Hi everyone,
I am planning to build a high efficiency active 3 way speaker with an open baffle mid.
I'm planning to use the CX3400 for crossover, with Fountek NeoCd2.0M ribbon tweeter and 4x8" peerless woofers per side from the NHT clearance. Although i am using an active setup, i would like to have similar sensitivities of the drivers of about 96dB.
I have narrowed down the choice of midrange driver between B&C 6MD38 and Eminence Alphalite 6A. I looked at the efficiency, Qts(for open baffle) and Xmax.
B&C 6MD38 | 96dB | Qts 0.44 | 2mm | 130Hz Fs
Eminence Alphalite 6A | 94dB | Qts 0.56 | 3.5mm | 126Hz Fs
I'm planning to cross the woofer to mid at about the BSC frequency of ~300Hz, and the mid-tweeter at about 4-5kHz depending on listening tests.
Although both drivers have similar Fs, the Eminence appears to go lower from the spec sheet chart (92dB at 100Hz vs 84dB), and has higher Xmax. Will both drivers be able to cross at ~300Hz?
Zaph has also reviewed the B&C quite favorably, but i couldn't find any reviews on the Eminence.
Any personal experience with these drivers, and recommendation for one over the other?
Many thanks!
I am planning to build a high efficiency active 3 way speaker with an open baffle mid.
I'm planning to use the CX3400 for crossover, with Fountek NeoCd2.0M ribbon tweeter and 4x8" peerless woofers per side from the NHT clearance. Although i am using an active setup, i would like to have similar sensitivities of the drivers of about 96dB.
I have narrowed down the choice of midrange driver between B&C 6MD38 and Eminence Alphalite 6A. I looked at the efficiency, Qts(for open baffle) and Xmax.
B&C 6MD38 | 96dB | Qts 0.44 | 2mm | 130Hz Fs
Eminence Alphalite 6A | 94dB | Qts 0.56 | 3.5mm | 126Hz Fs
I'm planning to cross the woofer to mid at about the BSC frequency of ~300Hz, and the mid-tweeter at about 4-5kHz depending on listening tests.
Although both drivers have similar Fs, the Eminence appears to go lower from the spec sheet chart (92dB at 100Hz vs 84dB), and has higher Xmax. Will both drivers be able to cross at ~300Hz?
Zaph has also reviewed the B&C quite favorably, but i couldn't find any reviews on the Eminence.
Any personal experience with these drivers, and recommendation for one over the other?
Many thanks!
Last edited:
A high dispersion ribbon crossing to a 6.5" midrange at 4-5khz will have serious off axis and mid-cone breakup issues. How about a high efficiency tweeter like the beyma AMT or the Fountek Neopro ribbons? That would allow for a lower crossover.
Thanks for the response. If it is not advisable to use this 2 mids up to 4-5kHz due to off axis and breakup issue, i can lower the XO frequency. You are saying the NeoCd2.0M is not suitable for a lower frequency XO, what would be the lower limit?
The NeoCd2.0M being high dispersion affects the XO frequency or the midrange choice?
The NeoCd2.0M being high dispersion affects the XO frequency or the midrange choice?
If it's like the Neo CD 2, then the recommended 2.5 kHz is about as low as you should go for the high-pass filter. (..and a "steep" filter at that.)
You might try "pinging" Dan Neubecker about the B&C driver. He has used it as an open baffle mid in his Basslines loudspeaker design. (..see what he thinks about a higher order low-pass for it around 2.5 kHz.)
I don't know if 2khz is too low for that tweeter, it might be ok with a very steep slope. Ribbons don't usually like to be crossed low but I think some of the NeoPro range can do it.
You'll find differing opinions on this however it's a good rule of thumb to try and match the dispersion of two drivers at their crossover point. The larger the diamter the driver, the quicker it's off axis response starts to fall off.
2khz should work ok for a 6" midrange but even a bit lower is better. You need to look at the off-axis plots of the drvier to see when it starts beaming.
B&C 6MD38 is the better mid and should be ok for a 300hz cross on a wide enough baffle to your woofer array. You can model the dipole peak and rolloff in Edge.
You'll find differing opinions on this however it's a good rule of thumb to try and match the dispersion of two drivers at their crossover point. The larger the diamter the driver, the quicker it's off axis response starts to fall off.
2khz should work ok for a 6" midrange but even a bit lower is better. You need to look at the off-axis plots of the drvier to see when it starts beaming.
B&C 6MD38 is the better mid and should be ok for a 300hz cross on a wide enough baffle to your woofer array. You can model the dipole peak and rolloff in Edge.
Thanks, i will look up the dispersion plots of the drivers.
For the XO slopes, i am planning to use a CX3400, so it will be active 24dB/octave slopes.
Based on the comments about dispersion, i have selected the Fountek NeoCd3.0MS from Madisound, as it has a shorter ribbon and so have better vertical dispersion.
I will look up Edge. Thanks!
For the XO slopes, i am planning to use a CX3400, so it will be active 24dB/octave slopes.
Based on the comments about dispersion, i have selected the Fountek NeoCd3.0MS from Madisound, as it has a shorter ribbon and so have better vertical dispersion.
I will look up Edge. Thanks!
Not to get ot but have you used that Behringer on anything yet? I got the CX2310 a year or so ago and it sounds horrible. Noisy as hell too, had to cut the gains way down to get the noise to an acceptable level. I got a miniDSP not long after that.......
Just wondering since I see you're planning on using some nice drive units.
I think you deal with the 5khz breakup with a notch filter and steeper XO low pass and take the 6md38 a little higher than 2khz, but you would likely have the midrange power response or directivity mismatch with the tweeter, unless you put the tweeter in a waveguide.
Here are some shots of the project Scott mentioned using a AE Lamda td12H, B&C 6md38 and a shallow waveguide mounted Peerless HDS dome tweeter. I think the sensitivity came out around 93db or so.
There is a build thread here (construction starts about page 8)
HTGuide Forum - The BaSSlines (was High Sensitivity Design)
Here are some shots of the project Scott mentioned using a AE Lamda td12H, B&C 6md38 and a shallow waveguide mounted Peerless HDS dome tweeter. I think the sensitivity came out around 93db or so.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
There is a build thread here (construction starts about page 8)
HTGuide Forum - The BaSSlines (was High Sensitivity Design)
Thanks for all the feedback. Is the 5kHz breakup worse than the datasheet graphs? I was thinking about the notch filter, how wide should i set the notch normally, for a 5k mid cone
breakup mode?
Midrange power response (mismatch) with the tweeter? What do you mean by power response mismatch?
Directivity - understand the concern about the 6 in mid beaming at about 3-4kHz and sounding weird with the wide dispersion ribbon tweeter. I was thinking maybe the off axis dropoff on the mid helps to tame the peaks at 4-5kHz, based on info from PE forum. Maybe i could use the speakers without toe-in, and at about 20-30deg off axis they would sound just nice?
Quote
Here is what I measured at 45 degrees off-axis.
2.0kHz, -1dB
2.5kHz, -3dB
3.1kHz, -5dB
4.0kHz, -6dB
5.0kHz, -12dB
I loved the open baffle use of the 6MD38 in the above speaker! But I am leaning towards the eminence for lighter weight, higher sensitivity, Qts, and better LF performance which gives more flexibility for the woofer - mid XO frequency.
breakup mode?
Midrange power response (mismatch) with the tweeter? What do you mean by power response mismatch?
Directivity - understand the concern about the 6 in mid beaming at about 3-4kHz and sounding weird with the wide dispersion ribbon tweeter. I was thinking maybe the off axis dropoff on the mid helps to tame the peaks at 4-5kHz, based on info from PE forum. Maybe i could use the speakers without toe-in, and at about 20-30deg off axis they would sound just nice?
Quote
Here is what I measured at 45 degrees off-axis.
2.0kHz, -1dB
2.5kHz, -3dB
3.1kHz, -5dB
4.0kHz, -6dB
5.0kHz, -12dB
I loved the open baffle use of the 6MD38 in the above speaker! But I am leaning towards the eminence for lighter weight, higher sensitivity, Qts, and better LF performance which gives more flexibility for the woofer - mid XO frequency.
Hmm, i guess i'm taking a gamble on the CX3400, it has some nice reviews. Worst case i can use it to experiment and find the optimal XO frequencies and build active XOs later?
I'm thinking that one limitation of these XO is there is no independent control for eg the mid LP and the tweet HP, they may not necessarily be at the same frequency, with driver acoustic rolloff and whatnot taken into account.
The CX3400 is what I have in my 3way system. There is no going back. I use the FBQ3102 for gentle tweeks to the mid LP and tweeter HP control...about 1.5 to 3 db max. Good specs for the CX and FBQ.
PeterC
PeterC
Good to know, my CX3400 shipped today! So did the Fountek!
I've been playing around with Edge, it seems almost impossible to get a remotely even response with an open baffle for a 6 inch driver. Are the responses really that jagged? How is a huge peak at 500Hz and peaks at 2-3kHz handled? Notch them all out?
Any rule of thumbs for open baffle mid, baffle dimensions? I have read the XO freq with the woofer and the BSC should come in about the same place, so ~300hz gives a 380mm width.
I was thinking about a trapezoidal baffle, similar to the pictures earlier. I also tried using the golden ratio to offset the driver from the center of the baffle, supposed to even out the dips? I plugged all this in Edge and got a horrible looking curve.
I've been playing around with Edge, it seems almost impossible to get a remotely even response with an open baffle for a 6 inch driver. Are the responses really that jagged? How is a huge peak at 500Hz and peaks at 2-3kHz handled? Notch them all out?
Any rule of thumbs for open baffle mid, baffle dimensions? I have read the XO freq with the woofer and the BSC should come in about the same place, so ~300hz gives a 380mm width.
I was thinking about a trapezoidal baffle, similar to the pictures earlier. I also tried using the golden ratio to offset the driver from the center of the baffle, supposed to even out the dips? I plugged all this in Edge and got a horrible looking curve.
Attachments
This is the response I got in "Edge"
An externally hosted image should be here but it was not working when we last tested it.
Make sure the "mic" is positioned at the center of the driver, otherwise you're getting off-axis sims. Be sure to try different mic distances as well.
I thought it was better to plot a response off axis (15 degrees) for a dipole design.
Would I be right in thinking this is done if you plan set the crossover after the first dipole peak otherwise you would have a deep null in the crossover region when listening slightly off axis?
Would I be right in thinking this is done if you plan set the crossover after the first dipole peak otherwise you would have a deep null in the crossover region when listening slightly off axis?
I'd say it's good to consider off-axis response for any design. I measure out to 60 degrees in 15 degree steps.
Crossover before the onset of beaming (near the dipole peak) and nulls aren't an issue 😉. With the chosen tweeter though I guess you just have to live with dispersion mis-match (or get a couple Neo3s)
@rhapsodee, yeah you're not going to get a flat response with a baffle that small. One of the trade-offs of going with a smaller baffle. I think the benefit of controlled directivity at higher frequencies is worth the needed eq to get it flat. Offsetting the driver will just smooth the response off-axis to one side, the other side will be different and probably much worse.
Keep it centered and eq is your friend!
Are we talking the woofer-mid XO or the mid-tweet XO? The dipole peak is right about 600hz right? Before meaning higher or lower than 600Hz?
Using a pair of Neo3s on each speaker? How does that help the dispersion?
I'm thinking about using a phase plug on the Eminence mid, should help dispersion right?
I'm also considering an aperiodic transmission line for the mid (eg the Statements), instead of open baffle. That seems to reduce the dipole peak.
Attachments
I have had very little luck with notch filters taming breakup, at least in tweeters. Have you considered using a smaller "full range" like a 4" that could reach the ribbon much easier?
You ideally want to operate before the dipole peak (this is about the point where the driver begins beaming, or narrowing directivity) for both xo's, if the woofer was dipole as well. With your modeled baffle yeah, 600 is about right, though I think in reality may be a bit higher. Your tweeter obviously can't go this low, so the mid would be beaming like crazy at the upper end of its range and the tweeter would have wide directivity at the bottom of its own range. Total directivity miss-match. You'd have to listen on-axis and even then the power response would have a serious dip from 1-2k, this would be audible.
I mention the Neo3s because they can be used as a dipole, so you'd only need one per side. With a narrow baffle (or no baffle!) for the mid and tweeter you'd get a much smoother transition as far as dispersion is concerned at the crossover, and IMO a much better sounding speaker.
I agree with tvrgeek that a smaller driver would be best to meet your tweeter, but that would make it a 4-way (still need the 6" or use an 8 to get down to 300 comfortably), and there aren't any small drivers that meet your sensitivity requirement that I know of. This is what I do personally, with an 8" from 300-1k, Neo8 1k-3.5k, and a Neo3 3.5k and up with no baffle.
There's always tradeoffs. Unless you use a large baffle for your mid, the eq needed to get a flat response is going to eat up the sensitivity of the mid anyways. Unless you have flea power amps sensitivity shouldn't really be that much of a concern in an active speaker IMO.
- Status
- Not open for further replies.