i`m intending to build a TWQT that was ment for a Fostex FE103E.
but becuase the FE103 is small, i`ve decided to try to replace it with a FE126E.
they differ mainy with the FE126 having a lower Fs and a much lower Qts. from 0.43 to 0.25. the Qms is lower but the Qes is higher.
would the FE126 work in a TWQT ment for a FE103.
i try figuring it out by using martin kings mathscad sheets but to no avail. after reading a couple tutorials, i don`t think it would make a large difference.. is that right?
but becuase the FE103 is small, i`ve decided to try to replace it with a FE126E.
they differ mainy with the FE126 having a lower Fs and a much lower Qts. from 0.43 to 0.25. the Qms is lower but the Qes is higher.
would the FE126 work in a TWQT ment for a FE103.
i try figuring it out by using martin kings mathscad sheets but to no avail. after reading a couple tutorials, i don`t think it would make a large difference.. is that right?
No bass is the result. A true TL mimics the response of a sealed box; an easy way to see whether a driver is suitable for a TL to the cutoff you want to achieve is to model the driver in an optimal size sealed box. If it won't work sealed it won't work TL. With a Qts of .3 or less you really have to go to a horn if you want strong bass response.
Yes, as already noted it would make a big difference, though you can use a series resistor to increase the effective Qts to match the 103's higher Qts, with the attendent loss of efficiency.
To calc a first approximation resistor value if actual specs are not known:
Qes' = Qes*(1+Rs/Rvc), where Rs is the series
resistance and Rvc is the driver's DC resistance, so:
Rs = (Qes'/Qes-1)*Rvc. Remember to include the series
resistances of your amplifier/cable/inductors in Rs,
so the actual resistor you buy may be somewhat
smaller than calculated value.
Qts' = Qes'*Qms/(Qes'+Qms)
Vas is unchanged
n0 = 9.6352*10^-10*Fs^3*Vas(liters)/Qes'
SPL = 112.018+10*Log(n0)
Unless Fs and Vas are a fairly close match to the 103 (couldn't find specs for the 126), it could be either an overdamped ('lean') or underdamped ('fat') sounding speaker, though experimenting with damping can lean out an underdamped design somewhat.
As noted though, a back horn design for this specific driver is the best cab to use if space and building complexity are not major issues.
GM
I am afraid I have to totally disagree with BillFitzmaurice on his opinion of low Qts drivers.
Over the years I have built a number of TL's for my self and friends, and apart from the first ones they all used low Q drivers.
My present speakers are a two-way using Seas CB17 RCY with a spec Qts of 0.25. By BillFm criteria I should have little or no bass.
WRONG, my speakers have a deep tight and well controlled bottom end. Even at low volumes they still provide excellent definition. I can not speak for Fostex but the drivers I have used work superbly in TL's.
A while ago I built a friend a pair of of TL's which used Audax 8in bass/mid with a Qts of 0.31 ( still low! ). He then aquired a pair of Scanspeak 8554 units with a spec Qts of 0.22, with a view to a new TL. As he used a simple first order XO on the bass we decided to dropin the 8554 to try it out.
The result proved a revelation. Not only was the mid clarity and imaging improved but the bass took on a greater level of depth and definition, not less !! This was further helped no dought, by the Scans have lower Fs and larger Vas, but still very low Qts
There are a lot of excllent TL's out there using low Qts drivers. One of the first was the oringinal Daline's desinged by Robert Fris, which used the Kef B110 which had a Q of 0.31. These little were capable of generating amazing bass in those smallish Tl's
Amoung the benifits of low Q drivers is they are less trobled by room placement . I and my friends have found the enclosures can placed close to a rear wall and still retain a controlled bass. A cause for domestic bliss !
I have always found that the so called classic TL with well stuffed lines can sap the life out of music, with low Q drivers you can reduce the amount of damping. David L Field expressed the same opinion as above ( though better written ) in his article in Speaker Builder 4/96.
In search for a better understanding of TL's, I had several emails from Martin King answering my questions. One reply stated " I also like the low Qts drivers and have found similar results. The higher Qts " 0.3- 0.4 !!, have a bigger bass but not as tight or controlled "
With my prefered drivers I have found I only need to line the walls with wool felt or profiled foam. The highly regarded PMC TL's use the same arrangment so I am in good company
AS I mentioned if you check out the net for TL's you will find that the majority use the lower 0.3-0.35 Q drivers. So I would try out those low Q Fosteks you might be pleasantly surprised
Over the years I have built a number of TL's for my self and friends, and apart from the first ones they all used low Q drivers.
My present speakers are a two-way using Seas CB17 RCY with a spec Qts of 0.25. By BillFm criteria I should have little or no bass.
WRONG, my speakers have a deep tight and well controlled bottom end. Even at low volumes they still provide excellent definition. I can not speak for Fostex but the drivers I have used work superbly in TL's.
A while ago I built a friend a pair of of TL's which used Audax 8in bass/mid with a Qts of 0.31 ( still low! ). He then aquired a pair of Scanspeak 8554 units with a spec Qts of 0.22, with a view to a new TL. As he used a simple first order XO on the bass we decided to dropin the 8554 to try it out.
The result proved a revelation. Not only was the mid clarity and imaging improved but the bass took on a greater level of depth and definition, not less !! This was further helped no dought, by the Scans have lower Fs and larger Vas, but still very low Qts
There are a lot of excllent TL's out there using low Qts drivers. One of the first was the oringinal Daline's desinged by Robert Fris, which used the Kef B110 which had a Q of 0.31. These little were capable of generating amazing bass in those smallish Tl's
Amoung the benifits of low Q drivers is they are less trobled by room placement . I and my friends have found the enclosures can placed close to a rear wall and still retain a controlled bass. A cause for domestic bliss !
I have always found that the so called classic TL with well stuffed lines can sap the life out of music, with low Q drivers you can reduce the amount of damping. David L Field expressed the same opinion as above ( though better written ) in his article in Speaker Builder 4/96.
In search for a better understanding of TL's, I had several emails from Martin King answering my questions. One reply stated " I also like the low Qts drivers and have found similar results. The higher Qts " 0.3- 0.4 !!, have a bigger bass but not as tight or controlled "
With my prefered drivers I have found I only need to line the walls with wool felt or profiled foam. The highly regarded PMC TL's use the same arrangment so I am in good company
AS I mentioned if you check out the net for TL's you will find that the majority use the lower 0.3-0.35 Q drivers. So I would try out those low Q Fosteks you might be pleasantly surprised
My comments were based upon the groundbreaking work in TL design by George Augsperger, who has backed up his opinions with measured results that display the superiority of high Qts drivers in true transmision lines. The TLines I've built and measured cause me to agree with his conclusions. Keep in mind this refers to true 1/4 wavelength TLs, and not TL hybrids, which I have neither built nor measured.
Damping in a true TL is not there for the purpose of altering bass response, it's for damping midrange response bumps. In the case of Field's T-Rex there is a very good reason why he found it worked better not totally stuffed: it was not a transmission line. It was actually a hybrid TL/reflex, and therefore it didn't work as either a TL or a reflex, but as something in between.
Bill
You wrote :
I just skimmed through George Augspurger's AES paper on TL theory and design. I could not find any claim that high Qts drivers were prefered in TLs. Where specifically did you find statements in the AES paper that high Qts drivers are superior? If I look at his alignment tables, he addresses Qts values between 0.31 and 0.58. It would seem to me that Roy's observations are consistent with the results obtained by Augspurger.
George Augspurger did some very good work on TL theory and design. In his paper he has selected a set of variables to optimize and arrive at a set of alignment tables consistent with these assumptions. But they are not the only set of alignment tables that could be generated. A different set of assumptions would result in different geometry alignments. If you are designing a BR box for a given driver, there are several different filter alignments that can be selected each with its own strengths and weaknesses and each yielding a different box size and port length. I consider George Augspurger's articles to be one piece of the TL theory puzzle but not all encompassing documents that have to be followed explicitly.
You wrote :
I just skimmed through George Augspurger's AES paper on TL theory and design. I could not find any claim that high Qts drivers were prefered in TLs. Where specifically did you find statements in the AES paper that high Qts drivers are superior? If I look at his alignment tables, he addresses Qts values between 0.31 and 0.58. It would seem to me that Roy's observations are consistent with the results obtained by Augspurger.
George Augspurger did some very good work on TL theory and design. In his paper he has selected a set of variables to optimize and arrive at a set of alignment tables consistent with these assumptions. But they are not the only set of alignment tables that could be generated. A different set of assumptions would result in different geometry alignments. If you are designing a BR box for a given driver, there are several different filter alignments that can be selected each with its own strengths and weaknesses and each yielding a different box size and port length. I consider George Augspurger's articles to be one piece of the TL theory puzzle but not all encompassing documents that have to be followed explicitly.
my drivers have a Qts of 0.44. Fs of 17hz and Vas of 380 liters, Sd of about 420cm2, Xmax (one way linear) of 1 cm, Mms of 80gms and BL of 7.8Tm. Other specs are also available but i dont think they are signifcant to my question.
what determines the length and area of a TL? What are the advantages/disadvantages of tapering the TL from say 1.5 X Sd to say 0.75 x Sd?
I intend to built a TL that is say 100" tall with an area of say (35cmx12cm) 420cm2 (Sd) that tapers to (35cm x 7cm) 250cm2 (0.6 x Sd) at the exit. Is this too much of a taper?
The lines will be placed against a wall and stood on end like tapered columns.
what determines the length and area of a TL? What are the advantages/disadvantages of tapering the TL from say 1.5 X Sd to say 0.75 x Sd?
I intend to built a TL that is say 100" tall with an area of say (35cmx12cm) 420cm2 (Sd) that tapers to (35cm x 7cm) 250cm2 (0.6 x Sd) at the exit. Is this too much of a taper?
The lines will be placed against a wall and stood on end like tapered columns.
Navin,
My alignment tables would give you a quick sizing of many different tapered enclosures. For further optimization of the length and areas, I would recommend running the MathCad worksheets starting with the alignment table results. Unfortunately there is no free lunch and some work has to be done to answer your questions. Don't expect instant answers, sometimes I spend several weeks (an hour here and an hour there) tweaking a design until I am satisfied that it has enough potential to warrent building it.
My alignment tables would give you a quick sizing of many different tapered enclosures. For further optimization of the length and areas, I would recommend running the MathCad worksheets starting with the alignment table results. Unfortunately there is no free lunch and some work has to be done to answer your questions. Don't expect instant answers, sometimes I spend several weeks (an hour here and an hour there) tweaking a design until I am satisfied that it has enough potential to warrent building it.
Augspergers's reference to high Qts drivers is noted in his 2/2000-4/2000 Speaker Builder articles, in which he defines the relationship between Qts and F3 in a TL. His tables only go as high as a .58 Qts as a matter of practicality, since quality drivers with Qts figures higher than that are not near as common as those with lower Qts figures. What is made perfectly clear is that in a pure TL the F3/Qts relationship is concise and predictable, and as Qts is raised the F3 of a TL of finite length is lowered, in the same fashion as holds true in a sealed alignment.
Specifically in the 4/2000 article he notes:" it is a little discouraging to see that f3 is 30% higher than fs at best, and this ratio requires Qts to be at least 0.5" and: "By using a driver with Qts greater than .707, it is also possible to extend low-frequency response below F3..."
These observations weren't new with Augsperger, he was just the first to quantify them with formulae. Bailey's 1972 design used a driver with a Qts of .6, and in the aforementioned T-Rex article by Field he states: " A traditional TL should have a high Qts driver..." as part and parcel of why his T-Rex was not a true TL, but in fact a hybrid reflex/TL intended to make better use of a lower Qts driver than a pure TL was able to.
This is not to say that there aren't problems attendant with higher Qts drivers, but that should one wish to construct a pure TL with the aim of achieving strong bass response one of the requirements in doing so is to use a driver with a high enough Qts.
MJK said:
MJK, I am sorry if i gave the impression that i expected someone to design a TL for me. I just gave as much info as I though anyone would need to decide if making a TL using these drivers (audio concepts DV12) made sense. thanks for the website link.
ok. you got me confused. 228 liters. wow. that is huge! 8 cu. ft. wow! I was looking at something half that size or less.BillFitzmaurice said:
Bill,
I'll have to dig back through my back issues of SB to find the articles. I am surprised that he came to that conclusion, I definitely don't agree. I have heard TL's built with lower Qts drivers (0.3-0.4) and also one that was built with a higher Qts driver (~0.7). All of the drivers were quality products, so my comparison is not based on one low quality high Qts unit. These were all "classic" TL geometries, no hybrids or other variants.
I found that the bass reproduced by the high Qts driver TL design was louder and more plentiful then the others but at the same time very bloated and boomy. The control and detail was not as good as produced by the lower Qts designs. Based on many simulations, using many different Qts values, I think the systems were typical of what could be expected. So I guess it is up to the DIYer which result they are after. My strong preference is to use a driver with a Qts between 0.3 and 0.5. Obviously your preference is different.
The problem I have with the advice you provide in your many TL posts is that you phrase your preference for high Qts drivers as a fact of nature that cannot be questioned. In reality, your advice represents one point of view that is based on a set of assumptions made by George Augspurger leading to a group of alignments in his articles. There are other assumptions that could be made leading to different alignments and alternate recommendations for driver Qts values. Everything is a compromise and there is no clear black and white answer to which Qts driver will work best. Different people will have different objectives and requirements for bass reproduction in a TL design.
I'll have to dig back through my back issues of SB to find the articles. I am surprised that he came to that conclusion, I definitely don't agree. I have heard TL's built with lower Qts drivers (0.3-0.4) and also one that was built with a higher Qts driver (~0.7). All of the drivers were quality products, so my comparison is not based on one low quality high Qts unit. These were all "classic" TL geometries, no hybrids or other variants.
I found that the bass reproduced by the high Qts driver TL design was louder and more plentiful then the others but at the same time very bloated and boomy. The control and detail was not as good as produced by the lower Qts designs. Based on many simulations, using many different Qts values, I think the systems were typical of what could be expected. So I guess it is up to the DIYer which result they are after. My strong preference is to use a driver with a Qts between 0.3 and 0.5. Obviously your preference is different.
The problem I have with the advice you provide in your many TL posts is that you phrase your preference for high Qts drivers as a fact of nature that cannot be questioned. In reality, your advice represents one point of view that is based on a set of assumptions made by George Augspurger leading to a group of alignments in his articles. There are other assumptions that could be made leading to different alignments and alternate recommendations for driver Qts values. Everything is a compromise and there is no clear black and white answer to which Qts driver will work best. Different people will have different objectives and requirements for bass reproduction in a TL design.
What you note brings up the other edge of the sword: that high Qts drivers also may have other characteristics that make them less than ideal, most notably a low Bl, which is not the best situation when you get below 40 Hz or so.
There comes a point in the quest for bass where you can get too much; I've found that because of room lift issues that going flat to lower than 40 Hz isn't that good of an idea, but try to tell that to someone who is bound and determined to be flat to 8 Hz anechoic. If someone does insist on getting flat to lower than 40 Hz with any box the response could get booming and flabby, and low Qts drivers will certainly assist in that result.
For that reason I don't use a TL for my sub; I use a horn that is quite comfortable with a .25 Qts driver. I use TLs for my mains, with .7 Qts drivers, because they work quite well to 80 Hz in a 38 inch line and don't exhibit any bad habits so employed.
My personal feeling is that TLs are fine to a 40 Hz or so f3, but that below that the advantage shifts to a horn, with low Qts high Bl drivers. On the other hand, if one desires to run a TL 'all the way down' then a Qts of at least .4 is required to do it, short of using a twenty foot pipe. Whether it will actually sound good is a different question entirely.
There comes a point in the quest for bass where you can get too much; I've found that because of room lift issues that going flat to lower than 40 Hz isn't that good of an idea, but try to tell that to someone who is bound and determined to be flat to 8 Hz anechoic. If someone does insist on getting flat to lower than 40 Hz with any box the response could get booming and flabby, and low Qts drivers will certainly assist in that result.
For that reason I don't use a TL for my sub; I use a horn that is quite comfortable with a .25 Qts driver. I use TLs for my mains, with .7 Qts drivers, because they work quite well to 80 Hz in a 38 inch line and don't exhibit any bad habits so employed.
My personal feeling is that TLs are fine to a 40 Hz or so f3, but that below that the advantage shifts to a horn, with low Qts high Bl drivers. On the other hand, if one desires to run a TL 'all the way down' then a Qts of at least .4 is required to do it, short of using a twenty foot pipe. Whether it will actually sound good is a different question entirely.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.