TL = transmission line
In this context Lionel is referring to the pathway from the rear of the driver to the terminus (vent). He is recommending finding a way to not run the driver input wire through that pathway.
A discussion of wire management is always relative to speaker building. As with many things, placement of the binding posts (or not using them at all) is a compromise between the best from a performance view and best from a cosmetic or usability view. From a performance view the best place for the cabinet is question may be on top.
thanx Tim,
i never even thought of that. the wire and the little holes where the leads enter the cabinet/?? nope,it never ocoured to me.
so,how do i route the wires without bringing a harmfull "noise" into the box??( i have a feeling that a simple question does not have a simple answer)..
also Tim, what did you think of that speaker i put out? i said i thought it looked alot like a salesmans "sample" thing he might carry to show his wares,but that it might not be a "real" speaker. what do you think about it? you think it may be real?? if so,id think about building it. but i cant aford to go making somthing that wont be any good.
so what are your thoughts on the lil table speek??
roy
i never even thought of that. the wire and the little holes where the leads enter the cabinet/?? nope,it never ocoured to me.
so,how do i route the wires without bringing a harmfull "noise" into the box??( i have a feeling that a simple question does not have a simple answer)..
also Tim, what did you think of that speaker i put out? i said i thought it looked alot like a salesmans "sample" thing he might carry to show his wares,but that it might not be a "real" speaker. what do you think about it? you think it may be real?? if so,id think about building it. but i cant aford to go making somthing that wont be any good.
so what are your thoughts on the lil table speek??
roy
Roystr
I really like that construction method. I've always wanted to use it, but my speakers are always so big that it would take 20 layers- too much work! Of course the side layers can just be board also-no need for them to be plastic. I thin kthe original comment about the wires was from an appearance point of view.
Hey guys- wouldn't that be a good way to build our reference speaker?
It allows nice smooth radius turns. Costruction is either more or less complicated than box type construction depending on how nice it needs to look! Best way to make it I guess is tracing the pattern on 3/4" layers and cutting the path freehand with a router?
A bandsaw would be faster.- you coulalso use 1.5" MDF as it appears the photo ones are.
I really like that construction method. I've always wanted to use it, but my speakers are always so big that it would take 20 layers- too much work! Of course the side layers can just be board also-no need for them to be plastic. I thin kthe original comment about the wires was from an appearance point of view.
Hey guys- wouldn't that be a good way to build our reference speaker?
It allows nice smooth radius turns. Costruction is either more or less complicated than box type construction depending on how nice it needs to look! Best way to make it I guess is tracing the pattern on 3/4" layers and cutting the path freehand with a router?
A bandsaw would be faster.- you coulalso use 1.5" MDF as it appears the photo ones are.
hey variac, when i saw that pic
it was one of them "eureeka" moments for me. i never thought of doing it like that.what a time saver. make one perfect,then just trace it out with a router and make how ever many youll need. all the same ,all perfect.
my first thought was a bandsaw,to cut thicker pieces,but then your tracing a line,with the router,your not tracing a line,your tracing the first piece exactly. the only thing is,how many router bits would you go through???? and would you need a 3HP router?? i think they go up to 3.5. and with the good bits going for 60-90$ each,, if you burned a few up to fast,the cost would skyrocket in no time.
if i had a router,id try it that way.course youd better be sure you have the rite shape first,thats alot of wood and work if after you glue it all up it dont sound good.
that would stink
roy
it was one of them "eureeka" moments for me. i never thought of doing it like that.what a time saver. make one perfect,then just trace it out with a router and make how ever many youll need. all the same ,all perfect.
my first thought was a bandsaw,to cut thicker pieces,but then your tracing a line,with the router,your not tracing a line,your tracing the first piece exactly. the only thing is,how many router bits would you go through???? and would you need a 3HP router?? i think they go up to 3.5. and with the good bits going for 60-90$ each,, if you burned a few up to fast,the cost would skyrocket in no time.
if i had a router,id try it that way.course youd better be sure you have the rite shape first,thats alot of wood and work if after you glue it all up it dont sound good.
that would stink
roy
Having done routing with 3/4" plywood (for a very large desk with a plywood drop lip edge) I don't think power would be a big issue. I did that job with a 2 hp router and REALLY cheap bits and it worked fine...
Anyone else have some experience with this kind of composite layup technique they can share? It seems quite interesting for these kinds of unique shapes.
Anyone else have some experience with this kind of composite layup technique they can share? It seems quite interesting for these kinds of unique shapes.
tleonard said:
Quite a few have done speakers using the Translam technique.... Andy Graddon has a bit of a gallery -- andy was an early proponent of this tech --
http://users.tpg.com.au/users/gradds/curved/curved sided loudspeakers.htm
A number of commercial loudspeakers use this technique too.
It can be very wasteful of material thou.
dave
Gio,
While I tested the drivers right out of the box, the response graphs for the FE127E were done after 24 hours of "conditioning" using a low frequency sine wave just under driver power limit. This is the standard break in conditioning in the literature. You can find a cab tuning experiment using this break in procedure in the Loudspeaker Cookbook, sixth edition. Except for the expected decrease in Fs and increase in Vas, there was no measurable difference in performance.
Also, a prefilter attempts to correct problems with a driver or loudspeaker by a complementary altering the signal fed to the driver. Baffle step correction is an example of a prefilter.
I am also attaching a corrected version of the driver with prefilter response plot to this post. When I plotted the data, I used the incorrect time base. The top end response of the FE127E begins to die just over 16kHz instead of just under 20 kHz as shown in the previous graph. All the other points of reference (the peaks dips in response) are slightly shifted up in frequency. Sorry for the confusion.
Now, back to the first paragraph. Since that original period of conditioning, I have driven the driver with broadband test signals and swept sine waves as I attempted to make sense of the TL enclosure designs. There is, however, more number crunching than actual voicing of the driver. This entire process has added only a couple more hours of operation. Far less than the 200 hours that has been suggested in this thread.
What, however, should be the conditioning process for breaking in a driver? Will music break the driver in differently than a low frequency sine wave? If so, how important is the type of music and the level? Someone suggested raggae music would break the driver in faster. But wouldn't the loudness also matter? And since pop music is usually compressed and has a higher average to peak loudness in comparison to classical, wouldn't that mean that it would take much longer to achieve the break in period if you listen to classical?
I am rather skeptical of a 200 hour break in period. Think about it. First, if the driver continues to change over the first 200 hours, what possible reason do we have to believe it will stop changing after 200 hours?
Besides the question of break in period, I also have some questions and concerns about the TL designs posted to this thread. I have built and tested three of the cabinet designs: the cheese wedge for the FE126E, the tall, thin monopole for the FE127E, and the tombstone monopole for the FE127E. The testing and analysis is becoming a huge project with an almost unmanagable amount of data. The problem, however, is that across all the tests the driver and enclosures are not acting as I would expect or predict for a TL.
I would find it helpful if Tim and Dave would explain what they predict their enclosures will do to the response of the driver. What gains will the TL enclosures provide? Are there any negatives? How and where do they expect the enclosures to make the response of the driver better or worse? And lastly, what should the nearfield response from the driver be like and what should the nearfield response from the trasmission line aperture be like?
Mark
While I tested the drivers right out of the box, the response graphs for the FE127E were done after 24 hours of "conditioning" using a low frequency sine wave just under driver power limit. This is the standard break in conditioning in the literature. You can find a cab tuning experiment using this break in procedure in the Loudspeaker Cookbook, sixth edition. Except for the expected decrease in Fs and increase in Vas, there was no measurable difference in performance.
Also, a prefilter attempts to correct problems with a driver or loudspeaker by a complementary altering the signal fed to the driver. Baffle step correction is an example of a prefilter.
I am also attaching a corrected version of the driver with prefilter response plot to this post. When I plotted the data, I used the incorrect time base. The top end response of the FE127E begins to die just over 16kHz instead of just under 20 kHz as shown in the previous graph. All the other points of reference (the peaks dips in response) are slightly shifted up in frequency. Sorry for the confusion.
Now, back to the first paragraph. Since that original period of conditioning, I have driven the driver with broadband test signals and swept sine waves as I attempted to make sense of the TL enclosure designs. There is, however, more number crunching than actual voicing of the driver. This entire process has added only a couple more hours of operation. Far less than the 200 hours that has been suggested in this thread.
What, however, should be the conditioning process for breaking in a driver? Will music break the driver in differently than a low frequency sine wave? If so, how important is the type of music and the level? Someone suggested raggae music would break the driver in faster. But wouldn't the loudness also matter? And since pop music is usually compressed and has a higher average to peak loudness in comparison to classical, wouldn't that mean that it would take much longer to achieve the break in period if you listen to classical?
I am rather skeptical of a 200 hour break in period. Think about it. First, if the driver continues to change over the first 200 hours, what possible reason do we have to believe it will stop changing after 200 hours?
Besides the question of break in period, I also have some questions and concerns about the TL designs posted to this thread. I have built and tested three of the cabinet designs: the cheese wedge for the FE126E, the tall, thin monopole for the FE127E, and the tombstone monopole for the FE127E. The testing and analysis is becoming a huge project with an almost unmanagable amount of data. The problem, however, is that across all the tests the driver and enclosures are not acting as I would expect or predict for a TL.
I would find it helpful if Tim and Dave would explain what they predict their enclosures will do to the response of the driver. What gains will the TL enclosures provide? Are there any negatives? How and where do they expect the enclosures to make the response of the driver better or worse? And lastly, what should the nearfield response from the driver be like and what should the nearfield response from the trasmission line aperture be like?
Mark
Attachments
hiya mark,
wow thats alot of information. someday ill be able to read it and understand it.
what im curious about is your referance to TL speakers.. is that the plan for the speaker were supposed to build with the 127? or do we build anything we want as long as its a single driver unit??
and do you have a website so i can see the ones you mention in the above article? how theyr built,what the measurements are,ect..
how did the 127 perform overall? you said you gathered quite a bit of info on the tests. but didnt mention how it turned out. was it (using the 127) a waste of time? sould you have tried a different driver instead?
thanx,
roy
wow thats alot of information. someday ill be able to read it and understand it.
what im curious about is your referance to TL speakers.. is that the plan for the speaker were supposed to build with the 127? or do we build anything we want as long as its a single driver unit??
and do you have a website so i can see the ones you mention in the above article? how theyr built,what the measurements are,ect..
how did the 127 perform overall? you said you gathered quite a bit of info on the tests. but didnt mention how it turned out. was it (using the 127) a waste of time? sould you have tried a different driver instead?
thanx,
roy
Roy,
This is a really long thread and I realize that it can take a while to read it all.
Dimensioned drawings of all three enclosures are a part of this thread. I believe the three enclosures I mentioned appear within the first 200 posts. Further, and only a couple of pages ago, I believe Gio restated the purpose of this thread. As I understood it, it was for everyone to build the same loudspeaker. If we do not, then we really cannot have a common reference sound.
As I read this thread, those dominating the discussion have decided on the FE127E in a transmission line enclosure.
Recently, there have been posts about plexiglass enclosures utilizing horn shaped apertures for the TL. ( Just for clarification, in the historical literature, a horn is a type of TL.) I feel that these plexiglas posts and Dave's latest series of enclosure drawings is providing far too many choices for enclosure when we have no good data on the performance variables of any of the enclosures. But then, that is a common problem with all discussion formats. Talk is always faster than doing.
I am just trying to better understand the intent of the early enclosure designs before I will do anything with the more difficult to build plexiglass rear horn TLs. I also want to hear more detail about the theory and design without being contaminated by actual performance results.
Best,
Mark
This is a really long thread and I realize that it can take a while to read it all.
Dimensioned drawings of all three enclosures are a part of this thread. I believe the three enclosures I mentioned appear within the first 200 posts. Further, and only a couple of pages ago, I believe Gio restated the purpose of this thread. As I understood it, it was for everyone to build the same loudspeaker. If we do not, then we really cannot have a common reference sound.
As I read this thread, those dominating the discussion have decided on the FE127E in a transmission line enclosure.
Recently, there have been posts about plexiglass enclosures utilizing horn shaped apertures for the TL. ( Just for clarification, in the historical literature, a horn is a type of TL.) I feel that these plexiglas posts and Dave's latest series of enclosure drawings is providing far too many choices for enclosure when we have no good data on the performance variables of any of the enclosures. But then, that is a common problem with all discussion formats. Talk is always faster than doing.
I am just trying to better understand the intent of the early enclosure designs before I will do anything with the more difficult to build plexiglass rear horn TLs. I also want to hear more detail about the theory and design without being contaminated by actual performance results.
Best,
Mark
Roy,
I don't understand what you are trying to learn? I understood your questions to be about the decisions/direction of this thread.
Mine has only been a very minor voice in the history of this thread. The decisions about the direction of the thread have been made by others. If that was the intent of your questions, then Tim would be the best person to ask.
If your questions were about loudspeaker designing and building, then the answers may not be found in this thread.
For example, you said, "you said you gathered quite a bit of info on the tests. but didnt mention how it turned out. was it (using the 127) a waste of time? sould you have tried a different driver instead?"
By transmission line theory the important variable is the length of the line and the quarter wave length of driver or system Fs. Since there will be some variance of Fs from one example to another of a particular driver, it would seem that there must be some degree of tolerance in the length of line to Fs. As such, another driver with similar Fs may work just as well in the same enclosure design. If, however, the Fs of the driver is not so close, perhaps only a small adjustment in the length of the line is necessary.
If, however, the enclosures presented in this thread are hybrid designs, utilizing variables other than that of a pure TL design, then they may be much more sensitive to driver selection. That is one of the things I would like to hear from Dave and Tim.
So, should I have tried a different driver? I do not know. The dominant voices in this thread have already made a group purchase and have the drivers. For those new FE127E owners at least, the decision of driver selection has been made. With this consideration, my contribution seems limited to documenting the performance of that driver and whatever enclosure is eventually set. While I would not have chosen the FE127E, I did say I would try to help those who did, at least to some degree. Would optimizing an enclosure for the 127 be a waste of time? I do not believe it would be.
If I may be allowed to make a suggestion to the entire group at this point, it would be to start talking to Fostex about improving the cone vibration performance of their drivers. The sound would be much improved and that improved sound available to more people if only the manufacturer produced a better product. It can be done.
If you would want to buy a better 127, write to Fostex and ask them to do better.
Best,
Mark
Let me just add one final thing. If I am a "stereo hound," (or in other words a dog and not a human being), then WOOF, WOOF.
Of course, dogs do have excellent hearing.
I don't understand what you are trying to learn? I understood your questions to be about the decisions/direction of this thread.
Mine has only been a very minor voice in the history of this thread. The decisions about the direction of the thread have been made by others. If that was the intent of your questions, then Tim would be the best person to ask.
If your questions were about loudspeaker designing and building, then the answers may not be found in this thread.
For example, you said, "you said you gathered quite a bit of info on the tests. but didnt mention how it turned out. was it (using the 127) a waste of time? sould you have tried a different driver instead?"
By transmission line theory the important variable is the length of the line and the quarter wave length of driver or system Fs. Since there will be some variance of Fs from one example to another of a particular driver, it would seem that there must be some degree of tolerance in the length of line to Fs. As such, another driver with similar Fs may work just as well in the same enclosure design. If, however, the Fs of the driver is not so close, perhaps only a small adjustment in the length of the line is necessary.
If, however, the enclosures presented in this thread are hybrid designs, utilizing variables other than that of a pure TL design, then they may be much more sensitive to driver selection. That is one of the things I would like to hear from Dave and Tim.
So, should I have tried a different driver? I do not know. The dominant voices in this thread have already made a group purchase and have the drivers. For those new FE127E owners at least, the decision of driver selection has been made. With this consideration, my contribution seems limited to documenting the performance of that driver and whatever enclosure is eventually set. While I would not have chosen the FE127E, I did say I would try to help those who did, at least to some degree. Would optimizing an enclosure for the 127 be a waste of time? I do not believe it would be.
If I may be allowed to make a suggestion to the entire group at this point, it would be to start talking to Fostex about improving the cone vibration performance of their drivers. The sound would be much improved and that improved sound available to more people if only the manufacturer produced a better product. It can be done.
If you would want to buy a better 127, write to Fostex and ask them to do better.
Best,
Mark
Let me just add one final thing. If I am a "stereo hound," (or in other words a dog and not a human being), then WOOF, WOOF.
Of course, dogs do have excellent hearing.
Tim is responsible for the designs, i just turned them into pretty pictures. The designs were generated using Martin king's worksheets and are all ML-TLs or ML-TQWTs (or in Augspurger terminology, TLs with restricted terminus). These seem to be better suited for low Q drivers than a straight TL which likes a higher Q. Results from the MJK sheets have proven to be fairly accurate. The limitation to the design will be the accuarcy of the T/S parameters used (in this case the factory specs). Given that the output of the sheets is dependednt on the things the designer tries out, i am sure there are other suitable designs hiding out there in quarter-wave space.
Fortunately the MLs are fairly tolerant of driver variation, and can be tuned both with stuffing and the restricted terminus (ie port).
All the existing design suggestions are still beta with desgnations less than v1. As input comes in from those building actual boxes further refinement, even different designs, or even drivers may have to be considered. What this thread will contain when we are done -- besides a "family" of related designs (no one design will suit all purposes), is a rich record of the design process, with the option of people taking off on a tangent when they see something of particular interest to us....
I see roystr's comments perhaps leading us toward potential horn designs -- perhaps for the FE126 since it will probably be happiest in a horn.
dave
Fortunately the MLs are fairly tolerant of driver variation, and can be tuned both with stuffing and the restricted terminus (ie port).
All the existing design suggestions are still beta with desgnations less than v1. As input comes in from those building actual boxes further refinement, even different designs, or even drivers may have to be considered. What this thread will contain when we are done -- besides a "family" of related designs (no one design will suit all purposes), is a rich record of the design process, with the option of people taking off on a tangent when they see something of particular interest to us....
I see roystr's comments perhaps leading us toward potential horn designs -- perhaps for the FE126 since it will probably be happiest in a horn.
dave
Thanks for the clarification.
Explains why they are testing like degraded ported/tuned vented boxes. For the 127, besides the cone loading problems of the enclosures, they can be made to perform much better by changing the amount and placement of the internal filler material.
I assume ML stands for mid line. If it does, then the testing is showing a possible problem with MLs. When fully damped, it is small and therefore probably can be ignored. When fully damped, however, low frequency output is way down. Reducing damping brings up the bass, but also shows much increased magnitude of a spectral repetitive narrow bandwidth dip and then peak. First instance occurs around 400 to 500 Hz.
Mark
Explains why they are testing like degraded ported/tuned vented boxes. For the 127, besides the cone loading problems of the enclosures, they can be made to perform much better by changing the amount and placement of the internal filler material.
I assume ML stands for mid line. If it does, then the testing is showing a possible problem with MLs. When fully damped, it is small and therefore probably can be ignored. When fully damped, however, low frequency output is way down. Reducing damping brings up the bass, but also shows much increased magnitude of a spectral repetitive narrow bandwidth dip and then peak. First instance occurs around 400 to 500 Hz.
Mark