It's completely possible to have good bass from dipoles. The trick is that you need about 10dB more sensitivity from the bass compared to the midrange.
I ran a Seas T18RE with two Eminence Beta 15A's in series (otherwise too much bass...) in an open baffle configuration for a while. Now the project is on a hold, for various reasons, but bass output wasn't one of those.
Actually dipole bass, when done correctly is magnificent. You get clean punch and a truckload of resolution, but they need a lot of space behind them. I had mine placed 1,4 metres from the front wall, and the output went down to 35Hz approximately straight, falling quite sharply after that.
One thing, though, is that dipole bass doesn't pressurise the room like monopoles. Makes it sound a little dry at first.
Three points I've learned about dipoles in the past four or five years:
- Bass driver sensitivity needs to be really high compared to midrange
- Find the lowest Fs drivers, this will practically be your low cutoff point
- Cone area, can't have too much of that. Long excursion doesn't hurt either.
Thanks a lot, Inductor and Korkkiruuvi,
thousands of questions and you have probably heard them all hundreds times here allready... 🙂
I´m now trying to decide what I´m going to do.
Isn´t it possible to "silence" the speaker, bring it down a bit? I remember reading it somewhere.
Can I feed woofer amps from rec-out output?
And what exactly means active crossover?
thousands of questions and you have probably heard them all hundreds times here allready... 🙂
I´m now trying to decide what I´m going to do.
Isn´t it possible to "silence" the speaker, bring it down a bit? I remember reading it somewhere.
Can I feed woofer amps from rec-out output?
And what exactly means active crossover?
You can attenuate a driver, but on a middriver it's a bit more difficult. Unlike tweeters, they usually don't respond well to L-pads - ie. resistive dividers. If the midrange has a rising response you can counter that with a single inductor in the best case. "Hit that rising response in the head with a large coil"...
The rec-output is a bad idea since it is not volume controlled.
What's meant with an active crossover is a crossover at the line level. That means you need a separate poweramp for each driver. They can vary from opamp based or just buffered passive filters in the analog domain to DSP-based solutions. Hypex has a plate amp with a built-in DSP and two poweramp stages.
One thing though... That Alpha of yours isn't really the best driver for OB uses, regardless of what MJ King theorises. Tried it, the Beta wipes the floor with Alphas. Longer throw, better capacity, and subjectively more punch - and low end extension.
Last edited:
Visaton BG20
What enclosure type and size do you have right now?! (For modeling the same you have.)
Use the conjugate 2. on the picture (top) with 1mH inductor and 15 Ohms resistor.
You can use intermediaries like circuit 1. in the bottom.
You can read here to understand.
Attachments
Thanks again.
BG20:s are now in open baffle with alpha15, both without filters. I´ll try your thing, as soon I get the list of all components I need ready and order out.
I know now my final goal: Jamo 909 copies, with alphas and maybe I get two Betas so I get best of both!?
Please understand that my funds are _limited_ so I have to go step by step...
I have designed the mold for curved baffles allready, checked the local store for 8mm MDF-boards, (I´m carpenter so this is familiar to me) and looked around for midrangers and tweeters.
What should I look? Sensitivity? Qts?
Midrange: Seas MCA15RCY
Tweeter: ScanSpeak R2604/832000
Maybe?
I quess I leave BG20:s out of that plan and use them something different, like you mentioned earlier.
BG20:s are now in open baffle with alpha15, both without filters. I´ll try your thing, as soon I get the list of all components I need ready and order out.
I know now my final goal: Jamo 909 copies, with alphas and maybe I get two Betas so I get best of both!?
Please understand that my funds are _limited_ so I have to go step by step...
I have designed the mold for curved baffles allready, checked the local store for 8mm MDF-boards, (I´m carpenter so this is familiar to me) and looked around for midrangers and tweeters.
What should I look? Sensitivity? Qts?
Midrange: Seas MCA15RCY
Tweeter: ScanSpeak R2604/832000
Maybe?
I quess I leave BG20:s out of that plan and use them something different, like you mentioned earlier.
+1
I heard a design with a horn loaded Akai orthodynamic tweeter (from 7KHz).
audiodiyers.hu - TKreso "Lodola"
without the tweeter it's unlistenable.
No opinions?
Other canditate for midrange would be Tangband W5-704S ...?
Tweeters impendance is 4 ohms, is it a problem?
I´d like to put it there because I allready have them.
Other canditate for midrange would be Tangband W5-704S ...?
Tweeters impendance is 4 ohms, is it a problem?
I´d like to put it there because I allready have them.
I'd strongly advise not to try pairing Betas and Alphas. In any case you'd need a different low pass filter for both in the X-O, and the Alpha would in any case "fart out" long before the Beta, since the latter also has a larger xmax.
The midrange could be next to anything you like, sensitivity up to 89-90dB/W/m. The treble driver should have slightly higher sensitivity. A simple solution would be to use the Seas T18-coaxial. The tweeter and midrange are quite easy to pair and require basically just a cap and a coil.
Be prepared though, the crossover will get expensive really quick. Large inductors are needed for the bass section, and they come with a price. The coils should have low Rdc, and since you might need up to 22mH values (as I did with my open baffles), be prepared to pay through the nose. Jantzen C-coil is a reasonably priced option here.
3-ways aren't easy, and making them open just complicates things more. After four or five years of banging my head on the wall with these I'm still not finished. Though it might also be because of my striving for perfection unachievable with the drivers I'm using. For a beginners project a 3-way OB is quite a hill to climb.
If you're not already registered, you might also get some support from the finnish forum hifiharrastajat.org. They also have some measuring equipment paying members can borrow and use, if you already don't have your own measuring equipment.
I had a feeling about that. I´ll stick with Alphas, at least they are cheap. 🙂
Well, then Scanspeak for tweeter and Tangband or MCA15 would be something to start with. Tangband´s curve seems a bit too nice...? T-18 is bit too expensive for me now. And little bit too big for 909-clone...
About the coils, I made some calculations with this: 2-Way Crossover Designer / Calculator , and if I placed the crossover-point in 200 Hz, coil was 5,5 mH... Did I miss something? 1st order Butterworth. 2nd order coils were 8 and 9 mH.
This is starting to feel overwhelming...
If I sell Alphas and buy and make some diy-kit... Then I have working set of speakers, propably way better that I ever manage to "design" myself. And in the end I´ll get them propably cheaper. But. Been there done that.
Doesn´t matter if this would take couple of years or decade, I´ve learned something already, thanks to you all, but somehow I know I´m not even started yet. 🙂
Well, then Scanspeak for tweeter and Tangband or MCA15 would be something to start with. Tangband´s curve seems a bit too nice...? T-18 is bit too expensive for me now. And little bit too big for 909-clone...
About the coils, I made some calculations with this: 2-Way Crossover Designer / Calculator , and if I placed the crossover-point in 200 Hz, coil was 5,5 mH... Did I miss something? 1st order Butterworth. 2nd order coils were 8 and 9 mH.
This is starting to feel overwhelming...
If I sell Alphas and buy and make some diy-kit... Then I have working set of speakers, propably way better that I ever manage to "design" myself. And in the end I´ll get them propably cheaper. But. Been there done that.
Doesn´t matter if this would take couple of years or decade, I´ve learned something already, thanks to you all, but somehow I know I´m not even started yet. 🙂
Actually, currently Uraltone in Finland lists the Beta 15A at 68,55€ and the Alpha at 79,65€. So it's not even cheaper. I never got satisfactory bass out of the Alphas.
An electrically first order filter at the bass is going to get you nowhere. The only thing it might do, is tilt the bass response straight at the given "crossover" frequency, from whence it would continue until the natural rolloff of the driver. So you would have a rising response from ~40 to ~180Hz, and from there it would be quite level until dropping off after a kilohertz or so.
The bass response from a bass driver in a dipole is always going to be rising. You take a big coil and hit that rise in the head. Then add the "real crossover." It will most certainly look nothing like a textbook crossover.
My baffles had a 3rd order low pass on the bass combined with a first order high pass to the midrange. Acoustically it was somewhere around second order on both drivers. You kinda have to work with the natural abilities of the drivers, and work your crossover accordingly. Data sheets and textbook calculations are of little use here.
I don't want to discourage anyone. On the contrary. But be prepared to tackle a problem that isn't really as simple as it might seem on the outside. But first, go get yourself a measuring mike if you already don't own one. Trust me, it's the first thing you need. Before any drivers.
Here wee go...
An externally hosted image should be here but it was not working when we last tested it.
- Status
- Not open for further replies.