[Timoxx4]

Hey guys.

I have just taken the plunge into Exodus Audio heaven and bought a LCR kit and a pair of 2641's. ( not built yet ).

I have just found that the LCR is going to be WAY too big where i want to put it in front of my TV so am wondering if anyone has plans for a sealed LCR i can work from (print out)?

I have subwoofers so the loss of bass is not really a big thing.

Its the depth i can't deal with.  I can possible fit it in if it where 10"or 12" deep instead of 16.75".

Any help is much appreciated.
Tim.

[kfr01]

Tim:

Download something like WinISD Pro and see what the estimated response looks like.  As long as you keep the spacing and size of the front baffle the same, the only range that should be seriously affected by sealing up the rear is the bass below 200hz.  So, from a rank amature, I'd say seal up and shorten up at will.  

Also, congrats on a great selection of speakers.  I own the 2641s and absolutely love them.  

I'm sure Kevin will have something to say on the sealed design.

Karl

[Kyle R.]

Keep the front baffle design the same as well as the dimensions on the sealed chamber for the tweeter and mid and then you can shorten the depth and seal it up.  If you do this then all that is affected is the low end response and if you are crossing that over in your receiver/preamp then you wont miss anything.

[Kevin Haskins]

Both Kyle & Karl are right on.   You can run them sealed in a much smaller cabinet.    Using the high pass feature on a processor will remove all the really low frequency content anyway.  

For multichannel music the bandwidth is probably more important but the sealed LCRs still have impressive LF performance.

[Timoxx4]

Thanks for the replies guys  

So any recommendations on how small i can go?   I am guessing the larger the better but what will going to 12" deep do regarding the bass?

If you do this then all that is affected is the low end response and if you are crossing that over in your receiver/preamp then you wont miss anything.

Yes that's what i had in mind. Using the Xover in the receiver set to small for the centre but large for the main 2641's.     And with 4 12" subwoofers in a smallish room i should be ok with a sealed LCR centre yes?

I guess there are no plans as such for a sealed LCR ?  Would i build it all the same with regard to the internal bracing?  Would completely stuffing the woofer part with poly fill be a good idea ? Or rigid fibreglass ? Which one?

Time to start building some more cabinets     Got one 2641 box about done.

Thanks again,
Tim.

[kfr01]

You can actually go relatively small with sealed designs.  I implore you to download an easy speaker design program and see for yourself what your estimated response below 200hz will look like with various internal volumes.

Shoot for a qtc of .8 or so before stuffing and you'll be fine.  You'll find this actually results in a pretty small box size.  Keep as much bracing as possible, imo.  Use something like black hole or a cheaper derivative for the enclosure sides and also use poly fill.

[Watson]

You may want to actually measure Qts of your Extremis drivers before deciding how shallow to make your cabinets.

Adire quotes the Extremis as having Qts=0.332, but John Krutke measured his as having Qts=0.482.  That's a huge difference when planning a sealed cabinet -- several inches of depth difference.  If you measure the drivers you have on hand, you'll know for sure how large to make your cabinets.

[Timoxx4]

I have tried WinISD but i find it too difficult to use and enter driver data.

Unless anyone has a WinISD file for the Extremis drivers they can send me ???

Just thought i would add that i am no speaker designer and have never built a speaker in my life.  That's pretty much why i am asking you guys for the help   And i appreciate it very much  

As for measuring the speakers, I agree it would be the thing to do but i wouldn't know how and don't really have the smarts to do it anyway  

So i am hoping someone has already done a sealed LCR before that i can copy, or have figured out what size to build the thing without actually building one?

Oh by the way kfr01,  What is the black hole stuff you speak of?  I don't think we have anything like that here in Australia     I have plenty of 1" and 2" rigid fibreglass.... Though that might take up too much internal air volume and need a bigger box to make it back up again ..

Thanks,
Tim.

[Watson]

Okay, I modeled the Extremis for you.  In each case I targeted Qtc=0.707.

If you go by Adire's specs (which, to be honest, seem likely to be inaccurate), your cabinet needs to be 9.39 litres and the -3dB point will be 65 Hz.

If you go by John Krutke's measurements (likely more reliable), your cabinet needs to be 17.24 litres and the -3dB point will be 68 Hz.

Pretty big size difference, huh?  That's why I would suggest measuring the drivers you actually have on hand.  If you can't do that, build the 17.24 litre cabinet.

[Timoxx4]

Update.....


Ok so i had one last crack with WinISD and IT WORKED        

So with 2 extremis in a sealed box with a Qtc of .8 gave me a box of 13.4L or 0.473 ft^3   Dose that sound right?   -3db point is just a touch under 70Hz  and its flat all the way to about 2000Hz where it starts to roll off slowly      

19L or 0.670 ft^3 box gives a Qtc of 0.707 roughly and -3db at 67.33Hz

So if i have done this all correct should i go with the Q 0.707 box or the Q 0.8 box and stuff it ( with stuffing that is   )?


After all that WinISD work i found the ADIRE AUDIO® EXTREMIS 6.8™ MIDWOOFER DATA SHEET   and found some recommended enclosure sizes. ( should have looked there first ay)  

They said for a sealed box, 9.5 litres with 100% stuffing...  So for 2 drivers that would be 19 litres yes?  So i must have got WinISD right after all

[Watson]

By the way, each of the examples I gave above assumes one driver.  If you have both Extremis drivers sharing the same cabinet area, you need to double the volumes I gave above.

[kfr01]

If you go by John Krutke's measurements (likely more reliable), your cabinet needs to be 17.24 litres and the -3dB point will be 68 Hz.

Actually, I don't know that I'd make that assumption.  IIRC Adire's tests were done by Dr. Klippel on very expensive equipment that I can only assume John does not have access to....

I'd go with the smaller cabinet.  When used with a subwoofer and full range mains, a 3db difference in the -3db point isn't going to make a squat worth of difference.

[Watson]

If you go by John Krutke's measurements (likely more reliable), your cabinet needs to be 17.24 litres and the -3dB point will be 68 Hz.

Actually, I don't know that I'd make that assumption.  IIRC Adire's tests were done by Dr. Klippel on very expensive equipment that I can only assume John does not have access to....

I don't see anywhere on Adire's website where they claim the measurements were done by Klippel.

There are a couple of reasons why I don't trust Adire's specs:
1) they published the specs before they moved manufacturing to Chile; there is likely some manufacturing variation in recent production
2) manufacturers have a strong incentive to inflate their specs (e.g. Fs of 30 Hz for a 6.8 inch driver?  John's measurement is more believable)
3) I tend to trust independent measurements more.

20% manufacturing variation is common, so it's always good anyway to measure the drivers you have on hand, particularly with low Qts drivers and sealed cabinets.  In this case, the quoted Qts of 0.332 vs the measured 0.482 is a pretty darn big difference, so it makes sense to try and verify what's actually going on.  Measuring Qts is easy enough that there's no reason to take anyone's word blindly for how the drivers measure.

[kfr01]

I don't see anywhere on Adire's website where they claim the measurements were done by Klippel.

There are a couple of reasons why I don't trust Adire's specs:
1) they published the specs before they moved manufacturing to Chile; there is likely some manufacturing variation in recent production
2) manufacturers have a strong incentive to inflate their specs (e.g. Fs of 30 Hz for a 6.8 inch driver?  John's measurement is more believable)
3) I tend to trust independent measurements more.

20% manufactu ...

Yeah, it isn't on the web, but was discussed on various forums.

I agree somewhat with your reasons 1, 2, and 3 and agree with your guideline that actual drivers should normally be tested when designing speakers.  

However, Adire has been known to be pretty honest.  While independent measurements are nice, and John is well respected in the DIY community, to my knowledge he has hobbyist / diy level equipment whereas Adire has their own build house and all the equipment necessary to take measurements to please clients placing large custom driver orders.  

I personally think all of this is pretty moot in this case given the fact that our friend is using a kit crossover and the change he's making isn't really application critical here.  

Cheers,

Karl

[Timoxx4]

Thanks for the help guys.

I think i will just trust Adire and if it turns out bad i will blame them  

Anyway, So now i know how many litres to make the sealed part for the woofers how dose that relate to the size i make the LCR box?  Dose anyone know the woofer volume as it is standard for the LCR?

Or even better how much volume in litres dose the bracing and tweeter/mid box take up inside the total enclosure volume?

I just need to know how much to shorten up the depth to reach the target 19 litres?

Thanks again,
Tim.

[kfr01]

Thanks for the help guys.

I think i will just trust Adire and if it turns out bad i will blame them  

Anyway, So now i know how many litres to make the sealed part for the woofers how dose that relate to the size i make the LCR box?  Dose anyone know the woofer volume as it is standard for the LCR?

Or even better how much volume in litres dose the bracing and tweeter/mid box take up inside the total enclosure volume?

I just need to know how much to shorten up the depth to reach the target 19 litres?

Thanks again,
Tim.

Sounds like calculator time for you.  Sorry, this one should be simple math.  

[Kyle R.]

Just a bit of clarification without intruding too badly:

1) they published the specs before they moved manufacturing to Chile; there is likely some manufacturing variation in recent production

The Extremis has always been made in the same location and that is not in Chile so there will be no variances  based upon manufacturing facilities.

Dr. Klippel did measure the Extremis at CES a few years back (and he was very impressed with it FWIW), but I'm not sure if these are the specs that are listed on the site.

Also, I'm not sure about background information on the measurements done by John Krutke, but be careful of independant user measurements.  There can be huge variances in results from test to test because of different equipment, different test methods, and whether the driver is tested right out of the box or if its been broken in, etc.

[Kevin Haskins]

First off I've measured  between 20-30 Extremis randomly for T/S parameters and they all have been within production tolerances.  

Second... we have little to nothing to gain by reporting T/S parameters that are wrong.   In fact it is bad business and unethical in my opinion.  

I'll try to clear up some misconceptions.   There are six fundamental physical properties that determine all the other driver parameters:

Le =  Inductance of the VC

Re =  DC resistance of the VC

Mms =  Moving mass of the driver's moving parts (cone, VC, former, dust cap etc...)

BL =  Motor force factor

Sd = Area of the cone/surround.

Cms = Suspension Compliance


Fs is determined by only by Mms & Cms.   Of the two Cms varies while Mms is held relatively constant.

Of those properties by far the larges variable in production tolerances is Cms.   Suspension compliance tolerances are typcially -+15% while things like Le, Re, BL, Sd & Mms are all held to much higher tolerances.  

How you measure has a great effect on Cms.   A broken-in driver will typically measure 3-10% different than a non-broken in driver.    Drive level effects the measurement.   Do a measurement with a 100mV input signal and your measurement will vary up to 10% in relation to measuring at a 1-2V drive level.    It also varies with atmospheric pressure, your not only measuring the compliance of the suspension but the resistance of atmosphere upon the cone movement.   At sea level you get different measurements than you will at 5000-10000ft above sea level.  

One thing to keep in mind are that the T/S model is an overly simplified explanation of transducer behavior.   The parameters are not static... they change with power level (VC temp), the suspension of a driver is not linear (why we have a range of Cms measurements) and the atmospheric pressure changes.   How all of these effect what enclosure to be used with a given driver is more complex than your simple software model is going to tell you.   But you know who understands the system better than some guy on the internet?   The guy who designed the driver!   He has built more, designed more and has more patents to his name than some guy posting on an internet forum.   If you don't trust his work in determining a relatively simple set of parameters I'd really say it's unwise to buy a product completely designed by him.

[Watson]

Second... we have little to nothing to gain by reporting T/S parameters that are wrong.

No one has accused you of doing this.  You don't manufacture the drivers, you just resell them.  

One thing to keep in mind are that the T/S model is an overly simplified explanation of transducer behavior. The parameters are not static... they change with power level (VC temp), the suspension of a driver is not linear (why we have a range of Cms measurements) and the atmospheric pressure changes. How all of these effect what enclosure to be used with a given driver is more complex than your simple software model is going to tell you. But you know who understands the system better than some guy on the internet? The guy who designed the driver! He has built more, designed more and has more patents to his name than some guy posting on an internet forum. If you don't trust his work in determining a relatively simple set of parameters I'd really say it's unwise to buy a product completely designed by him.

T/S parameters are remarkably accurate at capturing the small signal performance of a driver.

Krutke is not "some guy on the Internet"... he's a professional engineer, just as Dan Wiggins is.  This is science, not a popularity contest.  No need for personal attacks on either person; it adds nothing to the discussion.

The reality is that Adire's numbers are suspicious.  Take a look at their own posted impedance plot:
http://www.adireaudio.com/Home/Images/ExtremisZ.gif
Notice how the impedance peak measured by Adire is around 37Hz, exactly the same as what John Krutke measured.  Why, then, does Adire claim a Fs of 30Hz?  Adire's claim, using their own measurements, doesn't make sense.  Perhaps reality takes a back seat to marketing.

[Timoxx4]

Sounds like calculator time for you. Sorry, this one should be simple math.

     Thats fine if you are good at math...

Me ?   Not so good