Scan-Speak 21W/8555-10

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Your bald-faced lie was saying that the impedance dips to 3 ohms. It absolutely does not in the 20 Hz to 20 kHz range, and it only drops very slightly below 4 ohms very briefly between 3 and 4 kHz, hardly anything to be at all concerned about. Your concerns about the potential "shorts to earth" are unrealistic as long as the driving amplifier isn't defective or a poor design. I liked the Sonatello's performance and sound, so did "Thunk" from his thread about his build, so did our piano tuner I sold them to, as well as many fellow DIYers who heard them. Please get off of your high horse.
Paul

What did I say that was a lie? Charts below. :D

684445d1527884493-sb-acoustics-3-sonatello-speaker-sonatello-paul-kittinger-jpg


I took the trouble to sim your speaker, and found it looked more like a test of concept. Forgiveable because it was 6 years ago, and really nobody knew much about SB drivers.

A 2.2kHz cross to the tweeter doesn't really play to a 3 way's strengths. The tweeter filter is underdamped, which is why the impedance dips so low at 3kHz. An attenuator would work better and perhaps sound better too. The tank notches are overly deep and there are two paths to Earth in the bass and the mid. I did say originally that you just take your beating on a bass reflex impedance.

I would call it an interesting failure. Sorry. Nowadays we think it is far better to use a dedicated 4" midrange (or even two, wired in series!) that is flat rather than a difficult 5" midbass. Or at least, me and Troels Gravesen do. Then it all falls into place, simply.

Back on-topic. Some great woofers discussed in this thread. The SEAS Nextel is admirable:
E0045-08S W22NY001

But quite pricey, I think. I'm sure it's not hard to find a bass that is comfortable up to 600Hz. Then the mid does its magic up to 3.5kHz. Seemples.
 
My goal was to make a classic 3 way. 8" woofer+3-4" midrange+tweeter. Floorstanding speakers in 35-40L.

Dome midrange was my object of lust for a long time :). MOREL EM 1308. That is the speaker I want to implement in my design.

But cone midrange is a option too. W4-1320 for example.

Seas H1471, Seas H1252, Seas H1288 are the woofers to which I incline right now.
 
Contrary to what some "others" might say, you certainly can build a very nice 3-way with an 8" woofer and dome midrange. I did just that back in 2007 with the canTiLena. It uses an SS 22W/8857T00 with a dome Morel MDM55 and Fountek Neo CD1.0 ribbon tweeter. The woofer is crossed to the midrange at 750 Hz, and the midrange crosses to the tweeter at 4.8 kHz. All of the cabinet's volume of ~55 liters is devoted to a single-fold ML-TL for the woofer featuring a predicted anechoic f3 of 30 Hz. A smaller cabinet would definitely be possible if you could live with a higher f3. If you are interested in my write-up on this design, send me a private message with your email address included.
Paul


My goal was to make a classic 3 way. 8" woofer+3-4" midrange+tweeter. Floorstanding speakers in 35-40L.

Dome midrange was my object of lust for a long time :). MOREL EM 1308. That is the speaker I want to implement in my design.

But cone midrange is a option too. W4-1320 for example.

Seas H1471, Seas H1252, Seas H1288 are the woofers to which I incline right now.
 

Attachments

  • CanTiLena, no grille.JPG
    CanTiLena, no grille.JPG
    45.8 KB · Views: 351
I'm all for people following their dream, an all that, but dome mids aren't easy.

They don't play nicely on simple filters. There's a phase and slope problem to match with a cone bass, IIRC. And crossover needs to be higher than you expect.
SP38/13

Be worth replicating that circuit in a Sim. Or the classic Celestion Ditton 66. See how it looks. Unfortunately I had a clear out of old sims, so would have to set one up again. But I haven't the time.

But TBH, the Celestion Ditton 22 was their best speaker by a country Mile. A conventional 8" 3 way with a cone mid. I was listening to some recently. Still good.
 

Attachments

  • Celestion_Ditton_66_XO.JPG
    Celestion_Ditton_66_XO.JPG
    69.3 KB · Views: 353
The D3806/8200 is one of the least capable "dome mids" i've seen. 1.5" and 0.4mm xmax is just a tweeter by modern standards.

Assuming the EM1308 is just the MDM55 with a metal faceplace, about 700Hz LR4 should be a comfortable place to be unless you want it to play ludicrously loud (>105dB/1m).
 
All dome mids are glorified big tweeters. They don't have a spider, except for the Volt designs.

I just resurrected my Celestion Ditton 66 sim. It was a nightmare.

523543d1452190177-pioneer-cs-77-speaker-mod-suggestions-aequal4_steen_duelund_filter-jpg


The Duelund model assumes a flattish mid. Which is not what a Dome Mid is. The curve is bell shaped. So everything that works well in Duelund goes to pieces. Impedance, phase, slopes. We spent a lot of time on a Volt mid and 12" bass design with simple filters a while back, and everything was horrible about it.

I got something to line up with 4th order filters on bass and tweeter, but I really didn't like the demands on the mid, which gets shallower 2nd order electrical slopes which don't protect it from bass so well. And, the very point of a three way is the mid gets an easy life, so that is fail.

Seems like the few successful dome-mid designs use a 6" bass and a very loud tweeter, and I did replicate that.

I could be surprised here, but I couldn't find a solution that matched the simplicity and elegance of a cone mid. In short, IMO using a dome mid is entering a world of hurt. But don't let me stop you. :D
 
Last edited:
Sure most dome mids are constructed like tweeters, good ones included. The D3806/8200 has about half the radiating area and half the linear excursion of the average 2" dome so it's not really a good demonstration of what is possible. With a 2" dome mid having 4-8 times the displacement capability of a 1" tweeter, theoretically you expect to cross at about 1/2 to 1/3rd the frequency of a 1" tweeter and reality matches that.

Dueland isn't the only way to design a 3-way. Trying to work a dome mid into a 2nd order acoustical slope rarely works out well, just the same way that trying to work a dome tweeter into a 2nd order slope rarely works unless you cross unnecessarily high or the tweeter is an exceptional design. Domes naturally work themselves into a significantly narrower bandwidth than a larger cone mid. What you tend to get in return is significantly lower distortion, higher sensitivity and wider dispersion.

For most 2" domes what works well is 700-900Hz LR4 and 2-3kHz with slope dictated by the tweeter's demands. The narrower bandwidth of the mid means that its average level comes down a few dB, reducing distortion of the mid and you still get a big improvement in power response versus a 2-way. Phase tracking is easy with 4th order slopes, just make them slightly asymmetric and everything will line up.
 
Last edited:
All I can tell you is it wasn't pretty used as a regular mid. I used a 2" dome. :D

Cutting the bandwidth and level might get you out of the hole.

And that, FWIW, is Duelund too. Below. But really is the maths behind aligning a three way. The insurmountable mathematical rules of the Universe. :cool:

Woofer, low tweeter and supertweeter is a slightly different ball game where symmetry is sacrificed. You see that one quite a lot, and it works well.
 

Attachments

  • Aequal2_Steen_Duelund_filter.JPG
    Aequal2_Steen_Duelund_filter.JPG
    36.6 KB · Views: 294
I've changed my mind a bit. I chose W4-1320 as a midrange. More easy to work with. Simple enough network in low end.Xover point will be about 500-600Hz.

Woofer... well. I've decided to try Al. woofer. Rigid, low Fs, low distortions in working area.





Sealed box 36-37L. First impressions are good.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.