Extremis 6.8 or XG18

[b_force]

Hi everyone,


I'm planning to build a dipole design with the XG18 or Extremis as a mid-woofer.

I've a few questions about the drivers because there are some things where a can't find the answer.

The crosspoint with the woofers (2 x SLS12 probably) will be around 200-250Hz, so the limited xmax is not an issue.

I've read a lots of posts, articles and test from both drivers, but never an 1 to 1 comparison between the two.
The distortion and overall performance from both drivers is very well.
(almost the same)

I've also read some stories that the extremis will have some roughness, coloration and distortion above the 1,8-2kHz.

The tweeter will be a NEO3 PDR, so the cross point will be in that range.

So the big question is, which driver should it be, the extremis or the XG18 ?

[augerpro]

I don't think either of those drivers is available anymore.

[b_force]

I don't think either of those drivers is available anymore.

I have the 2x XG18, and one (yes one !) Extremis 6.8.

I can buy another Extremis 6.8.

The question is which driver will be perform better.

[Dennis H]

Well, if you use the Extremis, Dan Wiggins has already designed a 2-way crossover with the Neo3. Don't know about the XG18.

[b_force]

Thank you !

I'm more looking for the performance between the two drivers. Which driver perform better and why......

The design will be custom, so other designs can be used as a start, but it will be completely different.

[jkrutke]

Used in a dipole, the Extremis will perform better. Mainly because it's much better behaved at the high excursion levels required by a dipole. It's very low sensitivity however, but it will handle a lot more power and ultimately get much louder without distortion. The XG18 is a great driver, but not one to push too hard.

[b_force]

I have simulated both drivers in Boxsim (with the known dipole trick) and the max spl en sensitivity are almost the same.

Except lower than 100Hz, but, that's not a big deal because a will cross around 200-250Hz.

What about the quality of the mid (and upper mid) of the drivers.
Are they very different in mid quality ?

[ThomasW]

http://www.htguide.com/forum/showpost.php4?p=146676&postcount=1

[b_force]

I've seen those test.
Great, well done !

But I thought the test on the first page is different form the second page ?

Maybe jonMarsh can tell ?

The funny thing is that I've read that a few people had experience that the Extremis will have some roughness/distortion/colouration around 2kHz, but I can't see in on the tests on Zaph's site ?
Only the CSD is a bit different from both drivers.

(The Extremis a little bit worse maybe)


[offtopic]
I noticed a strange difference between the test from jonMarsh and Zaph.
It's all about difference in the tests. In the Zaph test, the XG18 perform better then the 830833, but in the jonMarsh test it's the other way around ?
[/offtopic]

[ThomasW]

Different testers use different software, different test mics, different test baffles and different testing techniques. It's no wonder the results differ.

Maybe jonMarsh can tell ?

He's unavailable for comment.

Since you have examples of both these drivers, you're in the unique situation of being able to experiment and answer your own questions.

[Mark K]

Hi Bart,

I'd tend to agree with John. The extremis is probably a better choice. Excursion related distortion will still exist for the xg18 crossed at 200 more so than the extremis.

I say probably because Thomas is also right. It is very hard to make this determination without putting them both on a dipole baffle and testing. The off axis behavior of the two drivers will likely be a little different as will the upper end response.

My recommendation would be to test both baffle configurations. In the absence of that, I would probably pick the extremis. (In reality I wouldn't do this without testing...)

[b_force]

Ok, I will test both drivers then.
A lot off different opinions.

Unfortunately my mic is a mess. (upgrading to a DIY "super" mic :P)

But I think I can borrow one from Martijn_H :P

[---k---]

Isn't someone going to be mean and tell him that neither of those, that he must use Kevin's new Exodus EX-6.5 Midwoofer? :P

http://www.diycable.com/main/product_info.php?cPath=24_93&products_id=538&Cid=7f170cab6926878656a17ec8462348d6

[b_force]

Does someone has experience whit that (great looking) woofer ?

[ThomasW]

Probably not it was just recently released into the DIY market

[---k---]

You're supposed to be our test pilot. :P

[cjd]

I was actually pondering doing a 2-way (maybe 2.5-way) with the OW1's I have and that little gem. But I have other projects to finish before I can even think such things.

C

[b_force]

If I look to the CSD and compare the opinions, I think I'm going for the XG18.

Let's look to the CSD on Zaphs site, you can see that de Extremis is "slower" than the XG18.
A metal woofer (like the AL130 for example) would be even "faster".
(faster and slower are NOT the exact words I mean, but I don't know a better word )

The CSD of the Extremis explains why some people hear some resonances and problems.

Maybe I will upgrade later for a different driver (maybe de Exodus driver).

A total different questions, is about the woofers.
Does anyone have experience with the Lambda/AE Dipole12 drivers ?
(or better some THD and CDS results)

[Saurav]

I recently got a pair of the cheaper alternative, the AE OB12 (based off the IB15, but modified to work better as a dipole woofer rather than a subwoofer). I hope to take some measurements this weekend, but my wife's going to be off at an art class and I have baby duty, so I'm not sure how much I'll get done.

[b_force]

I recently got a pair of the cheaper alternative, the AE OB12 (based off the IB15, but modified to work better as a dipole woofer rather than a subwoofer). I hope to take some measurements this weekend, but my wife's going to be off at an art class and I have baby duty, so I'm not sure how much I'll get done.

I'm very interesting in your results !!!!

[Dennis H]

CSD is just another way to look at the frequency response. One can be (and is) calculated from the other. Zaph says he keeps posting them because people like to see them.

[b_force]

Saurav,any results jet ?
(THD/waterfall)

[Saurav]

Nope, I didn't measure CSD or THD, I don't trust my measurement setup enough to try those. I have FR measurements, and assuming you're the same b_force on the AE forum, I posted links in your thread there. They're basically links to discussion threads here.

[b_force]

Yes I've read the links you posted, but a FRD measurement isn't good enough.

Which mic are you using ?

A CSD is easy to make with de impuls respons/FRD plot) and a THD is also not that hard to measure. A output of 2,84V will do.
(The disclaimer is that the THD cannot compared to other test, but it give an indication)

Can you try to measure it ??

[Saurav]

An uncalibrated ECM8000, and Speaker Workshop for software. I've tried generating CSD plots from the impulse response, but I always get weird results. I must be doing something wrong.

Can you try to measure it ??

Maybe at some point I might, but for now I'm enjoying listening to it Seriously, I still have a lot of work left on my crossovers, I'll have house guests in a month, so the priority right now is to fix the mess of wiring in the living room.

And one last comment:

a FRD measurement isn't good enough.

Agreed, but poor/wrong measurements aren't necessarily better than no measurements. I'm fairly casual about this hobby, and that includes how far I go to get proper measurements, which is why I don't plan on publishing any of my results beyond the graphs I've posted. I think other people are in a better position to take the kind of measurements that you're looking for.

And I use a single-ended tube amp. As you can tell, THD isn't high on my list of priorities