1/8 wave transmission lines?

[Bill poster]

I understand that quarter wave is normally the way to go, but space restrictions for my cabinet design means that I am considering an 1/8 wave TL (using the CSS SDX7) --

what are the differences between 1/8 and 1/4 in performance terms?
would it ultimately offer any real advantages over a BR cab?

Cheers

Will

 

[MJK]

I understand that quarter wave is normally the way to go, but space restrictions for my cabinet design means that I am considering an 1/8 wave TL (using the CSS SDX7) --

what are the differences between 1/8 and 1/4 in performance terms?
would it ultimately offer any real advantages over a BR cab?

Cheers

Will

There is no such think as a 1/8 wavelength TL. Sizing the length based on 1/8 of a wavelength will produce a 1/4 wave TL with a tuning frequency twice as high.

Martin

 

[Scottmoose]

Martin beat me to it (good to see you around BTW Martin).

There's no such thing, as he says. All other things being equal, you'll just double the pipe Fo.

 

[Bill poster]

Thats told me then! Thanks Martin for clearing that up- I saw a mention on another thread but nothing else on google. btw What kind of line do Bose use in their compact systems?

 

[Scottmoose]

A relatively long, heavily folded and very probably sub-optimal 1/4 wave pipe.

 

[speaker dave]

Since most TL designs are heavily damped, does the length matter as much? Seems like one of the old (and discredited) vented box notions of "tune the box to the woofer resonance".

David S.

 

[Bill poster]

Cheers scottmoose, sub optimal meaning the cross section of the line I guess.

 

[Volenti]

Since most TL designs are heavily damped, does the length matter as much? Seems like one of the old (and discredited) vented box notions of "tune the box to the woofer resonance".

David S.

The old and discredited notion is that heavy damping makes the line appear longer (even if it might make it sound subjectively better)

The length, area and (if used) taper still matter.

 

[Scottmoose]

Cheers scott, sub optimal meaning the cross section of the line I guess.

I rather suspect it's the whole lot rather than just the line CSA.

 

[mashypie]

Hello,

I asked a similar q about would doubling the length of the line result in a 1/2 wave resonator...

Scottmoose posted some useful links about 1/4 and 1/2 wave resonators on the 2nd page if you're intersted in the theory behind it - helped me understand what I was talking about a lot more!!

http://www.diyaudio.com/forums/full-range/178286-would-sound-better-2.html

But essentially a TL will always be a 1/4 wave resonator, no matter what its length

 

[mondogenerator]

I have to sympathise with the OP. having seen a speaker project online titled '8th wavelength TL' i too, for a short while, thought maybe it was possible.

you could say that was my introduction to 'snake oil'.

MJK, scott and others have dispelled alot of shakey 'science' in here, during the time ive been a member, but in some ways the old tl philosophy was poetic rather than scientific. sometimes poetically the combination of tuning to Fs, and the chosen CSA of the line worked; but many didnt work so well im sure. Despite this i still really like the idea of a aperiodically stuffed TL.

im sure though, with tinklering i could model it with MJKS sheets all the same, whether non tapered tl or otherwise.

A 10:1 tapered aperiodic stuffed TL might be very nice indeed, methinks.

 

[speaker dave]

I understand that the pipe will be resonant at 1/4 wave and multiples. The question is what frequency the 1/4 wavelength should be ideally set to.

I have read (and liked) the Augspurger paper, but it seemed that he showed the key to performance was to get the damping just right. Too little will reveal the upper resonances, too much will kill the benefits. It wasn't clear that setting the pipe length to a frequency of woofer resonance was essential to the design process.

David

 

[mashypie]

I understand that the pipe will be resonant at 1/4 wave and multiples. The question is what frequency the 1/4 wavelength should be ideally set to.

I have read (and liked) the Augspurger paper, but it seemed that he showed the key to performance was to get the damping just right. Too little will reveal the upper resonances, too much will kill the benefits. It wasn't clear that setting the pipe length to a frequency of woofer resonance was essential to the design process.

David

From what I remember, his best model was where the tube was tuned to 2xFs. That for me fits in with something that Rick Shultz also wrote about it linking in to the driver Qts, as Augspurgers was around 0.5. So I'm inclined to agree that it not that important to tune to the Fs

The more I read though the more I think that the length of the pipe can be practically anything, again as you're using the damping to filter out the higher harmonics. I think its whether you'd prefer efficiency with ripple or low efficiency with a smoother response

I am currently using 2 mini TLs, a straight one tuned to 89Hz before stuffing and another which has the length of a 77Hz tuning but is tapered so possibly tuned lower, again before stuffing.

Both sound incredibly similar. I think the higher tuned one (which was built with the Qts in mind) actually has slightly more higher bass than the lower tuned one. Im not sure which I prefer at the moment. But they do sound good playing together and I cant wait to start building around bigger drivers...

 

[MJK]

I understand that the pipe will be resonant at 1/4 wave and multiples. The question is what frequency the 1/4 wavelength should be ideally set to.

I have read (and liked) the Augspurger paper, but it seemed that he showed the key to performance was to get the damping just right. Too little will reveal the upper resonances, too much will kill the benefits. It wasn't clear that setting the pipe length to a frequency of woofer resonance was essential to the design process.

David

The more fundamental question is : Do you want the output of the TL to augment the bass produced by the driver?

If yes, then mating a pipe to a driver is a similar phenominon to mating a ported box to a driver. Alignments for ported boxes are available and tend to tune the box/port to have a resonant frequency close to the driver's fs. Typically drivers that work well in a BR have a Qts between 0.3 and 0.5. For a Butterworth aligned BR the driver fs is equal to the box fb. For other alignments the box fb might be above or below the driver fs.

A similar set of tuning methods can be applied to TL's (no I am not saying you can use the T/S alignment tables for BR enclosures to design a TL). A driver that works well in a BR will probably also work well in a TL. You would still want to tune the TL to be close to the driver's fs. Correct amount of stuffing allows the TL's fundamental quarter wave resonance to help the driver produce adequate bass output while damping the higher harmonics. The low frequency cutt-off resulting knee in the SPL plot will tend to be more rounded compared to a BR design so the transition to the 24 dB/octave roll-off is more gradual.

There are no exactly correct or wrong answers, only a continuum of tuning options based on your ultimate goal. Tuning and SPL output are driven by length, taper, cross-sectional area, and stuffing placement and density. Trade them all off to get a design that is acceptable.

 

[planet10]

A 10:1 tapered aperiodic stuffed TL might be very nice indeed, methinks.

Worked really well as a midTL in Tysen, and similar is giving good results in the MTM i'm working on.

dave

 

[Bill poster]

I feel forever a noob on this forum ..couple more questions

If the TL is not exactly 1/4 wave(or half etc) length then wouldnt that mean phase issues?

Another probably stupid Q is - if you close the exit of a 1/4 wave line, does that automatically then become a 1/2 wave closed TL?

 

[mashypie]

I feel forever a noob on this forum ..couple more questions

If the TL is not exactly 1/4 wave(or half etc) length then wouldnt that mean phase issues?

Another probably stupid Q is - if you close the exit of a 1/4 wave line, does that automatically become a 1/2 wave closed TL?

My understanding is that there will always be phase issues, that is why you get a ripply output, there are freqs where the driver and port are in phase and other freqs when they are out of phase

If you look at figures 3 and 4 on the below link, even though the port output is high on figure 4 at about 120Hz, it doesnt result in a full peak on the combined output of fig 3 at 120Hz, more it peaks then troughs, theres a bit where they combine and another bit where they cancel

Pearls from Martin J King Quarter Wave Design

 

[speaker dave]

The more I read though the more I think that the length of the pipe can be practically anything, again as you're using the damping to filter out the higher harmonics. I think its whether you'd prefer efficiency with ripple or low efficiency with a smoother response

That was my take on it. Clearly we want more LF output, hopefully with less excursion/distortion. Beyond that the key is finding the best tradeoff between bass augmentation and ripple.

If that is the case, then the 1/4 wave aspect of it becomes secondary.

David S.

 

[MJK]

That was my take on it. Clearly we want more LF output, hopefully with less excursion/distortion. Beyond that the key is finding the best tradeoff between bass augmentation and ripple.

If that is the case, then the 1/4 wave aspect of it becomes secondary.

David S.

One end closed and one end open, it will always be a 1/4 wave resonance. The length and taper will detemine the frequency of the 1/4 wave tuning. At the 1/4 wave resonances the driver's cone motion is attenuated and the open end output is at +/- 90 degrees with respect to the driver.

 

[Bill poster]

i read somewhere that if you have 2 pipes in the same enclosure, the 1st one 'full length', the 2nd one roughly 15-20% of that length, it will help cancel out ripples