2 way dappolito with scan-speak drivers

[Audiophile100%]

Hi,
like my past post suggest,i'm tihnking to build a spekaer with scan-speak drivers.
the candidates for woofer are :

- 18w8545-00
- 18w8545-k
- 18w8546

the candidate for tw is :

- d9900 revelator

it must be in dappolito configuration.
the loudspeaker must be 2 way and 3 speakers (the 2 wf crossed at the same
frequency with the tw).
this is an idea for this project.....i'm thinking about it.
someboby of you does projected a think like this with the speakers i mentioned ?
some project in web already done ? like this : http://www.audioxpress.com/reviews/media/AX301FL.pdf
please advise me.

thank you
regards

[JonMarsh]

This idea is similar to a NorthCreek project done years ago-

Why ScanSpeak? Have you ever looked at the Thor project (TL cabinet), using Seas Excel Drivers? I think you'd spend about the same amount of money, and wind up with a better speaker.

Perhaps others know of something similar to what you're asking for in a published available design.

If it was my money, I'd either build the Modula MTM's in a Thor TL style enclsoure, or possibly go with the new Peerless Exclusive Nomex series woofers in a simlar plan, but with the RS28a tweeter. Put the money you save on drivers into better crossover parts (like AudioCAP Theta in the tweeter crossovers)

But don't let my opinions sway you if you have your reasons-

One little question:

Here's the D2905/99000 "Revelator" response curve:




Here's the D2904/98000 response curve:



The 98000 I do have some experience with, having designed and built a couple of different speakers using it. What it is about the 99000 you prefer?


~Jon

[TacoD]

In my current project, the Seas drivers are really good, though the AT woofers are more dynamic and have almost the same resolution. An big disadvantage of AT is the poor bas performance. I would say the Seas is a more allround performer, but you have to construct a more complex crossover.

Scanspeak have also good drivers but the 8545 and 8546 are not on par with the Excel Magnesium cones or AT. If you stick to Scanspeak I would prefer the sliced cone 18W. All SS units are tuned for the distinctive "Scanspeak" sound, and in some cases it sound really nice with minimal x-over components.

I use the 99000 tweeter, it has lots of details/ airiness for a softdome, but due to the hornloading it can sound @8 kHz somewhat harsh. Also it takes some advanced filterwork to get the sound right. I think the Seas Millenium or Morel Supreme 110 is a better choice.

I would not go for the Peerless units, I do not think they will sound as good as Scanspeak. Like the budget Seas tweeters they measure very well, but I still prefere the more expensive Millenium.

[Dennis H]

That 9900 plot may be somewhat relevant to the waveguide thread. My understanding is the 9900 is just a 9700 with a small, shallow waveguide. Note how the 9900 has boosted lower treble but more ragged upper treble than the 9700 below.

[JonMarsh]

Dennis,

Interesting observation, and I'd guess it's dead on the money. Which of course, leads me to wonder, who persuaded them to make one in the fashion of the 99000? Who wanted that sound, or response profile? It's pretty easy to voice the 97000 to "flat" in the crossover; I'm at a loss to understand the attraction of the 99000.


Taco,

That's how you tell the engineers from the artists, I guess, I'm a little too nuts and bolts oriented. The LF linearity of the Scanspeaker SD1 midwoofer motor is quite good, and overall harmonic distortion is low... it's just their cone and suspension designs that make me crazy.

Based on what I've seen with the 850467, I have high hopes for the new Peerless Exclusive series- have some on order, too- only hopes at this point, but we'll see. If the LF distortion numbers are good, and the cone remains pistonic and ETC plots are good, AND if they show the same remarkably low midrange THD as the 850467 (which puts them in the "A" class with many drivers costing 2x), then I'm definitely going to give them a go... the rest is mostly voicing, IMO. I love the mag Excel performance, but there's some Scotts in my blood, and I'd be happier getting close to the same performance for 1/2 the money, IF it's possible. Maybe not, but we'll see.

It's definitely easier to voice the Millenium Excel well- I agree it takes more work, I think, for the 27TDFC. But probably not $120 worth more work. And my hat's off to you if you can tame the 99000 to where it makes your ears happy.

~Jon

[dawaro]

Scanspeak Designs

Here are a couple that I have seen on the web.

http://www.speakercity.com/scan_mtm_18.shtml

http://www.solen.ca/ssur.htm

Not the exact drivers your were looking for but close.

[Dennis H]

Back on the original topic, if you are set on those drivers, Madisound has a kit using the 8545K and your choice of the 9500 or 9700.

http://www.madisound.com/solist.html

[ekkoville]

Interesting...I was just looking at the Solist kit last week. I am as raw is there is when it comes to DIY, but hard to believe an $1100 kit has componets that are suspect in the eyes of the gurus here. Really makes you appreciate the work and results one can achieve for half of that amount.

Erik

[Audiophile100%]

thank you very much people!!
i'm searching for scan driver because of i prefer to seas.
i like very much mtm-18 project http://www.audioxpress.com/reviews/media/AX301FL.pdf
but there is not the crossover schematic!!,somebody could help me??

thank you

[JonMarsh]

Hello "Rolph",

Yes, I remember that review from years ago. Given the review that they gave this speaker, I'm a little surprised that this would push you in this direction for a new speaker.

Try reading the review online at AudioXpress for the Thor system using Seas drivers, and compare the subjective evaluations, as well as the objective.

Please understand that we support you in the right to be welcome to your own preferences and opinions, but human nature being what it is, you might not find that many folks being interested in helping you with a design based on those drivers if they aren't interested in it themselves at all- it removes the intellectual curiousity factor, for example, if the design is based on drivers with a mediocre value proposition or known problems, when there are comparably priced or even much less expensive drivers with more linear performance.

And the crossover network is in the article (without values), so if you want to reverse engineer the design (hopefully in a much larger enclosure) using the techniques from the FRD consortium I described in another post elsewhere, it would just take a bit of time and patience. Look at the NorthCreek Rythm/Revelator design to get graphs of the acoustic targets you want to employ for the crossover network.

But if you haven't done any crossover design, don't have measurement equipment (and don't plan to buy it), then a "proven" kit like the Madisound Soloist, or the NorthCreek Rythm/Revelator, (have you checked that out? Lots of graphs and design hints) would be more likely to meet your needs well. At $1798, you get a very nice speaker kit, with good design comparable to some high end systems.

However, for the last time, I'm warning you that you may not be happy with the SS based MTM if you ever hear a good design like the Thor's based on the Seas drivers, or even the "lowly" little Modula 1 MTM, which can be built for much less than any of these systems, even with premium crossover components, and has very smooth response, and the inherently low distortion of the rather amazing (for the price) RS180 and 27TDFC.

One gentleman and his son locally are building this in a Thor style cabinet, for the extended bottom end the transmission line should support. It's a lot of speaker for the cost in materials.

Good luck with your project; I'd recommend you do plan on getting at least a decent inexpensive measuring setup (say, free software like Speaker Workshop, and a Behringer mic and preamp).


~Jon