M drivers do, but they're also tuned 5Hz higher at 50Hz in place of 45Hz. I'm seriously concerned about the claims of the TD10M drivers, as no FR graphs can be found anywhere. I'll reserve judgment until I can measure them properly and compare them to the B&C drivers. I'm hoping they're as good as claimed.
The crossover point is a little up in the air as it's a trade-off between the increased beaming of the 10" driver at higher frequencies, and the increased distortion of the ribbon tweeter at lower frequencies. I'm pretty sure that only listening to both extremes will tell me which of those drawbacks is easier to live with. Like all things, the final compromise will probably fall somewhere in between.
The TL loading is also set in stone, as it's by far my favorite bass loading design. The taper for both drivers will be 0.2 and the lines were designed using MJK's TL worksheets. The TD10M's seem to work best in a 45Hz line, and the B&C drivers seem to need a 50Hz line. I'm not looking for huge deep bass, but I do want these to integrate well with a larger format (18" or larger) subwoofer in the future. I also need them to stand their own ground above 40Hz so they can be used as a true standalone speaker without ever needing a sub.
The finished efficiency of the speaker will hopefully be right around 100dB and should reach easily down to 35-40Hz and up to 40kHz.
I'll welcome any suggestions, comments or criticisms, and I hope to hear from anyone who has any previous experience with either of the 10" drivers mentioned above.
I'll attach a few details over the next few posts, but for now, let's start with the box design for the TD10M midbass drivers.
Cheers,
Owen
I've spent the last few months researching drivers, alignments, bass loading and several other things, and I'm finally far enough along to start this thread.
This is where I'll document the design process, the construction process, and hopefully soon, the finished result.
I think a little background is in order first...
My current set of speakers is a very nice pair of Lynn Olsen's Ariels. These took a few years to build, and several more years to tweak to their current status. I love them, but I've been longing for the much larger and more dynamic sound that only a larger pair of speakers can offer.
The Ariels are reasonably efficient, at about 92dB but I'd like that figure to go up. They're also light on bass, and for some reason, they're quite difficult to integrate with a subwoofer. My last few attempts have ultimately ended in me not using a subwoofer at all.
The desire for "big" speakers started several years ago when I was lucky enough to come across three Altec "Voice of the Theater" speakers. I used two of them in my main system for a few months, and then sold all three because they were so bloody big. During those months though, I became completely obsessed with the incredible dynamics and "huge" sound that they were able to produce. The bass and midbass, although not incredibly low, was so powerful that it just captivated the listener, and demanded your utmost attention. Unfortunately, the rather mediocre horns turned out to be a let-down, and ultimately resulted in a rather fatiguing speaker.
What I hope to achieve with these new speakers is the best of both worlds. I want the detail and easy listening qualities of the Ariels, along with the dynamics and presence of the VOtT's.
So far, here's what I've got:
- NeoPro5i Ribbons.
- Either AE Lambda TD10M or B&C 10NW64 midbass.
- MTM arrangement.
- Independent transmission lines for each 10" midbass.
- 1600Hz - 2200Hz crossover point.
I want very high efficiency, so the NeoPro5i is pretty much set in stone. I've listened to and measured three different compression drivers in a variety of horns and waveguides, and I didn't like any of them. The NeoPro ribbons both measure and sound better than anything I can find in the 100+ dB range.
The midbass drivers are still a toss up. I've ordered 4 of both, and I plan on auditioning both to see which of the two is best. The B&C drivers fit in a smaller TL than the TD10M drivers do, but they're also tuned 5Hz higher at 50Hz in place of 45Hz. I'm seriously concerned about the claims of the TD10M drivers, as no FR graphs can be found anywhere. I'll reserve judgment until I can measure them properly and compare them to the B&C drivers. I'm hoping they're as good as claimed.
The crossover point is a little up in the air as it's a trade-off between the increased beaming of the 10" driver at higher frequencies, and the increased distortion of the ribbon tweeter at lower frequencies. I'm pretty sure that only listening to both extremes will tell me which of those drawbacks is easier to live with. Like all things, the final compromise will probably fall somewhere in between.
The TL loading is also set in stone, as it's by far my favorite bass loading design. The taper for both drivers will be 0.2 and the lines were designed using MJK's TL worksheets. The TD10M's seem to work best in a 45Hz line, and the B&C drivers seem to need a 50Hz line. I'm not looking for huge deep bass, but I do want these to integrate well with a larger format (18" or larger) subwoofer in the future. I also need them to stand their own ground above 40Hz so they can be used as a true standalone speaker without ever needing a sub.
The finished efficiency of the speaker will hopefully be right around 100dB and should reach easily down to 35-40Hz and up to 40kHz.
I'll welcome any suggestions, comments or criticisms, and I hope to hear from anyone who has any previous experience with either of the 10" drivers mentioned above.
I'll attach a few details over the next few posts, but for now, let's start with the box design for the TD10M midbass drivers.
Cheers,
Owen
A 3-way with the Neopro and Lambda 10" + 15" looks attractive. Avoids MTM spacing issues with 10" midbass, allows sealed box midrange, and 25Hz deep bass from mass loaded transmission line with a port on the bottom of a raised-leg cabinet.
Neopro5i ribbon LR4 at 1,400
AE Lambda 10M in sealed well stuffed and braced box LR4 at 80Hz
----read AE Lambda Midbass 10 Project? thread
AE Lambda 15S or 15H in ML TL down to 25Hz www.quarter-wave.com
Neopro5i ribbon LR4 at 1,400
AE Lambda 10M in sealed well stuffed and braced box LR4 at 80Hz
----read AE Lambda Midbass 10 Project? thread
AE Lambda 15S or 15H in ML TL down to 25Hz www.quarter-wave.com
Hi Guys,
Thanks for the early comments, very much appreciated!
Glen,
The spacing of the 10" drivers is a pretty big concern for me, but I'm going to build up a mock baffle and measure it with all the drivers in place to see how bad it really is. Apparently with the very limited vertical response of the ribbon, there's not much to worry about as far as ribbon to midbass interaction goes, but between the two 10" drivers is another concern.
As of right now, it looks like C-C spacing of the 10" drivers is going to be about 18.25" with the B&C drivers. the TD drivers will be a little farther apart.
Infinia,
I've attached a .pdf of the tweeter measured in the baffle I'm planning to use, with flat plates over the midbass holes. No smoothing, and measured at a 1m distance. These are gated measurements, so ignore anything below 500Hz and above 20k (microphone limit)
The ripples in the midband are either from the tweeter not being flush mounted, or the sharp edges on the mock baffle. In the finished speaker these would at least be rounded over, and the tweeter would be offset and flush mounted to mitigate the impact.
Off axis response is really quite good for the tweeter, so it looks like I'll have to see how the midbass units do before making the call.
LineSource,
I've been following a few of the threads with the AE drivers very carefully. Part of me keeps hoping for someone to post some measurements, but there never are any for the TD10M.
I see where you're coming from, but I'm not sure it's quite my cup of tea. You'd have to pad the tweeter down by over 6dB to get it to the level of a single 10", then even worse you'd have to pad down the 15" as well. Your final system would sit at a little less than 94dB and it would be larger than the MTM I'm looking at.
If I find this light on bass, then I'd probably look into a pair of B&C 21" drivers in folded TL's to fill in strictly from 40Hz down to 10Hz or so. I'm not really all that bent on getting response that low, as most of what I listen to doesn't have much going on down there. Just the same, this is kinda cost-no-object, so I'll see where it goes.
Keep it coming!
Owen
Thanks for the early comments, very much appreciated!
Glen,
The spacing of the 10" drivers is a pretty big concern for me, but I'm going to build up a mock baffle and measure it with all the drivers in place to see how bad it really is. Apparently with the very limited vertical response of the ribbon, there's not much to worry about as far as ribbon to midbass interaction goes, but between the two 10" drivers is another concern.
As of right now, it looks like C-C spacing of the 10" drivers is going to be about 18.25" with the B&C drivers. the TD drivers will be a little farther apart.
Infinia,
I've attached a .pdf of the tweeter measured in the baffle I'm planning to use, with flat plates over the midbass holes. No smoothing, and measured at a 1m distance. These are gated measurements, so ignore anything below 500Hz and above 20k (microphone limit)
The ripples in the midband are either from the tweeter not being flush mounted, or the sharp edges on the mock baffle. In the finished speaker these would at least be rounded over, and the tweeter would be offset and flush mounted to mitigate the impact.
Off axis response is really quite good for the tweeter, so it looks like I'll have to see how the midbass units do before making the call.
LineSource,
I've been following a few of the threads with the AE drivers very carefully. Part of me keeps hoping for someone to post some measurements, but there never are any for the TD10M.
I see where you're coming from, but I'm not sure it's quite my cup of tea. You'd have to pad the tweeter down by over 6dB to get it to the level of a single 10", then even worse you'd have to pad down the 15" as well. Your final system would sit at a little less than 94dB and it would be larger than the MTM I'm looking at.
If I find this light on bass, then I'd probably look into a pair of B&C 21" drivers in folded TL's to fill in strictly from 40Hz down to 10Hz or so. I'm not really all that bent on getting response that low, as most of what I listen to doesn't have much going on down there. Just the same, this is kinda cost-no-object, so I'll see where it goes.
Keep it coming!
Owen
I should add:
Black trace = On Axis
Red trace = 30 degrees horizontal
Blue trace = 60 degrees horizontal
There's no scaling, but it looks like about 100dB @2.83V and 2kHz with a decent SPL meter at 1m away.
I've attached a modeled tweeter response with a 1800Hz 3rd order crossover and a little correction on that rise between 8k and 15k
I'm currently building this crossover for testing... results will follow.
Cheers,
Owen
Black trace = On Axis
Red trace = 30 degrees horizontal
Blue trace = 60 degrees horizontal
There's no scaling, but it looks like about 100dB @2.83V and 2kHz with a decent SPL meter at 1m away.
I've attached a modeled tweeter response with a 1800Hz 3rd order crossover and a little correction on that rise between 8k and 15k
I'm currently building this crossover for testing... results will follow.
Cheers,
Owen
opc said:
Hi
Thanks for the off axis data. I wonder why the large peak in response.. it almost looks similar to breakup in a cone. How did you tame it? Seems it would take a major correction to get your final response.
Edit> I doubt the peak is breakup related. My speculation.. mostly because off axis follows on axis closely. Is there a resonance nearby on the raw impedance charts?
Hi Guys,
Brucemck2:
The high efficiency thing is really a big part of what I'm looking for, so it was the first thing I checked when I got those ribbons. I'm not sure if there was some differences between production runs, but mine are absolutely between 100 and 101 dB at a one meter distance with a 2.83 volt RMS sine wave at 2kHz. The SPL meter was flickering between 100 and 101, and it's an accurate enough meter to be trusted. You can use that to scale the graphs, and you'll see they're definitely at or above 100dB for the whole range.
If the efficiency thing was less of a big deal, I would have gone with RAAL ribbons, but at 94dB, they're not quite there.
infinia:
That peak is present on the published datasheet, but the one I measured is a good 2 or 3 dB higher in level relative to the rest of the response. I don't think it's a resonance as the impedance is perfect in that region. I've attached the raw impedance data, and I measured this three or four times, along with re-measuring a few known drivers to be sure. My impedance measurement correlates quite well with the published spec, except they cut their graph off at 600Hz, which is where mine gets a little confusing.
It looks like the actual ribbon resonance is the little bump between 700 and 1K, but I have no idea what the 40 ohm spike is at 200Hz. I had the ribbon directly connected to the impedance jig, so there was no cap or anything else between to cause that.
I'm guessing it has something to do with the transformer, but I really don't know.
Fixing the bump in the FR just requires a parallel notch with a wide band. The simulated values are 60uH, 3.2uF, and 10 ohms. I'll be testing them in the next few days with the filter in place.
Cheers,
Owen
Brucemck2:
The high efficiency thing is really a big part of what I'm looking for, so it was the first thing I checked when I got those ribbons. I'm not sure if there was some differences between production runs, but mine are absolutely between 100 and 101 dB at a one meter distance with a 2.83 volt RMS sine wave at 2kHz. The SPL meter was flickering between 100 and 101, and it's an accurate enough meter to be trusted. You can use that to scale the graphs, and you'll see they're definitely at or above 100dB for the whole range.
If the efficiency thing was less of a big deal, I would have gone with RAAL ribbons, but at 94dB, they're not quite there.
infinia:
That peak is present on the published datasheet, but the one I measured is a good 2 or 3 dB higher in level relative to the rest of the response. I don't think it's a resonance as the impedance is perfect in that region. I've attached the raw impedance data, and I measured this three or four times, along with re-measuring a few known drivers to be sure. My impedance measurement correlates quite well with the published spec, except they cut their graph off at 600Hz, which is where mine gets a little confusing.
It looks like the actual ribbon resonance is the little bump between 700 and 1K, but I have no idea what the 40 ohm spike is at 200Hz. I had the ribbon directly connected to the impedance jig, so there was no cap or anything else between to cause that.
I'm guessing it has something to do with the transformer, but I really don't know.
Fixing the bump in the FR just requires a parallel notch with a wide band. The simulated values are 60uH, 3.2uF, and 10 ohms. I'll be testing them in the next few days with the filter in place.
Cheers,
Owen
Hi Guys,
Just a quick update... the B&C 10" drivers arrived today, and initial impressions are very good!
I haven't had time to measure these yet, but that's what I'll be doing over the weekend. I have listened to them with a rudimentary crossover and I was very impressed. Looks like they're plenty efficient to keep up with the tweeter, so I think the finished efficiency will actually be 100dB if all goes well.
I have attached a picture of the drivers sitting on the floor, and one of the back side of the B&C driver.
They're quite well built, simple construction but excellent quality.
Cheers,
Owen
Just a quick update... the B&C 10" drivers arrived today, and initial impressions are very good!
I haven't had time to measure these yet, but that's what I'll be doing over the weekend. I have listened to them with a rudimentary crossover and I was very impressed. Looks like they're plenty efficient to keep up with the tweeter, so I think the finished efficiency will actually be 100dB if all goes well.
I have attached a picture of the drivers sitting on the floor, and one of the back side of the B&C driver.
They're quite well built, simple construction but excellent quality.
Cheers,
Owen
Attachments
High quality choices.
I'm considering something cheaper and OB.
BG neo3, open back with 2 Beyma 10BR60 as mids.
http://www.bmm-electronics.com/Product.asp?Product_ID=3632
Problem is the cross I had in mind would be 2000hz.
That's on the lower limit of the tweeter and about the upper limit of the mid.
Plus I can't find an 30 degree off axis graph of the mid. Or an independend distortion test.
And the lobbing thing is also a concern.
Visually I like mtm better but have you thought about mmt?
I'm considering something cheaper and OB.
BG neo3, open back with 2 Beyma 10BR60 as mids.
http://www.bmm-electronics.com/Product.asp?Product_ID=3632
Problem is the cross I had in mind would be 2000hz.
That's on the lower limit of the tweeter and about the upper limit of the mid.
Plus I can't find an 30 degree off axis graph of the mid. Or an independend distortion test.
And the lobbing thing is also a concern.
Visually I like mtm better but have you thought about mmt?
Hi Guys,
Things are progressing well with these, but I've been so busy over the past little while that I haven't posted an update.
I finally picked up my AE TD10M drivers last Wednesday, and I'll be taking some measurements today of both the B&C and the AE drivers.
I have listened to both now, and I have to say that just listening to them full range I think the AE drivers will do better. Although the B&C drivers do have the published Fs of about 51Hz, they just don't produce much below 300Hz, which isn't going to work for me.
I took some quick nearfield measurements of the B&C's to confirm this, and it looks like there's a 7dB drop at around 200Hz which isn't box related. I'll take some better measurements over the day today and post the results this evening for those interested.
I'm also considering going in a different direction with the box. I'm starting to lean away from the dual TL and towards a smaller sealed box for the 10" midranges, and then an 18" driver below each to fill in the bottom end. Either a Faital Pro 18FH500, or a Ciare NDC 18-4 S.
If I go with the TL, the box will be quite large, and it'll need a stand to sit on. I'm also having my doubts that either driver will produce good solid bottom end. So I think it may be best to replace the stand with a large sub, make the midrange boxes smaller and hopefully make the whole speaker smaller and more manageable.
Last time I checked, nobody has ever added an 18" driver to their speaker in order to make it smaller, but in this case I think it'll work.
I have attached a picture of a mock baffle I made to give you an idea of where I want to go with these. It would be made with a horizontal "Trans Lam" process where I'd layer 30 pieces of 3/4" Baltic Birch ply to get that shape all the way back. I'd also round over the entire from and back perimeter with a 1.25" radius.
The general shape still needs some work, but it'll be something similar to that. I think it'll be good for reducing diffraction, and hopefully it'll look a little more interesting than the usual square box. I'm still debating between using MDF and an automotive paint finish, or Baltic Birch with a natural finish
I'll welcome any thoughts or suggestions, and hopefully I'll have some measurements to post tonight.
Cheers,
Owen
Things are progressing well with these, but I've been so busy over the past little while that I haven't posted an update.
I finally picked up my AE TD10M drivers last Wednesday, and I'll be taking some measurements today of both the B&C and the AE drivers.
I have listened to both now, and I have to say that just listening to them full range I think the AE drivers will do better. Although the B&C drivers do have the published Fs of about 51Hz, they just don't produce much below 300Hz, which isn't going to work for me.
I took some quick nearfield measurements of the B&C's to confirm this, and it looks like there's a 7dB drop at around 200Hz which isn't box related. I'll take some better measurements over the day today and post the results this evening for those interested.
I'm also considering going in a different direction with the box. I'm starting to lean away from the dual TL and towards a smaller sealed box for the 10" midranges, and then an 18" driver below each to fill in the bottom end. Either a Faital Pro 18FH500, or a Ciare NDC 18-4 S.
If I go with the TL, the box will be quite large, and it'll need a stand to sit on. I'm also having my doubts that either driver will produce good solid bottom end. So I think it may be best to replace the stand with a large sub, make the midrange boxes smaller and hopefully make the whole speaker smaller and more manageable.
Last time I checked, nobody has ever added an 18" driver to their speaker in order to make it smaller, but in this case I think it'll work.
I have attached a picture of a mock baffle I made to give you an idea of where I want to go with these. It would be made with a horizontal "Trans Lam" process where I'd layer 30 pieces of 3/4" Baltic Birch ply to get that shape all the way back. I'd also round over the entire from and back perimeter with a 1.25" radius.
The general shape still needs some work, but it'll be something similar to that. I think it'll be good for reducing diffraction, and hopefully it'll look a little more interesting than the usual square box. I'm still debating between using MDF and an automotive paint finish, or Baltic Birch with a natural finish
I'll welcome any thoughts or suggestions, and hopefully I'll have some measurements to post tonight.
Cheers,
Owen
Attachments
Here's a quick shot of the AE TD10M next to the B&C. The AE driver dwarfs it to the point that you'd swear it was an 8" beside a 10", but they're both 10" drivers with nearly the same Sd.
Both will be going in the test baffle shortly. It's pretty daunting pulling all this stuff out of storage, what I wouldn't give for a dedicated lab where I could leave all my speaker stuff.
Cheers,
Owen
Both will be going in the test baffle shortly. It's pretty daunting pulling all this stuff out of storage, what I wouldn't give for a dedicated lab where I could leave all my speaker stuff.
Cheers,
Owen
Attachments
And the back... this is why I love Neodymium magnets. With both AE drivers and the NeoPro mounted in that baffle, I can hardly lift it. The B&C weighs 8lbs, the AE weighs 22lbs, both have the same BL value.
Another interesting thing I noticed while breaking in both drivers was a whooshing sound made by the B&C woofer at high excursion. I break the drivers in by playing a near full excursion 5Hz tone for 24 hours, and that's when I noticed it. The magnet has some sort of cooling system that forces air over the coil and out those holes on the back. When the driver moves enough, you can really hear it rushing through the holes. Not a good thing.
Another interesting thing I noticed while breaking in both drivers was a whooshing sound made by the B&C woofer at high excursion. I break the drivers in by playing a near full excursion 5Hz tone for 24 hours, and that's when I noticed it. The magnet has some sort of cooling system that forces air over the coil and out those holes on the back. When the driver moves enough, you can really hear it rushing through the holes. Not a good thing.
Attachments
Good stuff, keep it coming! Near and far responses of both would be very appreciated! Also try running an impedance sweep with the cone pushed in/out to about xmax, maybe 5mm or so for both. This is a very rough gauge of the motor linearity.
I don't know if I agree that the noise on the B&C is a terrible thing. If it's in a closed box-or even OB crossed high enough-you won't hear it, but what you will gain is arguably better power handling due to better motor cooling.
Strange bump in the impedance. Do you have another to test? How was the driver mounted when you did this test? Hopefully no sitting flat on a table with the vents covered up...
I don't know if I agree that the noise on the B&C is a terrible thing. If it's in a closed box-or even OB crossed high enough-you won't hear it, but what you will gain is arguably better power handling due to better motor cooling.
Strange bump in the impedance. Do you have another to test? How was the driver mounted when you did this test? Hopefully no sitting flat on a table with the vents covered up...
opc said:
I noticed this on the B&C 12 TBX100 doing exactly the same thing as you. Its a non issue IMO since I never heard a thing once I got it in a cabinet. I tanned that thing as well.
BTW Neopro 5i has rising distortion below 2Khz with a large peak at around 1.6Khz. So be conscious about where and how you XO. The RAAL measures better than Neopro5i in distortion at same levels regardless of sensitivity. I don't use it below 2.5Khz 2nd order. You might find you come to a similar conclusion. Despite these ribbons having lots of radiating area, they are behind domes for distortion at lower SPL but from measurements, RAAL handles 113dB @ 2m with only 2% THD using 2.5Khz 2nd order LR filter which a dome couldn't do.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.