Site Links
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
Quick Links
Categories
In this Discussion
Please review the site Rules, Terms of Service, and Privacy Policy at your convenience. Rules, TOS, Privacy
Get familiar with the reaction system: Introducing the Reaction System
TL Mega Paper
Scott Hinson recently posted a new mega paper on transmission lines. It is 43 pages long! I just finished reading it and found it very revealing, especially the measurements comparing TL's to BR alignments using new and improved "swoopy" ports. Check it out at Scott's facebook link below:
Comments
Thanks for the link Bill. I just downloaded it and will give it a read this week.
Moral of the story: Generally TLs are a whole bunch of fiddle farting to end up with the same (or worse) performance as BR. Any possible gain would be minimal (maybe a couple db) and come with a whole list of caveats... Get a 3d printer to mess around with swoopy ports for BR instead.
What is a "swoopy port"? Google got me nothing.
Seems to be a shaped port, kinda sorta like two low flare horns connected at the throats, then a radius makes the transition of the mouth to the baffle. I think I've seen 2d/slot load style with shaped sides called "waveguide ports" before.
Really funny this is. Jeff always said he could match performance of a TL Paul K drew up with a simple vented box, and really only semantics separated the 2. He felt saying as such would be disrespectful and step on the toes of the builder's that prefer TLs, so he declined to state his opinion on this publicly. I guess it's closer to the truth than presumed prior.
InDIYana Event Website
Each to his own preferences. Scott published the first version of that paper a few weeks ago and got "burned" by several folks. One myth - that a TL must have a larger box than the equivalent BR. I like TLs much better than BRs.
Here is a link to the AES paper describing "swoopy" ports. They call it "Loudspeaker Port Design For Optimal Performance And Listening Experience."
https://www.aes.org/e-lib/browse.cfm?elib=20683&fbclid=IwAR3Sx7n9GLnUjYOayBdbWzXBo7yt5Z7KFSmVQLFLn9wTMkYLSDPmdCxKBVY
Preference is preference. Nothing can change that.
I often appreciate the extra output of BR, but something about it just doesn't tend to be the most pleasing. For that I end up gravitating towards sealed. Clay's eggs may be some of the only TLs I've experienced and I'd have to say, to me they strike a real nice balance of extension while remaining detailed and resolving in the bass region. I just haven't heard enough yet to be sold that it is a general TL thing rather than a fluke, or a driver-specific thing.
This has been my experience with my own TL adventure and every TL I've ever looked at.
Thanks Drew. I appreciate the positive feedback. I read through the mega TL paper this morning and while I did not understand it fully, I think I detected some bias against TL. like when he rejected the notion that the wave is inverted at the terminus because it is resonant and too complicated.
Keep in mind Scott does a lot of empirical testing prior to publishing his findings. Obviously there is the chance he is testing towards a predetermined conclusion but he has gained a lot of respect from industry professionals and experienced DIYers so there is that to consider.
That being said, when I accidentally did a MLTL on my Taiga build I found the results to be stellar. However, I modeled it as a basic BR and stuffed the bottom of the cabinet until the pipe resonance was sufficiently suppressed. I had no idea all a MLTL is is exactly that.
As far as traditional TL goes, I have heard good and bad implementations. Like any other aspect of audio design, it is no panacea.
...also why I advocate sealed when possible, and avoid tall cabinets with ports unless some basic care is taken to manage the pipe resonance. TL/MLTL modeling helps quite a bit with that and that outcome (IMHO) should be how we view these alignments.
Yes I agree. We all have a potential for bias. However since we are on a subject that interests me, and of which I strive for a fuller understanding, I do not understand how a TL can be considered 4th ord. An MLTL yes. A MLTL has no line, instead it has a port, two separate volumes of air acoustically coupled. So I can see how it is 4th ord. I agree with Ben that it is just a BR in a tall form with stuffing. Please correct me if I am wrong, because I often am, but isn't the order of the enclosure determined by the number of chambers. And more chambers can often times mean more group delay?
I would not say the number of chambers dictates the order. You have one volume in a sealed, TL, and vented box; UNLESS this is defining the vent as one as well. I've never been solid on what makes the order except the outcome. Sealed is 12dB rolloff, vented is vented is 24dB rolloff or 4th order slope. TL boxes of any sort still have a reflex and resonance, and typically are 4th order or just under that mark, but never as low as 12dB rolloff.
Single reflex 2 chamber Bandpass boxes are '4th order'. I see the lowpass to be 4th order, but is the highpass portion really as well? The sealed volume will rolloff at 12dB on its own.
6th order is usually 2 vented volumes, but I've not investigated and seen if that applies as I've never built one. The process I would see or suspect is 4th order on both ends, and would not figure it to roll off at 36dB/oct.
Thus is why it gets fuzzy for me when there are claims of 8th or 12th order boxes. I just don't buy the claim that they roll off that steep.
InDIYana Event Website
The order is determined by the slope of the rolloff, just like a filter's transfer function (12,18, 24... dB/octave).
Thanks guys. I will be posting often because I had knee replacement one week ago. So get ready smart guys for some questions and non provocative comments.
I know it hurts, but that is a replacement that has to keep moving or it will stiffen. You seem to be in fairly good shape so I have no doubt you can do it.
We welcome your queries, egg-man!
InDIYana Event Website
As someone who has suffered a pretty serious knee injury, I know your pain. Please do all the exercises.
TLs have never been magic for me, reaching the same rolloff in the same sized cabinet as BR, but the possible advantages always seemed to be a larger port terminus for less chuffing and good management of internal midrange reflections. In practice, I think these two items can be managed well in BR and I've never heard an identifiable difference between TL and BR in the 100s of speakers I've heard, so yes, to me, it seems like a lot of messing around to hopefully achieve what a properly implemented BR can achieve. I'll be interested to read Scott's full write-up though.
My last cabinet was a TL design, just because I wanted to try it out.
Does it sound good? Yes.
Was it bigger than an equivalent BR would have been? Yes.
Is it better than a BR? Doubtful, I'm not going to make that claim for many reasons, mainly a lack of direct apples to apples comparison. It would be absurd to claim that TLs are better simply because I like the sound of this one, I would be comparing the experience to different drivers in different cabinets with different responses. How do I know that I like the sound because of some inherent TL quality, or perhaps its just the FR and distortion profile of this particular speaker, and a BR providing the exact same output would be equally as enjoyable?
Time and time again I see people jump to conclusions based on some anecdotal evidence such as above, I built a TL and it sounds great, therefore TLs are great, right? Where's the validation to confirm that the reason I enjoy the sound is because of the TL? I would have to build 2 cabinets, load them with the same driver and confirm with confidence that they provide the same response, and make comparison and judgments from there. I didn't do that, so I'm not going to make any claims one way or the other.
Reality is that as much as we try, we often don't truly search for the truth, but for validation of what we already believe to be true, conclusions are more easily found with this notion, and your mind is at ease from knowing the answers to life's questions. I once heard a long time DIYer explain that he had heard two different speakers in two different settings, one was passive, the other active DSP. He preferred the passive, therefore passive designs are better. Hmm, something doesn't seem right with that conclusion, and that is just one example.
Scott has put forth a great effort to try and dispel some of the myths surrounding TLs, and while the paper reads with a healthy bit of confirmation bias, I applaud the effort. I think one thing to consider is that the design tools for TLs are often a bit more complex than the BR design tools many people use, so the results of TL models may be a bit more accurate / true to reality, where a simple BR model without complex considerations such as cabinet shape, driver and port location, complex driver impedance, etc. leaves a bit of grey area between simulation and reality. Perhaps people prefer TLs because the result was a bit better predicted? Who's to say.
Many years ago I had the "TL Guru" design a MLTL for me and I thought the bass sucked.
I repurposed those same woofers in a BR design of my own and loved them. That said, I have heard TL's that were really great and a few that just weren't right. As accurate as I think those Martin King algorithms are I think the room has way more impact.
I've heard maligned and good versions of all of TL and BR, or active and passive. I've heard active next to passive on the same pair in the same venue (that had matched transfer functions), as well as different venues. I am allowed to have an opinion and form thoughts on these matters just as anyone else here would or does.
From my perspective, they are bass alignments and xovers, venues matching or different. Sometimes things sound better to me one over another that very well may not be perceived that way to someone else. That is okay. I'm just being an informant of my own thoughts. Shooting the messenger for his message doesn't do any good. Same goes for Scott here, having the courage to challenge the common thought processes about TLs.
All of us can learn from the rest of us.
InDIYana Event Website
Good stuff. Thanks. But I wonder who among us agrees that a MLTL is really a TL?
I say it is not. And amber Bock should be called a mild brown. Sorry if I'm being provocative.
Yes, a MLTL IS a TL.
For instance, all of Paul Kittinger's TL designs (that I know of) are MLTL.
Despite ease of construction, I have switched to tapered offset TLs, and no longer need to worry about port length or port area size to set Fb.
For my FTRS design, I built a bass-reflex ported box for 2 x RS180 woofers. Sounded good.
Then I added an internal divider that made it into an offset MLTL. Sounded better.
Here are the Winisd and Hornrespo sims.
Here are the close miked port outputs for the BR and MLTL measured wavelet
Honestly, I can't tell if one is better than the other.
But what I DID measure a difference was the impedance.
Martin King says the lower peak shows that the woofer doesn't unload below Fs. I think that makes a big difference in bass SQ.
The lower hump of the impedance is reduced simply because the reduced efficiency of the resonant system through damping. You can get a similar result from a BR by sticking some damping material over the port to reduce it's output.
Actually, now that I look at the wavelet graphs in detail, I see some port resonances at about 500Hz on the MLTL that aren't there on the BR graph. My next experiments will try lining the TL plenum with 1/2 or 1 inch open cell foam to see if that damps any port resonances without affecting F3. Should be possible if I don't obstruct airflow.
Thanks, Very cool stuff. I have read through Martin's work several times. He is a very intelligent man, I do not consider myself one of those. However what I can do is notice trends and tradeoffs and which attributes of a system benefit my goals the most, and pursue those. The cookies on the bottom shelf taste pretty good to me. Perhaps the ones up higher taste better but I cant reach them quite yet.
I guess I'm saying that I desire a thorough understanding of the basics. I wonder if depicting the different systems as equivalent electrical circuits with LCR components might help me to understand, trouble is that's stretching myself a bit.
In an analogous circuit of a vented speaker the air mass adds a reactive component compared to a sealed box, increasing the roll off rate and group delay. I think a TL without a large coupling chamber, the air mass couples directly to the cone mass, therefore not adding another reactance. What if someone could build a TL with a taper ratio fairly high, approaching 10, containing 3 lines of equal length but radically different taper ratios? Also no driver offset and no fiber stuffing. The attenuation of overtones would be achieved by multiplying their occurrences and thereby lessening their severity.