I would like to design a transmission line speaker around the dynavox LY401F full range speaker.
Dynavox LY401F 4" Full-Range Driver
I have never built a TL speaker before, but from what little I knew, these drivers seemed like a pretty decent choice to play and experiment with.
I have done a fair bit of research and one great resource I found was this video. YouTube
The problem is that when I try to replicate his equations , I get two graph lines that never intersect. When I add his 1000 extra cubic inches, I get other ridiculous numbers as well.
Did I choose a driver that will never work in a TL speaker? Or do I need to do something different with the math to figure out the TL length.
Dynavox LY401F 4" Full-Range Driver
I have never built a TL speaker before, but from what little I knew, these drivers seemed like a pretty decent choice to play and experiment with.
I have done a fair bit of research and one great resource I found was this video. YouTube
The problem is that when I try to replicate his equations , I get two graph lines that never intersect. When I add his 1000 extra cubic inches, I get other ridiculous numbers as well.
Did I choose a driver that will never work in a TL speaker? Or do I need to do something different with the math to figure out the TL length.
here's a little end drive constant cross section t-line of about 18 liters airspace. Its tuned around driver Fs but is shorter than a 53Hz quarter wave. Interestingly, when the distance between driver center and vent center are accounted, there's no dip. (maybe someone with MJK sheet or Leonard Audio TL experience will comment)
With classical approach a TML driver should have high QTs, reasonable low Fs od a large linear excursion.
The driver you selected, is well suited for a horn.
Now comes the newer way of looking at it: Any loudspeaker in an enclosure can be seen as a special case of a horn. So, if you start simulating with a really good program, the best simulation will be a TL that has strong indications of beeing a horn. So you keep on changing parameters, in the end your plan willbe a back loaded horn.
There are many "in between" constructions that working in some way or the other, like horn reflex, tapped horn or open baffel. As with any loudspeaker system, if the chassis fits the construction and deliveres a usefull response, the result may be good.
If the chassis is not well suited for a housing, you get what most first "own" constructions are, a waste of time and money. Some are so impressed by their own work, that they don´t want to see what kind of ill camel they have build...
My first quesion,do you have a measuring microfone and are familiar with a program like ARTA (no better I know for construction, cheap)? No? Learn first. Don´t build and "maybe later" measure.
My prediction, if you stick to this driver and put it into a real TM housing, it might play, but will be at the end of it´s cone travel at low volume.
My advice, for a first "own" construction, build a closed or reflex speaker, measure it (!!!)and listen in comparison to well known products. If you start to understand, how to build a good enclosure and how the physics of loudspeakers work, start constucting a TL. Most probably you will not want one any more.
Best,if you are magicaly drawn to TL, find a well proven design. For a beginner it is hardly possible to get this right, even if the plans are good.
Just my 2 Cent from 40 years of loudspeaker building.
The driver you selected, is well suited for a horn.
Now comes the newer way of looking at it: Any loudspeaker in an enclosure can be seen as a special case of a horn. So, if you start simulating with a really good program, the best simulation will be a TL that has strong indications of beeing a horn. So you keep on changing parameters, in the end your plan willbe a back loaded horn.
There are many "in between" constructions that working in some way or the other, like horn reflex, tapped horn or open baffel. As with any loudspeaker system, if the chassis fits the construction and deliveres a usefull response, the result may be good.
If the chassis is not well suited for a housing, you get what most first "own" constructions are, a waste of time and money. Some are so impressed by their own work, that they don´t want to see what kind of ill camel they have build...
My first quesion,do you have a measuring microfone and are familiar with a program like ARTA (no better I know for construction, cheap)? No? Learn first. Don´t build and "maybe later" measure.
My prediction, if you stick to this driver and put it into a real TM housing, it might play, but will be at the end of it´s cone travel at low volume.
My advice, for a first "own" construction, build a closed or reflex speaker, measure it (!!!)and listen in comparison to well known products. If you start to understand, how to build a good enclosure and how the physics of loudspeakers work, start constucting a TL. Most probably you will not want one any more.
Best,if you are magicaly drawn to TL, find a well proven design. For a beginner it is hardly possible to get this right, even if the plans are good.
Just my 2 Cent from 40 years of loudspeaker building.
Eh sort of. I understand his point very well. I am the type of person who likes to DIY everything because I firmly believe that if you want something done right you have to do it yourself. I also understand it requires a huge amount of time, money and effort in order to achieve something that is even half decent.
I have seen a ton of people who think they can just slap dash a product together and will end up with something really good. Of course these people end up either really disappointed in what they have made, or they falsely believe its the greatest thing since sliced bread and they go off to sell it at their local hifi show : P
What turbowatch said is definitely great advice for some if not most people. But the difference is that I am literally doing this for pure research. I already have the wood, a wood shop, and free time. I got the drivers on sale for 20 bucks for the pair soooo meh? I am not really worried about the sink of time and resources. Plus I have already designed sealed box speakers before. I just haven't done a TL speaker.
Of course it now looks like I need to make some sort of horn speaker. That is both an exciting and terrifying notion because I got into audio by listening to western electric speakers. So I feel like I have VERY unrealistic expectations of myself if I go down the horn path lol.
Any driver can be TL loaded, it just depends on what your size, performance goals are.
WRT to the video, at the beginning he states he's following Jon Risch's 'classic' TL design, which if you look at the top of it there's a note in red that states it's an obsolete way to design TLs, so knowing better from experience decades before ever being exposed to him/it, 'listened' no further: Classic TL Design -- Jon Risch
MJK's extensive research is considered the 'Bible' [with good reason] and has had an excellent 'classic' freeware design routine available since 2006 with another forum member? sharing his Excel SS a few years later to save some of us from having to wade through the instructions and somewhat tedious tables:
http://www.quarter-wave.com/TLs/Alignment_Tables.pdf
http://www.quarter-wave.com/TLs/Alignment_Tables_Calculator_3_3_09.xls
There's also another, simpler way developed around the same time by George Augspurger where tuning [Fp] = Fs/Qts' where Qts' = Qts + any added series resistance [Rs]: http://www.mh-audio.nl/Calculators/newqts.html
I don't have quick access to his and/or Rick Schultz's Alpha TL design routine docs based on it, but a good rule-of-thumb for the TL's net volume [Vb] = ~4-10x Vas depending on driver Qts', Vas specs.
Driver location normally depends on the app, but always at a pipe harmonic [ditto for vents] i.e. for max acoustic efficiency it's located at the closed end [1/5 = ~L*0.21] and for overall smoothest response with the driver offset down the pipe [1/3 = ~0.349, 2/5 = ~0.424].
Anyway, TL design being what it is, whichever design routine you use for a given [Fp], [Vb] or even by 'sliding' your way to it using Hornresp's Loudspeaker Wizard [LW], the simmed response will be near enough the same, though stuffing density of MJK's inverse tapered variant [TQWT] is much less, so should sound more 'open' than the more heavily stuffed straight pipes regardless of whose design routine you use and IME is the best suited overall for low Qt drivers since they most often require acoustically small cabs with large, long vents for good efficiency, so might as well morph them into one long, tapered box.
GM
As previously noted, it would normally be some form of BLH, so again, MJK has done extensive research on it and front loaded: Horn Theory
If you want to start [super] simple, the 'BIB' parabolic pipe horn is a perennial favorite: https://www.diyaudio.com/forums/ful...x-craft-handbooks.html?highlight=terry+cain's
MJK's thoughts on the BIB: http://www.quarter-wave.com/General/Fostex_FE-167E_BIB_Design.pdf
Do a 'Search: Keyword(s): terry, cain's ; Forum: Full Range and sub forums' for dozens of relevant [build] threads.
GM
that was a very helpful overview GM. I've only had two TL - one - Sander's 10 ft sub for Eclipse W1038R - (weak/flabby upper bass - not enough displacement at the low end ) and ESS Transtatic w. B139 in a short (maybe 5-6ft?) line, B110 midrange in stuffed open end pipe TL and electrostatic array at the top. I think there was a problem with the electrostat's power supply and it didn't have a lot of subjective impact nor power. Using a Karlson K15 "the wrong way" (EXEMPLAR) tuned low / boosted like 6th order reflex and loaded with Altec was much more impressive - I don't understand why it might work/sound better than the same woofer in a simple 6th order reflex.
Last edited:
Thanks!
Not familiar with these TL speakers, but if you mean the early '90s Madisound Eclipse; this and its 12" variant was the last 'el cheapo' speakers I did for some co-workers before retiring and worked a treat in the mid-bass, lower mids using Fs/Qts and still enough down lower to do DSOTM when near 1pi.
In a 10 ft straight pipe, my SWAG is it would need to be 25-30 ft^3!!
Big difference between a basic vent, even ducted, and 'K' slot.
GM
Not familiar with these TL speakers, but if you mean the early '90s Madisound Eclipse; this and its 12" variant was the last 'el cheapo' speakers I did for some co-workers before retiring and worked a treat in the mid-bass, lower mids using Fs/Qts and still enough down lower to do DSOTM when near 1pi.
In a 10 ft straight pipe, my SWAG is it would need to be 25-30 ft^3!!
Big difference between a basic vent, even ducted, and 'K' slot.
GM
yep - Eclipse W1038R was the butyl surround 10 (think made by "Carbonneau" - does that sound right?), sold both by Madisound and Meniscus . I had them in the line below. At one point to see what would happen, I cut a pretty long Karlson/Fulmer slot. That effectively shortened the line a bit and introduced two dips. I'd love to give those and some good big cabinets to someone to make some space. They need new speakers. It would be easy to put in a new plate but the stuffing is gone - an easy way to re-stuff would be to cut all the way up to the top of the final chamber - then stuffing for the middle chamber could be applied.) - IF anyone gets then I can't help lift anything very heavy with multiple hernia.
here's W1038R's general spec (I have the feeling 9mm was p-p but may have well been that much ?)
Re = 6.7 ohm
SPL – 88dB
Fs – 24Hz
Qts – .42
Vas -117L
Xmax – 9mm
Frame – 10.25 in
Cut out – 9 in
https://i.imgur.com/d9bz61o.png
here's W1038R's general spec (I have the feeling 9mm was p-p but may have well been that much ?)
Re = 6.7 ohm
SPL – 88dB
Fs – 24Hz
Qts – .42
Vas -117L
Xmax – 9mm
Frame – 10.25 in
Cut out – 9 in
https://i.imgur.com/d9bz61o.png
Last edited:
No clue who made them, but mine had quite different specs! Xmax was listed as 7.9 mm, but with only a ~13/16" [20 mm] VC winding height combined with just listening to them I'm sure it's p-p. For sure neither the 10" or 12" handled much power down low even in a corner.
Anyway, still have enough info to sim them both, so FWIW..........
GM
Anyway, still have enough info to sim them both, so FWIW..........
GM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.