[kyrill]

hi

I just read  gitarretype OB post ( http://www.audiocircle.com/index.php?topic=54178.0 ) and fell in love with his tender and tall OB design and the "magic" he found

I would like that design to accommodate
 different drivers
I am thinking of 2x2 SW-12-16FR Servo Sub Woofers from GR research
to support a Focal 6W audiom bass/mid driver, which i already own,  taking over at 100 hz (QTS 0.49)


and a Peerless 810921, 1inch  textile dome HDS tweeter taking over at 2khz

I expect hopefully "miracles" from my DEQX preamp who makes crossovers on the fly with slopes beyond 100dB/octave ( http://www.deqx.com/PDC26P-Preamp.html )

Will this work?

[scorpion]

Hi kyrill,

If you own the Audioms already I suppose you know their sound and like it. If any Band-Pass filter is needed to tame the rising 1 kHz response I suppose your DEQX preamp can fix that. For a baffle I think I would recommend at least 16" wide around the midrange to maintain linear response down to 250-300 Hz where Subs come in. I would complement the subs with a short 7.5" U-baffle behind á la MJK: http://www.quarter-wave.com/OBs/U_and_H_Frames.pdf. That would ease servo need and also linearise bass output. The U-baffle doesn't need to be greater than to make room for the sub-units.  This would be a very interesting OB speaker.

/Erling

[kyrill]

XX

[kyrill]

 
thx Erling

I haven't heard the audiom yet Kept it packaged until finally i would start the project of
making own speakers
but isn't Guitarr's design already an U frame?
 

[nodiak]

Hi kyrill, are you saying you would use a stereo pair of the GR ob subs? Two 16 ohm drivers in each sub (like Danny's w-frame example)? If so you can use any monitor you can dream of above them, right? The GR ob subs can reportedly reach 200 hz, so you could go easy on the Focal/Peerless ob monitor. Sounds like a very nice project!
Don

[scorpion]

kyrill,

Sort of, 'Topless-U' with damping. I have had good experience with this U-type in my 'Volks-OB'  http://www.audiocircle.com/index.php?topic=46951.0 and no damping. In MJK's version you don't need to damp either and depending on how the servo works I think you could set crossover formally to 125-150 and have flat response up to around 300 Hz for the next crossover.

/Erling

[Vapor Audio]

I've worked with the 6W's, they are a very nice sounding driver ... they'll actually be more sensitive than the Peerless HDS.  The one thing with the Audiom Mids that's kept it from gaining much DIY acclaim is it's break-up is very low (around 1200hz according to factory specs, my pair measures more like 1900-2000hz).  But that break-up is almost un-noticable when listening, assuming you bring down the rising response.  They have a great balance of detail and musicality. 

A single will probably go no lower than 150hz, the motor design isn't the greatest and starts to reach it's limits quickly, limit the travel as much as you can. 

I'd probably try to cross to the HDS around 1750hz, but that means you'll have to add some EQ to the HDS to bring up it's low-end a bit.  The HDS has excellent low-frequency distortion, but it's response droops, which makes a low x-over freq difficult without active eq'ing.  I'm undecided on my feelings toward how the HDS sounds though.  It performs very well on paper, but just doesn't have the micro-dynamics.  It's certainly capable, but I just don't find it does anything spectacular.

What sort of sound are you looking for overall?  I know drivers ... maybe I can help you choose. 

[gitarretyp]

Based on the specs of that driver (FS 117Hz, xmax 2.1mm), you're going to need to push it pretty hard to get a 100Hz crossover frequency. This may be acceptable if you only listen at lowish volumes, but i'd probably aim for 200Hz or higher to keep the driver in its linear range.

[D OB G]

Hi kyrill,

You don't need me to tell you about the verstility of the DEQX (I'm using the HDP3).

To people used to "normal" crossovers, the idea of high order slopes may seem flawed, and yet they work (they are not Butterworth, Bessel, Linkwitz e.t.c, but linear phase)!!

I don't know if you have kept up with the user group, but it is Kym's opinion, reinforced by others' experience, that mid to tweeter crossover is typically optimum at 60dB per octave.

This means that the Peerless tweeter probably has enough x-max to cross over at 1200Hz if necessary (I think Zaph shows low distortion at low freqs for that driver?).  I'm using a Dynaudio Esotar at 1200Hz very nicely. (Linkwitz is crossing his Seas tweeter at 1400 Hz at 24dB per octave).

It's possible that you can avoid the break-up freqs of the Focal driver (even though DEQX will correct for them), which seems to be an advantage, because their characteristics change off-axis (which DEQX can't simultaneously correct for).

David

 

[kyrill]

thx people

Erling
"could set crossover formally to 125-150 and have flat response up to around 300 Hz for the next crossover." Hmm, if i read you well, does this mean if i set Xover on the bassamp to 150, the drivers will continue f l a t to 300? How is this?

TurboFC3S
hi thx you for yr advise of drivers. I already have the Peerless HDs in house, still packaged.
Hate to read that maybe or is it not even maybe? that they may not let you hear deep into the micro details..I am a transparent lover, the less veil the better ( next to musicality and so on)

I dont like to EQ the tweeter low freqs level .This will lower the overall sensitivity of the tweeter; as i use a DECWARE triode TABOO tube amp which has only 6 watts I need all the senitivity i can get from the tweeter. 

if not.. i need a higher efficiency tweeter

the 6W Audiom will be tamed by the DEQX and i use Vinnie's RW 30.2 for it as amp. this is only 30 watts but the Audiom is over 92 db sensitivity which will make it loud enough, no?

I can play with the filters so i try out 100-150-200 hz to cross over to Danny's servo's

Guitarretype
thx for yr photoos. To me a picture is more clear than many words.  So yr "embedded" U design made me happy. Danny warned me  to make the "U" frame not too deep in order to avoid cavity resonances. I saw at yr photoo the "U" is not too deep but why do people not make the two free legs of the U not slightly unparalleled for that reason?
On the sensitivity graph of the audiom they are flat to under 100hz with  sensitivity of 92db. they are rated 100W-200W. Does yr warning means, that on higher levels (at 100hz) the driver becomes less sensitive, so needs more power than say, 300hz for the same loudness?

David
I upgraded my DEQX to nearly 3 status at DEQX,  the motherboard was too old to have the full mod)
you said: It's possible that you can avoid the break-up freqs of the Focal drive,
you mean by having this 1200 hz Xover to the tweeter? the overall sensitivity of the tweeter will get down and i explained my 6W amp cannot have that. The DEQX will correct the break up nicely within 0.5 dB. looking to the manufacturers graph i will Xover at 1800-2000Hz

I will in due time post the measurements graphs and photoos 

[MJK]

"could set crossover formally to 125-150 and have flat response up to around 300 Hz for the next crossover." Hmm, if i read you well, does this mean if i set Xover on the bassamp to 150, the drivers will continue f l a t to 300? How is this?

Because there will be a hump in the SPL response caused by the baffle itself. Below the hump the front and rear waves add destructively creating the low frequency roll-off. Then over some frequency range, determined by the size and shape of the baffle, the rear and front waves add constructively producing a hump typically 3 or 4 dB high. This occurs between 200 and 500 Hz for most modestly sized baffles. the EDGE program shows this well. If you place your low pass electrical crossover just below the hump, the acoustic crossover will be higher in frequency. So a low pass electrical crossover tuned to 200 Hz can have an effective acoustic crossover of 300 or maybe even 400 Hz. If you look in my OB articles or at the pdf's of my MathCad worksheets you will see pictures of this phenominon.

[scorpion]

kyrill,

Martin just answered the question for me , nothing to add. Also my advice about 7.5" depth has its roots in MJK's analysis. It will be really interesting to learn how the servo behaves. To me it seems it will react just the way simulations predict. That would be my first choice and then  follows the rest. With the short 7.5" U there will be a resonance somewhere around 300 Hz. By choosing a lower crossover frequency this is also adjusted and used to advantage for a higher real (acoustic) flat response crossover frequency.

/Erling

[kyrill]

thx martin (and others) read yr papers, transparent explanations

will order the servo's today but boy oh boy what is shipping to Europe expensive almost 500 $
hope to get my DEQX back next week from AU

am I too old for adventures?

[scorpion]

Who can be too old for adventures ? 

/Erling

[gitarretyp]

Guitarretype
thx for yr photoos. To me a picture is more clear than many words.  So yr "embedded" U design made me happy. Danny warned me  to make the "U" frame not too deep in order to avoid cavity resonances. I saw at yr photoo the "U" is not too deep but why do people not make the two free legs of the U not slightly unparalleled for that reason?
On the sensitivity graph of the audiom they are flat to under 100hz with  sensitivity of 92db. they are rated 100W-200W. Does yr warning means, that on higher levels (at 100hz) the driver becomes less sensitive, so needs more power than say, 300hz for the same loudness?

You could certainly make the sides of the u non-parallel. It just makes construction a bit more difficult. The way i did the u with it inset from the edges of the baffle was not the best idea, in my opinion. I ended up with a strong peak in the 160Hz region despite the u depth calculation indicating a resonant peak closer to 400Hz. If you want to try this idea, i strongly recommend building test baffles before you finalize the design.

The frequency response you posted for the audiom is either in an enclosure or on a very large baffle (the spec sheet doesn't say). As a dipole on a small'ish baffle (say 12-15") you're thinking about, it will be ~10 dB down at 100Hz from it's response at ~300Hz. You'd need to push the driver pretty hard (high excursion) at reasonable volume levels to reach a 100Hz crossover.

[kyrill]

ok I see guitaretype

[kyrill]

HAA  I ordered the servo's from gr research 
now i must experiment with different baffles and U-frames
i start with guitarretype's setup
and play a bit with the u-frame dimensions
I am convinced i will meet audiophile Nirvana at the end

If you really, i mean really enjoy music to the extend you close the lights
and listen in darkness
i will guarantee you, you will encounter that part of God
who's mandate it is to help you
recognize yr soul
in the abstract translations of music
as a further development of your soul as
 a creator

Don't ask me to explain 

[gitarretyp]

If you're going to use my style of design, you might try what i'm currently experimenting with. Make the width 15" or so to nearly the top of the upper woofer, then taper in to the final width at the top. You can then attach your wings to the lower, constant width portion. This design is working very well with the Acoustic Elegance IB15s i'm currently using.