OK, so I am tired of reading through everything with the search button and not really finding any answers to my question. How do you choose a speaker for a t/l? what makes a person choose one model over another? what are the limitations of a t/l? I think I understand how they work I just need a little more than a nudge... ( more like a shove) in the right direction.. I wish I could afford to go back to school for the missing knowledge just to support my hobby, it just isn't cost effective to do so. So reading and asking q's are are what I have to work with.
Mike
Mike
You guys need to spend some time over at Quarter Wavelength Loudspeaker Design studying Martin King's landmark work on modeling TL's. Click on his theory link and go thru all of his papers more than once and all your questions will be answered.
I'm starting on the T-Line road myself. I chose the Vifa PL18WO09 on a couple of recomendations for a cheap t-line driver and a SEAS Prestige 27TFFC (H0881) Textile Dome tweeter to match. I'd be interested if anyone has an opion on those choices!
Pearls from Martin J King Quarter Wave Design
Was a good place to start, after the basics of any speaker design.
Pearls from Martin J King Quarter Wave Design
Was a good place to start, after the basics of any speaker design.
Good lord.. I have a long way to go on this learning curve.. I will be back to this after I go back to school for degrees in physics and a lot more math.
I need a little matrix schooling plug me and away I go.
Discouraged! No! It may take me till I retire to figure it all out but someday. I think for now I will stay with the sealed and ported enclosures (I know these are not the proper terminology)that @ least I thought I understood. Some of you guys definately make me feel stupid on occasion. I didn't really come to appreciate sound reproduction until after listening to a very few demonstrations that blew me away. So what I do know and understand has all been from lots of reading ( I havent even gotten to the must read books yet) with a lot of trial and error, and the added perk of some small success that helps to keep me moving forward.
Mike
I need a little matrix schooling plug me and away I go.Discouraged! No! It may take me till I retire to figure it all out but someday. I think for now I will stay with the sealed and ported enclosures (I know these are not the proper terminology)that @ least I thought I understood. Some of you guys definately make me feel stupid on occasion. I didn't really come to appreciate sound reproduction until after listening to a very few demonstrations that blew me away. So what I do know and understand has all been from lots of reading ( I havent even gotten to the must read books yet) with a lot of trial and error, and the added perk of some small success that helps to keep me moving forward.
Mike
Last edited:
On the contrary Dave I do understand transmission lines very well.
What I don't understand is the constant if not assertion then at least implication that they posses some sort of mystical property other methods do not have.
I also find the assertion that if a person does not agree with this then they obviously do not understand the technique, laughable, to put the best face upon it.
rcw.
What I don't understand is the constant if not assertion then at least implication that they posses some sort of mystical property other methods do not have.
I also find the assertion that if a person does not agree with this then they obviously do not understand the technique, laughable, to put the best face upon it.
rcw.
All speaker designs use a mix of science and black art. In transmission lines the Black Art is a little more prominent than say a bass reflex which uses a little more black art than a sealed. If your going to build a T/L for the first time I suggest using a design already tried and tested, then you can experiment.
Terry
Terry
With the introduction of Martin King's Model & the accompanying software, the black arts involved became no greater than any other speaker enclosure.
Before MJK (and Augspurger's parallel model) TL designers followed a reciepe, and not a very good one.
One significant discovery was how, if you take a bass reflex with normal dimensions, and then start stretching the aspect ratio (keeping the port at the far end of the cabinet) the BR behaviour transitions into something completely different (dubbed an ML-TL). The impedance plot is broadly similar to a BR's but one look at the ANSYS simulations of the behaviour of the air inside the box quickly shows that box is doing something quite different (published in MJK's ML-TQWT article). More than one designer has found that their tower "bass-reflex" is not funcioning as simmed... it is because the box is not a BR it is an ML-TL.
Or you can take a strongly tapered pipe and damp fairly heavily to get a rool-off similar to a sealed box, but with a dramatically lower impedance peak. The example is somewhat less than could be achieved, but were constrained by other aspects of the box.
TLs in their broadest sense are very versatile and useful devices. Like any speaker design you need to choose your compromises.
Personally i avoid standard bass reflexes, i find that being tightly tuned (using a set of scalar T/S parameters) they move around in the bass as the T/S change as f(V). For a standard BR to work well you would have to choose a driver with flat T/S curves.
TLs are naturally tolerant, and you can mutate a BR to be more tolerant.
dave
Before MJK (and Augspurger's parallel model) TL designers followed a reciepe, and not a very good one.
One significant discovery was how, if you take a bass reflex with normal dimensions, and then start stretching the aspect ratio (keeping the port at the far end of the cabinet) the BR behaviour transitions into something completely different (dubbed an ML-TL). The impedance plot is broadly similar to a BR's but one look at the ANSYS simulations of the behaviour of the air inside the box quickly shows that box is doing something quite different (published in MJK's ML-TQWT article). More than one designer has found that their tower "bass-reflex" is not funcioning as simmed... it is because the box is not a BR it is an ML-TL.
Or you can take a strongly tapered pipe and damp fairly heavily to get a rool-off similar to a sealed box, but with a dramatically lower impedance peak. The example is somewhat less than could be achieved, but were constrained by other aspects of the box.
TLs in their broadest sense are very versatile and useful devices. Like any speaker design you need to choose your compromises.
Personally i avoid standard bass reflexes, i find that being tightly tuned (using a set of scalar T/S parameters) they move around in the bass as the T/S change as f(V). For a standard BR to work well you would have to choose a driver with flat T/S curves.
TLs are naturally tolerant, and you can mutate a BR to be more tolerant.
dave
Well it has been a while since I read about them, and I'm certainly no expert. Yes transmission lines don't posess some sort of mystical properties that other methods do not (in isolation). The thing that attracted me to TL's was that they apparently have the bass extention of a bass reflex, but had a transient response similar to a sealed enclosure.
A BR box trades off transient response for deeper bass, A closed box has much better transient response without the deeper bass (for similar sized cabinets and speakers) A transmission line as far as I know has both the deep bass and good transient response.. it is this property that I would personally consider to make TL's worth pursuing over other alignments.
Unfortunately when I looked into it my 10" woofers don't appear to be particularly suitable for a TL (didn't model well, probably because of the low QTS).
Mike as a starting point This bit from MJK's site might help you to choose a suitable driver: Choosing a driver with a QTS above 3.5 would probably be a good starting point. (mine have a qts of 3.2 and I really struggled to get a good model and decided to abandon the idea. Although there is heaps of math on MJK's site, you really don't need to understand it to do the modelling in mathcad, if you did need to I would probably not have bothered starting either
Whilst understanding the math can help with designing something, IMO it is not necessary if someone else has built a tool that takes care of the math for you Tony.
Last edited:
Good lord.. I have a long way to go on this learning curve.. I will be back to this after I go back to school for degrees in physics and a lot more math.
Discouraged! No! It may take me till I retire to figure it all out but someday. I think for now I will stay with the sealed and ported enclosures (I know these are not the proper terminology)that @ least I thought I understood. Some of you guys definately make me feel stupid on occasion. I didn't really come to appreciate sound reproduction until after listening to a very few demonstrations that blew me away. So what I do know and understand has all been from lots of reading ( I havent even gotten to the must read books yet) with a lot of trial and error, and the added perk of some small success that helps to keep me moving forward.
Mike
Don't loose heart. Bobocow's link simplifies MKK's thesis pretty well. Just read down to where the math starts and you should have a good understanding of the basics. Also, Keith Webb's Excel alignment tables found under MJK's theory page allow you to plug in your driver parameters and choice of SL/SO ratio and the rest is done for you based on King's original alignment tables.
I do prefer TL design to an acoustic suspension and BR cabinets. There are drawbacks to either design and nonliterary caused by each particular loading. IMO TL has a better transient response, can be build cheaper with better bass extension. I have build a few BR, some Acoustic Suspensions but TL became my choice for the last year or so. Either loading can be made to sound great or horrendous depending on the designer and driver\crossover implementation.
Normally I'd recommend drivers with 0.4 qt on average but some people prefer lower q drivers (and I have done so as well), some higher. X-mass is not so much of an issue because in the TL driver moves significantly less.
I do recommend reading G.L. Augspurger's articles.
Normally I'd recommend drivers with 0.4 qt on average but some people prefer lower q drivers (and I have done so as well), some higher. X-mass is not so much of an issue because in the TL driver moves significantly less.
I do recommend reading G.L. Augspurger's articles.
No. A closed TL (ie 1/2 wave) will have the roll-off of a sealed, a reduction in the impedance peak, and no time smear coming back thru the driver.
dave
Hi Dave,
A while ago but you contributed to one of my TL builds - HERE.
I used Martin's Mathcad worksheets for that one and a (much) bigger pair before those.
I guess you are reacting my use of "accurate" - I really meant "easy". MK's models are accurate but not as easy to work with as a BR modeling program, like Unibox.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.