need info on sealed enclosure theory

[griffon]

I am building a high quality speaker in a design that I am hoping to bring up to audiophile quality with as flat a response as possible but am not sure on a couple of things.
My theory is to use a fullrange driver (probably a Lowther dx55) in a sealed enclosure cut off below between 80 to 150hz with a sub in the bottom of the speaker facing to the floor.

One of my problems is that the lowther wishes to have a fairly massive enclosure for its design uses (utterly full range) but I was wondering if by cutting it off at the above mentioned frequencies it would allow me to put it in a small sealed enclosure with out stressing the cone or removing transient handling and timing.

the enclosure would be a truncated pyramid design so there shouldn't be any standing waves in the small enclosure space which for the Lowther cone is 4640cm3 with 16.5cm high by 16.75cm wide and deep (approx) and the cone is only slightly bigger than this.

what thoughts on this and possible sizes that may work for me?
if you haven't seen Lowther or want to see their stats to help me then: http://www.lowtherloudspeakers.co.uk/

[Ron E]

In a sealed box with that driver you are going to be looking at a F3 of about 200Hz, which is ideally the lowest frequency where you would want the sub(s) to come in. If you want to cross lower, use a different driver with a lower resonance.

Use a strange shaped box because you like the way it looks, not because you think it will perform better. There are standing waves in an enclosure of any shape, but in odd shaped enclosures they are just not as predictable or easy to calculate. That's what damping material is for.

Transient handling and timing are audiophile buzzwords, not technical terms. As far as stressing the cone, ask the mfg. Based on their websites technical info and marketing copy, they are likely to tell you exactly what they think you want to hear.

[griffon]

was the f3 figure you gave me calculated on the box size or is it just down to the cones specs? and also by increasing the chamber dimensions would that f3 become lowered? too say 150 or lower at a push?

You see I think that the lower I can get the fullrange driver (without compromising sound quality) the more accurate the sound stage will become. I also feel that differences in the timing between the two drivers will be far less noticeable with a lower crossover point (which is what I meant by timing and this is not just a buzz word when relating to the differences between drivers distances to the listeners ear, especially when your trying to make a system as accurate as I want, or am I some how way of the mark here)

My reasoning for believing the timing may sound more accurate between the two drivers is because I would be using purely resonant frequencies and so all direct frequencies (which the human ear is more sensitive to) would be emanating from the one point.

at this point I should ask if I can make the crossover point higher with the same results or am I chasing a theory that wont work even

Anyway that was a bit more info on my aims here. If there is anything I am wrong about or any help with the top questions then please don't hesitate.

[Syncroniq]

Hey,

If youre going for liniarity, you should read this before building.

Troels G on Lowther

[Ron E]

Originally posted by griffon
was the f3 figure you gave me calculated on the box size or is it just down to the cones specs? and also by increasing the chamber dimensions would that f3 become lowered? too say 150 or lower at a push?

Just a guess

Off the top of my head:
Fc/Fs = Qtc/Qts = sqrt(Vas/Vb+1)

F3/Fc=sqrt((A+sqrt(A^2+4))/2)
where A=1/Qtc^2-2

For your box, slightly more complicated equations than the above give Qtc ~0.5 and F3~200Hz. A smaller or larger box will change Fc, but will not change F3 much. Note that Fc for your box (140Hz) is approximately F6 - the point where a LR2 lowpass crossover to a sub would be easy to implement. If you want to highpass the Lowther to reduce excursion and distortion, use an LR2 highpass at 200Hz and then cross the sub with an LR4 lowpass at 200Hz. Since this crossover uses the natural rolloff of the driver, you would need to verify T/S parameters to make sure it comes out OK.

Typically people use a back horn or TQWP with Lowthers, and live without the subwoofer.

Originally posted by griffon
My reasoning for believing the timing may sound more accurate between the two drivers is because I would be using purely resonant frequencies and so all direct frequencies (which the human ear is more sensitive to) would be emanating from the one point.

"Using purely resonant frequencies" does not make much sense here, but it does quite accurately describe the sound of most fullrange drivers. Notice the bumps in the impedance curve by Troels G. Those are all secondary cone resonances.

I intimate from all of your talk about timing and point source behavior that you fear or mistrust crossovers. Even in a single driver, the zero delay plane moves somewhat with frequency. A better way to get an "accurate" speaker is to use a coaxial unit.

There is a fullrange driver forum here where you may find opinions from fellow fullrange enthusiasts.

[wakibaki]

Lowthers are highly respected FR speakers that many people find perfectly adequate on their own in a vented box or horn enclosure.

If you wish to keep the whole installation small, then you could use a small vented box, at the cost of some bass volume and extension. There's no shortage of opinions on the design of such over in the FR section.

If you wish to improve the system with a subwoofer you have the option of keeping the lower frequencies out of the FRs with filtration at source. Depending on your installation, many players and TVs now offer a 'subwoofer' mode, and while this may not be exactly to your liking suitable filtration is not difficult to arrange anyway.

w

[griffon]

thanks for the help so far (especially appreciated Ron E)

I think that at this point after having seen the Lowther free air frequency responses and now the review by Troels G (thanks syncroniq) I may choose to drop the lowther. also I do suspect that their cones are designed for to large an enclosure for my needs.

As it is I was also considering the Fostex fx120 and a couple of tangband full range drivers with links to all(I'm in this forum now so I ain't moving the thread to the fullrange section ). how do you think these different drivers would fair in the same enclosure mentioned at the top.

as far as i can see it is really difficult to figure out the appropriate size of enclosure for a full range driver with the lower frequencies cut off.
All the enclosure size theories and Qes things seem to relate to subs and warn against having to small an enclosure, but with the base cut off shouldn't the correct enclosure volume be way less. (well obviously it should but how do I figure out the right volume). Is it a case of their being a minimum crossover point for the fullrange driver where above that the cones movements are small enough to match the small enclosure. is there a free software that does box sizes taking this stuff into account or something?

Also am I wrong here or is the 3" driver by tb-speakers the one with the cleanest response meaning it could prove to be the closest to a perfect speaker? or is it to quiet really?

thanks again all for the help so far.

http://www.fostexinternational.com/d.../pdf/fx120.pdf http://www.tb-speaker.com/detail/1208_03/w5-1611sa.htm http://www.tb-speaker.com/detail/1208_03/w3-871sc.htm

[Ron E]

Originally posted by griffon
Also am I wrong here or is the 3" driver by tb-speakers the one with the cleanest response meaning it could prove to be the closest to a perfect speaker? or is it to quiet really?

The Tangband is a well regarded FR driver, especially for the price. I think once it is filtered to give a flat response (see John Krutke's zaphaudio site for design and measurements) it has a sensitivity of maybe 84dB/1W. Whether that is too quiet or not depends on how much power you have to drive them and your idea of loud.

You could always use two in parallel, but then the highs would be quite directional - maybe you could do a focused array and put your head in a vice

Whatever you do, keep us posted.

If you want a less expensive FR driver that actually has some bass, there is always the venerable BUF20 (I think) from Pioneer. Ragged highs, but listenable/tweakable.

[griffon]

As it is the speaker I am building here is being built for sale for a high price eventually. the reason for the high price is the design which will incorporate an isobaric sub (using scanspeak subs) with an electronic 4th order electronic crossover to a full range driver that covers the rest of the spectrum up to 20khz. yes, this is because I don't like crossovers in the vocal range
and also because one of the best sounding speakers is a single driver full range system using no crossovers.
What has been generally accepted is that these systems don't generally have a good low base output and often use subs. Well I am combining all the best aspects of subs, full range speakers and floor standing monitors (I am essentially creating this as a highest specification monitor speaker).
The important thing for the full range driver is a straight response line from roughly 250hz to 20khz. To do this job I have looked at all sorts of different fullrange drivers yet never found such a straight line response as the 3" silver paper W3-871SC by tang Band. Is there something I'm missing about this driver. It costs £15 and yet I want to use it in a £5000 system because it appears to be the best.
are there likely to be large harmonic distortions of varying orders. would it undervalue my system to the point that i should go with an imperfect cone of higher value?

[Ron E]

Originally posted by griffon
As it is the speaker I am building here is being built for sale for a high price eventually.

If you want to sell at a high price, stick with a Lowther or Fostex.