Exodus 2641 Measurements

Kevin P · « **on:** 6 Jul 2005, 10:56 pm »

This is the first round of measurements for the 2641. The 2641 is a tower speaker 41" tall x 8" wide x 12.5" deep using a pair of Adire Audio Extremis drivers, WR-125 midrange and the Usher 9950-15C tweeter.

The tweeter-midrange crossover is around 4K with a roughly 4th order electrical network. The midrange-woofer crossover is around 800Hz with a roughly 2nd order electrical network. For those who want to sharpshoot the choice of crossover points it largely comes down to where I have to choose to get the best on/off-axis response. I do this through a lot of software manipulation followed by building and testing. It takes time and sometimes what you think should work best doesn't so you have to keep an open mind. Also keep in mind this is raw data. No smoothing was used. After 1/3 octave smoothing we are talking +/- 1.5db from 150Hz to 20K on-axis and the off-axis behavior is superlative.

Here is the reverse null (reversed the midrange) showing the phase stays linear across the crossover points.

Here is the impedance curve. Its hard to tell from this but the low is around 4 Ohms around 2.3K and it averages over 8 Ohms for most of the bandwidth. Not much energy up around 2.3K so it should be an easy load for most amplifiers.

The bass portion of the cabinet is currently designed (above dimensions) with a SBB4 alignment tuned to 30Hz. You get very good response well into the 20s and the SBB4 gives low group delay/transient behavior.

I'm doing some final build/measuring over the next 2-3 weeks. I hope to have it on the market sometime in Aug. I expect the price to come in around $400 per speaker.

Jason1 · « **Reply #1 on:** 6 Jul 2005, 11:08 pm »

That looks really good. Are finished cabinets going to be available?

Kevin P · « **Reply #2 on:** 6 Jul 2005, 11:26 pm »

I'm working with a couple cabinet guys. The larger the speaker the more complicated the shipping issues. I had a design that broke in half (tweeter-midrange one cabinet and two woofers in another) so that we could ease shipping issues with the finished cabinet. We will release some designs that work best when built as a kit where the builder doesn't have to worry about shipping issues and we will have a slightly modified version for people interested in finished cabinets.

All of that has yet to be determined though as I'm not yet done with the design phase.

Al Garay · « **Reply #3 on:** 8 Jul 2005, 07:38 am »

First, you need to finish the Kit 641.

Kevin P · « **Reply #4 on:** 8 Jul 2005, 02:21 pm »

Quote from: Al Garay

First, you need to finish the Kit 641.

Design wise it's finished. I'm just waiting on a few parts to arrive.

Voodoo Rufus · « **Reply #5 on:** 14 Jul 2005, 06:52 am »

There a reason for the weirdness in the 100-200Hz region? Or is it not-so-weird for some technical reason I'd love to learn?

Can't wait to see pics of the finished kit.

Kevin P · « **Reply #6 on:** 14 Jul 2005, 02:35 pm »

Quote from: Voodoo Rufus

There a reason for the weirdness in the 100-200Hz region? Or is it not-so-weird for some technical reason I'd love to learn?

Can't wait to see pics of the finished kit.

Sure... measurement wackiness. We measure from a platform in a warehouse where the nearest boundary is about 9-10ft away from the speaker. It gives us pretty good data from about 150-200Hz up. It's not an instant transition though. You can see some wiggles in the response from about 500Hz on down.

For a final response I'd do either a close-mic or ground plane measurement and splice the bottom end with the top from about 500Hz and higher. For crossover design we don't need to.... the data we have with impedance and good FR data from 200Hz & up is all we need for the crossover work.

Keep in mind this is unsmoothed data. If you look at the raw data from any speaker measurement (even outdoors) you get peaks and troughs. The ear has an effective 1/3 octave smoothing so it's not cheating to smooth data. You get a better idea of what the human ear will perceive with 1/3 octave smoothing.

Voodoo Rufus · « **Reply #7 on:** 14 Jul 2005, 05:16 pm »

Is this also part of the explanation?

Quote

.....Pink noise is often used with 1/3 octave band filters to
measure room acoustics. This idea has merit since 1/3 octave
is a convenient number near the limit of our ears ability to
detect frequency response irregularities, and because
averaging measurements over 1/3 octave bands smooths out the
numerous very narrow peaks and dips that arise due to
standing waves in rooms.

Kevin P · « **Reply #8 on:** 14 Jul 2005, 09:18 pm »

Yes.... they are not explaining WHY the ear is an effective 1/3 octave smooting device but the are alluding to what I'm talking about.

Its educational to use LSPCAD to model speaker location in the room. It isn't necessarily something that will correlate exactly to your REAL room but it at least will show you the effect of moving both the speaker & listening position in a "virtual" room where you can set the dimensions and the absorption coefficients of all the room boundries. It helps you visualize how profound the room is in system design.

wildfire99 · « **Reply #9 on:** 15 Jul 2005, 08:16 am »

Is it possible to build this as a very large bookshelf model? (I.e., it would be a vertically-oriented LCR.) I've been waiting for the ability to make a vertically-oriented LCR with non-compromising horizontal off-axis response.

Also, does it make any quality difference to build these or the LCR kits as sealed boxes (for the extremis') if you're using a good subwoofer?

Kevin P · « **Reply #10 on:** 15 Jul 2005, 04:28 pm »

Quote from: wildfire99

Is it possible to build this as a very large bookshelf model? (I.e., it would be a vertically-oriented LCR.) I've been waiting for the ability to make a vertically-oriented LCR with non-compromising horizontal off-axis response.

Also, does it make any quality difference to build these or the LCR kits as sealed boxes (for the extremis') if you're using a good subwoofer?

You could build a 41" tall bookshelf.

I'd not call the off-axis performance of the LCR compromised. The 2641 will have a more symmetric horizontal dispersion pattern but there is absolutely nothing wrong with the LCRs performance if you wanted something shaped like a bookshelf.

Nothing wrong with using any of them sealed either. Quality is non-specific. The sealed boxes will give up an octave of extension, use a smaller box, and have better power handling in some ways and worse in others. Sealed designs tend to have slightly higher distortion because the driver has to use more excursion but they have better transient response. The Extremis has plenty of clean excursion and you get plenty of headroom and clean output from 50Hz and up so sealed is a good design choice if you need a smaller box. They will get low enough that transitioning to a sub is easy work and there are fewer variables in enclosure construction to goof up on. Sealed box roll-off also tends to match room gain better than many ported alignments. Much depends on your room.

Give me a call if you want some suggested sealed box sizes.

klh · « **Reply #11 on:** 15 Jul 2005, 08:06 pm »

How much does the internal volume matter if the speaker was sealed? What are the advantages of having a smaller sealed enclosure vs the same enclosure size sealed? If one doesn't need or want the enclosure to be smaller, wouldn't it be alright to just plug the port? It seems to me having the flexibility of unplugging it would be advantageous. It's not too hard to imagine wanting or needing to move the speakers and subwoofer to separate locations. It would be nice to not have to build another enclosure if ports became desired.

Kevin P · « **Reply #12 on:** 15 Jul 2005, 11:41 pm »

Quote from: klh

How much does the internal volume matter if the speaker was sealed? What are the advantages of having a smaller sealed enclosure vs the same enclosure size sealed? If one doesn't need or want the enclosure to be smaller, wouldn't it be alright to just plug the port? It seems to me having the flexibility of unplugging it would be advantageous. It's not too hard to imagine wanting or needing to move the speakers and subwoofer to separate locations. It would be nice to not have to build another enclosure if ports became desired.

If you plug a port the enclosure is a REALLY low Q alignment sealed. This is a broad topic and preference among sealed alignments vary. I prefer a Q of about 0.75 on full bandwidth enclosures. I like a Q of about 1 on limited bandwidth loudspeakers. A lot depends on the room and where you plan to place speakers in relation to other room boundries.

Really if you can live with the larger size there is no reason to go sealed in that size enclosure. Your better off with the ported designs. The people who are going to want to use them sealed are going to be those looking for a smaller enclosure and they will be willing to give up some bandwidth to get it.