Extemis - Fact of Fiction

[sbolin]

I was just wondering if anyone has heard the status of the Adire Extremis...last I heard was, basically, coming soon, but I've heard that for a while now. Will I be getting a nice Christmas present this year , or....

[Grahamt]

On the CSS website they say they are expecting to get them this week. If that is true then I would expect them to arrive sometime next week hopefully. I already have my tweeters.

[sbolin]

Originally posted by Grahamt
On the CSS website they say they are expecting to get them this week. If that is true then I would expect them to arrive sometime next week hopefully. I already have my tweeters.

I hope this is true, I have quite a few on order. What tweeters are you going to use with your Extremis? I have several on hand that I will try, and see which works best.

[Grahamt]

I ordered two Extremises myself for a 2way with LPG 26T tweeters. I am copying a design that Jon Marsh is doing with a cauer elliptic crossover. I like Jon's design philosophies and he is very helpful with answering questions. I am doing this project to learn more because I am not ready to design on my own yet. Kevin at diy cable and Dan Wiggins are doing designs also but I dont know when they are coming out.

What tweeters are you going to try?

[sbolin]

I have some Focal TC 90TD5B that I got on sale at Zalytron (no longer on sale), some Seas 27TDFCs, and some Focal JM Lab TPCBJM, which I also got on sale at Zalytron. I haven't really decided too much on the design, but really would like to do a TL, maybe something reminiscent of the Seas Thor. Not sure, the speakers open up a lot of possibilities. Of course, I want to see what Dan comes up with, I am sure it will make the best use of the Extremis speakers. I would love to see an XBL-2 tweeter, too...

[kfr01]

I also have two on order. I'm not sure what I'll be doing with them yet. I'm also looking at the Seas 27TDFCs.

Grahamt, do you have a link to this Jon Marsh design or any more information about it?

[JonMarsh]

The design I'm going for/with Grahm will be similar to the M8 bookshelf design, as regards the crossoer type, but will be in a PE 3/4 cu. ft enclosure, tuned to about 32 Hz with a 2" Precision flared port; the crossover frequency I expect to be about 1.8 kHz, as a good balance between not taking the woofer too high, or the tweeter too low. The LPG26T tweeter is modified the way I use it; those details will be posted to; it's just removing the plastic acoustic lens from the grille, and adding some long fiber wool to the pole piece- no biggie, but pays some dividends.

The SEAS 27TDFC is a good choice, too, IMO; very nice non-linear distortion, especially if you have to push it low.

I'm going a design for another friend with this tweeter and the Peerless 850439 6-1/2" midwoofer; it would work well with the Extremis, too, I think. Maybe a little softer on the top. For this design, I'm using a somewhat different crossover than my normal; more like the quasi-series crossover I did for my daughter's speakers last year, which used a HiVi D6.8 woofer (a Dynaudio clone). It's a bessel alignment filter; fewer components. For their goals, and more limited SPL, it should work out. The Extremis is a bit rougher above 2 kHz than the Peerless, so I think it will mate better with the higher order crossover. (cauer elliptic tuned to 8th order L-R acoustical transistion band function).


The trouble with the Extremis is chosing a moderately priced tweeter that plays well with it's strengths and isn't a limiting factor in the system performance! With a more conventional crossover, say, 4th order L-R, it becomes a balancing act. Most folks wouldn't want to put a Millenium Excel tweeter with this driver, yet it's performance below 2 kHz probably warrants it...

~Jon

[Grahamt]

sbolin,

Cool tweeters. I have heard good things about the Focal domes and I want to try something with those Seas 27t tweets some day. I am interested to see what you come up with. Regarding the XBL^2 tweeters I hear they wont be out for a while, maybe later next year. Maybe sooner now though with CAC opening up. I was going to do a design by Mr. Wiggins too, but I am a little nervous when he mentions the high crossover point on the website. Also he was looking at expensive tweeters like Hiquphon, Morel and SS and I am sure he is busy with preparing for CES.

I think Jon is the most objective when looking at the drivers pros and cons and I like his sceptical (yet optimistic) approach. I like how he is trying to get the most out of relatively inexpensive drivers.

Jon,

Modifying the LPGs sounds like fun to me.
Are you going to do the full 6 dB for the BSC? I would like to see the full 5-6 dB myself. Also, could you explain how you like to do this, do you change one of the inductors in the low pass?

[JonMarsh]

Yes, I normally do the "full" 5-6 dB of baffle step comp. If someone is using a system truly on a bookshelf, then I'll do less, tailoring the response to that measured position.

My experience has been that a system which is "undercompensated" is annoying and fatiguing over time, whereas one that is "normally" compensated and then moved close to the wall is a little "riper" in the lower mids and upper midbass, but with a lot of recordings, or with HT, that's not necessarily a bad thing. From a viewpoint of balance, it's usually better. In A/B'ing the 8" two way with HiVi M8a against the Usher two way kit that was popular last year, for example, the latter sound thin and edgy in comparison. Some of that is lack of suppression of upper harmonic IM from the midwoofer, too (too high a crossover point), but some is the milder baffle step used, assuming near wall placement.

BSC is accomplished mainly by juggling the first inductor value against the upper midrange impedance, and in fact you can tweak the level over a narrow range, depending on what you use for an RC zobel for the woofer. Sometimes people just use an inductor paralleled with a resistor, so that there isn't as much interaction with the driver impedance, but then that requires having an extra inductor; same net insertion loss, typically. I prefer the single inductor approach. (fewer parts).

~Jon

[Grahamt]

Thanks for the explanation Jon.