Seas Excel W22 in 2-way

[Vikash]

Hi,

I've a pair of W22EX001 that I want to use in a 2-way. I still haven't decided on a tweeter, in particular between the 27TBFCG/27TDFC and Millenium. Amps are aplenty here so I was thinking of trying active 8th order elliptical @1200-1400Hz ala Jon Marsh (but will need a lot of help in the dept - hence why I'm here )

I see there are a few people around here familiar with the W22 units and was wondering if they have self measured T/S params? I've given my units a good bashing this weekend (over 50 hours at max excursion) and then measured them but the T/S params are still out by more than I believed would be so by Seas' standards.

Driver		Fs	VAS	Qms	Qes	Qts	
Specsheet	25	98	2.14	0.41	0.34
Measured A	30	68	3.04	0.50	0.43
Measured B	30	66	3.06	0.48	0.42

[Jed]

Hi,

I've a pair of W22EX001 that I want to use in a 2-way. I still haven't decided on a tweeter, in particular between the 27TBFCG/27TDFC and Millenium. Amps are aplenty here so I was thinking of trying active 8th order elliptical @1200-1400Hz ala Jon Marsh (but will need a lot of help in the dept - hence why I'm here )

I see there are a few people around here familiar with the W22 units and was wondering if they have self measured T/S params? I've given my units a good bashing this weekend (over 50 hours at max excursion) and then measured them but the T/S params are still out by more than I believed would be so by Seas' standards.

Driver		Fs	VAS	Qms	Qes	Qts	
Specsheet	25	98	2.14	0.41	0.34
Measured A	30	68	3.04	0.50	0.43
Measured B	30	66	3.06	0.48	0.42

If it helps- my W22s are in a 50-52 Liter ported box tuned to around 30HZ. When we took the measurements for impedance/FR, the box tuning was perfect.

[Vikash]

Hi Jed, thanks for the reply.

Strangely enough I've just noticed that my measured params are virtually bang on to the W21EX-001 specsheet. Not sure what to make of this really

Some modelling does indeed show that 55 litres ported box tuned to ~30Hz has the best set of compromises. What tweeter did you use? Any pics?

[TacoD]

I would suggest Seas Millenium / Eton ER4, and if money is an object, then you can use the 27TBFCG/27TDFC. But in my experience there is a audiable difference between all these tweeters.

[JonMarsh]

Taco, you'll probably like that my pair of Milleniums and W22's may end up in the M8ta's by Xmas time. I've got to model the box characteristics, and measure, too, but I don't think it's going to be an issue. Small chance that I'll go with the RS225 instead and a waveguide, but at this point I don't want to do anymore serious work on the cabinets. Won't decide until I finish driver measurements.

Re your T/S measurements, go with what you have- don't expect to see a lot of change once they're run in as much as you have, unless you've come close to wearing them out.

When I go ported, though, I tune for a slight downward taper on the LF response, so that the in room response isn't too heavy- usually about - 6dB in room at Fs. That works out to a typical enclosure size for the 8's I like of 60-70 liters, with LF tuning in the 27-34 Hz range, depending on driver. YMMV.

~Jon

[capslock]

Why Millenium and not RS28 or TB(F)C or TD(F)C? Looking at CSD from the HobbyHifi review, you get something in return for the extra cost of the W22 over the RS225, but according to various tests, this does not seem to be the case for the Millenium.

[TacoD]

Why Millenium and not RS28 or TB(F)C or TD(F)C? Looking at CSD from the HobbyHifi review, you get something in return for the extra cost of the W22 over the RS225, but according to various tests, this does not seem to be the case for the Millenium.

The dispersion of the Millenium is better than the RS28, but I don't know if that is the reason why I prefer the Seas. I even prefer the ScanSpeak 9800 above the RS, it sounds more refined than the Dayton (for a metal dome). The RS28 is for some reason a bit raw when directly compared to a 9800. Resolution wise there are no big differences between those metal domes, both these domes win in that respect from the Millenium.

I simply find the match of the Mill with the Magnesium Excel cones just wonderfull. I also tried metal domes, but while those have a more detailed presentation, their timbre doesn't match as well as in the case of the Seas. I have to point out that all these drivers are fine and performance wise above average, but that doesn't mean (for me atleast) that every combination of those units will have the same performance. For instance the SS 7000 tweeters measure not that well but paired with a sliced paper cone makes a lot of music. The sound is very convincing, and you have to try very hard to get the same "ease of producing music" with metal cones. With both types of units I have got very good results, with a slight advantage for the metal cones for their uncolored bass. But I prefer polyprop (Audiotechnology) for the upper mid... In the case of a 2-way I would choose a metal cone.

[JonMarsh]

I have a pair of Millenium's around that I should do something with, and this is one of the possiblities. I have heard them in other speakers at some length, including the Orion's, and their sonic character is pleasing, even if they have a smidge more upper order distortion products. But the off axis behavior is fairly nice. And no one has the SS 66000 in the US yet.

You know, to some extent, what we're doing it trying to pick the best "beauty queen" from among the worlds top super models; it becomes rather subjective, and ANY of the candidates will please considerably compared with the run of the mill...

With regards to the W22 and the RS225, the issue becomes how low I might cross the RS225, and whether it's better low frequency non-linear distortion offsets the lower midrange distortion of the W22. Decisions, decisions.

After I get done measuring, maybe I'll have a clearer picture of what I want to do. Or maybe I'll just be more confused.

~Jon

[Jonasz]

About the 66000, I talked to a fellow Swede (who actually ordered the tweeters) and he told me that the first batch didn't keep the specs so they will be a little delayed. But he who waits for something good...

[Dennis H]

I was hoping Davey would chime in here. Before the Orion came to be he documented an active analog crossver for the W22/Mill in a sealed box. The only thing missing from the circuit is the optional notch filter he mentions for the W22. I couldn't find the link to the doc file on his webpage so I posted it to mine temporarily. Grab it quick if you want it.

Davey's Seas Design

[Jed]

I'm using the W22 in a 3-way design with W15CH and Hiquphon OW1.

[Jed]

For instance the SS 7000 tweeters measure not that well but paired with a sliced paper cone makes a lot of music. The sound is very convincing, and you have to try very hard to get the same "ease of producing music" with metal cones. With both types of units I have got very good results, with a slight advantage for the metal cones for their uncolored bass. But I prefer polyprop (Audiotechnology) for the upper mid... In the case of a 2-way I would choose a metal cone.

TacoD,

Are you going to try the new Seas Nextel midranges? My next step to see how the Nextel 15LY compares to the W15CH that I'm currently using.

Regards,

Jed

[JonMarsh]

I was hoping Davey would chime in here. Before the Orion came to be he documented an active analog crossver for the W22/Mill in a sealed box. The only thing missing from the circuit is the optional notch filter he mentions for the W22. I couldn't find the link to the doc file on his webpage so I posted it to mine temporarily. Grab it quick if you want it.

Davey's Seas Design

I grabbed it just for fun... Always enjoy looking at other's work. Thanks, Dennis.

~Jon

[mikec]

I was hoping Davey would chime in here. Before the Orion came to be he documented an active analog crossver for the W22/Mill in a sealed box. The only thing missing from the circuit is the optional notch filter he mentions for the W22. I couldn't find the link to the doc file on his webpage so I posted it to mine temporarily. Grab it quick if you want it.

Davey's Seas Design

He also did a passive version using the L22 & T25CF001 Excel, which he was kind enough to loan me for a few weeks. It was a nice sounding design.

[Davey]

It's still there in my webspace.

http://home.comcast.net/~dreite/Seas/Seas_active.htm

I have quite a few measurements I made on that system that aren't on the page. One of the most interesting is the SPL of the W22 by itself via the 4th-order electrical crossover. The 5khz peak is still clearly visible, but the peak becomes a "shelf" about -20db below the nominal level. A notch filter really should be used, but the system still sounded very good without it.

Cheers,

Davey.

[TacoD]

I have a pair of Millenium's around that I should do something with, and this is one of the possiblities. I have heard them in other speakers at some length, including the Orion's, and their sonic character is pleasing, even if they have a smidge more upper order distortion products. But the off axis behavior is fairly nice. And no one has the SS 66000 in the US yet.

I plan to do another 2-way using the SS 66000 paired with SS 22W alu cone, using steep filters. We see how it will work out... Also those Nextel drivers are on my todo list.

But first I want to finish my own 3-way. And a couple of 2-way loudspeakers for friends. (using Dayton, Seas, Peerless).

[capslock]

Taco, have you seen the recent Klang und Ton issue? The Nextel drivers come nowhere near the Excels distortion and CSD-wise.

[TacoD]

Taco, have you seen the recent Klang und Ton issue? The Nextel drivers come nowhere near the Excels distortion and CSD-wise.

I have that issue, and I agree those Nextels are not on par with the Magnesium cones. I think that the mechanical damping of the cone itself is a bit better, but on this board it's called "evil stored energy".

The proof lies in eating the pudding, so I want to try those units for myself. With every woofer I comment on I actually built something. I think there are to many people basing their judgement on specs/ measurements without listening. I agree that specs are important like measurements, but as Ged Lee shows (see his site) its hard to correlate. Therefore I have confidence in almost every good engineered unit. And not just putting some big magnet, a lot of copper and a alu cone in/at a basket. Sorry I have to say that the Dayton is great but not the dragonslayer I always read on every board. (if someone is interested I will post the x-over schematic of the TM I did with the RS180 and a Scanspeak 9800).

[Jed]

Taco, have you seen the recent Klang und Ton issue? The Nextel drivers come nowhere near the Excels distortion and CSD-wise.

Could you summarize the contents of the tests? I'm interested primarily in the W15LY versus W15CH, but I'm not sure which nextel woofer they tested in the Klang und Ton issue. Besides, if distortion was so important, why is Klang und Ton using "old" eton drivers in the Duetta 2.5way as a "reference" loudspeaker? Sometimes a well damped cone, even if it measures worse with some harmonic distortion tests, sounds more pleasing than some metal cones that release high order harmonic distortion where the ear is most sensitive. Just playing the devil's advocate for a moment here. My magnesium cone 3-way sounds very fine indeed, and I want to experiment with paper cone midranges to compare.

[Branwell]

Vickash,

I use the Millennium in a system thats crossed over at 1300hz with 96db slopes. While it sounds amazing 99% of the time, the tweeter can occasionally sound strained, i.e. starts to reach its limit on loud passages.

I’m not sure the 27TBFCG or 27TDFC with half the xmax will serve well where the W22 needs to be crossed over.

Branwell

[capslock]

Could you summarize the contents of the tests? I'm interested primarily in the W15LY versus W15CH, but I'm not sure which nextel woofer they tested in the Klang und Ton issue. Besides, if distortion was so important, why is Klang und Ton using "old" eton drivers in the Duetta 2.5way as a "reference" loudspeaker? Sometimes a well damped cone, even if it measures worse with some harmonic distortion tests, sounds more pleasing than some metal cones that release high order harmonic distortion where the ear is most sensitive. Just playing the devil's advocate for a moment here. My magnesium cone 3-way sounds very fine indeed, and I want to experiment with paper cone midranges to compare.

Because the guys who do speaker designs are not necessarily those who do the testing. I believe Udo Wolgemuth had a hand in many of the Eton designs. While he used to do speaker testing, too, he seems to prefer medium to high distortion drivers. Many of his recent designs use Tang Band drivers...


I'd have to look at the review to do an adequate summary. From what I remember, midband distortion was worse in the Nextel, CSC below 4 kHz was a lot worse, and last not least, there were nasty resonances in a range where the Mag cone is still clean.

[TacoD]

I'd have to look at the review to do an adequate summary. From what I remember, midband distortion was worse in the Nextel, CSC below 4 kHz was a lot worse, and last not least, there were nasty resonances in a range where the Mag cone is still clean.

The distortion is higher than a Magnesium cone and also the first breakup is at a lower frequency, but overall for this type of cone it's performance is very good. It is in the same ballpark as the Peerless Exclusive with the Peerless a little bit better.

I like those Eton cones, maybe not the best motor attached, but those cones have in my experience better bass performance than Magnesium or Aluminium.

[AJINFLA]

I use the Millennium in a system thats crossed over at 1300hz with 96db slopes. While it sounds amazing 99% of the time, the tweeter can occasionally sound strained, i.e. starts to reach its limit on loud passages.

Here is a (the for me ) solution for the endless questions about pushing your tweeters (like the H1212/1189) XO lower and lower without having to resort to ultra steep sloped (and usually more complex) networks: http://www.zaphaudio.com/WaveguideTMM.html
- layed out much more coherently than I ever could.
Of course you could use both a WG and high order network (as I do) if you really want to get lowww .

Cheers,

AJ

[Branwell]

Funny you should mention wave guides. I read that article a few days ago and ordered a few from MCM to experiment with. My main concern is getting the guide to mate well with the shape of the tweeters face plate. Burn that bridge when we get too it………. :-)

Branwell

[JonMarsh]

Vickash,

I use the Millennium in a system thats crossed over at 1300hz with 96db slopes. While it sounds amazing 99% of the time, the tweeter can occasionally sound strained, i.e. starts to reach its limit on loud passages.

I’m not sure the 27TBFCG or 27TDFC with half the xmax will serve well where the W22 needs to be crossed over.

Branwell

One wouldn't expect them to, but in side by side testing we did with my samples at MarkK's, the others hold up to very high output levels with tri-tone clusters at 800Hz, 1000Hz, and 1200Hz. 100 dB. Didn't expect that, but the spectral analysis doesn't lie. This isn't dissing the Millenium, it's just praising how good the less expensive (and not quite as flat) others are. My Milleniums, my low buck Seas Tweeters. Virgin Milleniums, too.

~Jon

[GrahamT]

Vickash,

I use the Millennium in a system thats crossed over at 1300hz with 96db slopes. While it sounds amazing 99% of the time, the tweeter can occasionally sound strained, i.e. starts to reach its limit on loud passages.

I’m not sure the 27TBFCG or 27TDFC with half the xmax will serve well where the W22 needs to be crossed over.

Branwell

Branwell,

Can you explain how you got such high attenuation rates? Thanks.

[Branwell]

Hello GrahamT,

A digital crossover called a DEQX. Does lots of wonderfull things including XO rates up to 300db.

http://www.deqx.com/

Branwell

[GrahamT]

Thanks, I was thinking it was something passive like the Joseph Audio Infinite slope.

[cjd]

Thanks, I was thinking it was something passive like the Joseph Audio Infinite slope.

Which is really just a crazy way of doing what Jon does with the Cauer Elliptic filters. :P

C