Hi,
Here I go again with my new project. After the success of my Detaché project, I will try to get more bass.
Now I have another beyma product: The 6B30P, a 6,5" poly woofer. Fs=55Hz. Vas=18l. Qts=0.62 Xmax=4mm. 90dB/m.
I liked very, very much the 5MP60/N, so I'll try its big brother.
I'm trying to design a TL, but I'd like to clarify some points:
- The TL target is to get rid of the backwave as long as the wavelenght shorter than Fs is eaten in the tube by the filling.
- The other purpose is to extend the bass response, since the wave that is not eaten exits through the "port", adding low output to the main wave.
Is this correct?
If so, at what frequency should I tune the 1/4 of the wavelenght? to the resonant frequency of the driver on free air? Or maybe to the resonant frequency of the driver on the selected enclosure? Why should it be to the resonant freq. of the driver? Taking the bass reflex as example: I can fine tune the port to my taste, getting plain response, peaked one, or whatever I'd like, not exactly to Fs. Should this be applied to the TL?
Thanks,
Here I go again with my new project. After the success of my Detaché project, I will try to get more bass.
Now I have another beyma product: The 6B30P, a 6,5" poly woofer. Fs=55Hz. Vas=18l. Qts=0.62 Xmax=4mm. 90dB/m.
I liked very, very much the 5MP60/N, so I'll try its big brother.
I'm trying to design a TL, but I'd like to clarify some points:
- The TL target is to get rid of the backwave as long as the wavelenght shorter than Fs is eaten in the tube by the filling.
- The other purpose is to extend the bass response, since the wave that is not eaten exits through the "port", adding low output to the main wave.
Is this correct?
If so, at what frequency should I tune the 1/4 of the wavelenght? to the resonant frequency of the driver on free air? Or maybe to the resonant frequency of the driver on the selected enclosure? Why should it be to the resonant freq. of the driver? Taking the bass reflex as example: I can fine tune the port to my taste, getting plain response, peaked one, or whatever I'd like, not exactly to Fs. Should this be applied to the TL?
Thanks,
Hello Raka,
If you would like to design your TL, and get some idea of the performance before you actually build it, then you might consider trying my MathCad worksheets. Many people have used them and have built TL's that worked as predicted by the math model results.
If you would like to design your TL, and get some idea of the performance before you actually build it, then you might consider trying my MathCad worksheets. Many people have used them and have built TL's that worked as predicted by the math model results.
Worksheets
Thanks for your work!
I've downloaded your worksheets and the Mathcad explorer 8. I opened the closed box file, and no graph is drawn. Do I have to do something? I haven't written any number yet, is the example complete? with the given values of the focal unit, in the Qtd field appears: "This variable or function is not defined above". Only if I write a Qed value >0 then it works. Anyway, in the graphs doesn't appear a graphic, just blank.
Help with mathcad!! I'm an excel boy!!
Thanks.
Thanks for your work!
I've downloaded your worksheets and the Mathcad explorer 8. I opened the closed box file, and no graph is drawn. Do I have to do something? I haven't written any number yet, is the example complete? with the given values of the focal unit, in the Qtd field appears: "This variable or function is not defined above". Only if I write a Qed value >0 then it works. Anyway, in the graphs doesn't appear a graphic, just blank.
Help with mathcad!! I'm an excel boy!!
Thanks.
Hi Raka,
All of the files should run when you open them, graphs should calculate as you scroll down. Does your computer use a comma or a period for a decimal point? It needs to use a period, if it is set to use a comma you need to change the computer set-up. This is the most common problem I have encountered in running the worksheets and it exhibits the behavior you describe.
All of the files should run when you open them, graphs should calculate as you scroll down. Does your computer use a comma or a period for a decimal point? It needs to use a period, if it is set to use a comma you need to change the computer set-up. This is the most common problem I have encountered in running the worksheets and it exhibits the behavior you describe.
I found out
I've already found the way. Just to change the country I live in.
Works fine your software. Now I need to use the files in order to get used to it and develop my design. I'll need your help, so be prepared to answer my boring questions!
Thanks a lot for sharing your knowledge.
I've already found the way. Just to change the country I live in.
Works fine your software. Now I need to use the files in order to get used to it and develop my design. I'll need your help, so be prepared to answer my boring questions!
Thanks a lot for sharing your knowledge.
Hi Raka,
Sounds like you are on your way, I am glad the fix was easy. Please do not hesitate to ask any questions you might have, I may know the answer to a few of them.
Sounds like you are on your way, I am glad the fix was easy. Please do not hesitate to ask any questions you might have, I may know the answer to a few of them.
I can tell you that there are many people watching this thread as it develops into a project. Keep us posted!
Dave
Dave
Any Hint?
I've been playing with your software, and seems to be so powerful that fears me (I don't know what direction should be followed). Well, as far as I've found out, the problem of the TL is to tune it until it loads acoustically the driver in a proper way. That gets a lower resonance of the driver in the enclosure (or damp it), and make the waves inside to behave, and to add output as desired.
Is this correct?
It looks like if a checkpoint for a good design is a flat impedance curve in the low end, isn't it?
I'm starting with a lenght equal to the 1/4 wavelenght of the free air Fs, 1.25Sd to 1Sd and constant stuffing. Are those good figures to start with?
I'm still reading and understanding your text, so be indulgent with my questions...
What is n_open?
In the "transmission line definition", I don't have to introduce any value, do I?
I've been playing with your software, and seems to be so powerful that fears me (I don't know what direction should be followed). Well, as far as I've found out, the problem of the TL is to tune it until it loads acoustically the driver in a proper way. That gets a lower resonance of the driver in the enclosure (or damp it), and make the waves inside to behave, and to add output as desired.
Is this correct?
It looks like if a checkpoint for a good design is a flat impedance curve in the low end, isn't it?
I'm starting with a lenght equal to the 1/4 wavelenght of the free air Fs, 1.25Sd to 1Sd and constant stuffing. Are those good figures to start with?
I'm still reading and understanding your text, so be indulgent with my questions...
What is n_open?
In the "transmission line definition", I don't have to introduce any value, do I?
Re: Any Hint?
If tapering the line better to start with 3 or 4 Sd tapering to Sd (the line volume is actually not related to Sd but to Vas {which does tend to go up with Sd}). A narrowing taper moves the resonant modes of the line such that you will find that a line shorter than 1/4 Fs is more appropriate.
dave
Raka said:
If tapering the line better to start with 3 or 4 Sd tapering to Sd (the line volume is actually not related to Sd but to Vas {which does tend to go up with Sd}). A narrowing taper moves the resonant modes of the line such that you will find that a line shorter than 1/4 Fs is more appropriate.
dave
Ok, I'll try and answer at least some of the questions the best I can. It was a busy day at my day job so I could not respond sooner. Working for a living is really starting to get in the way of the good stuff.
Raka wrote :
"It looks like if a checkpoint for a good design is a flat impedance curve in the low end, isn't it?
I'm starting with a lenght equal to the 1/4 wavelenght of the free air Fs, 1.25Sd to 1Sd and constant stuffing. Are those good figures to start with?
I'm still reading and understanding your text, so be indulgent with my questions...
What is n_open?
In the "transmission line definition", I don't have to introduce any value, do I?"
(MJK) I do not use the impedance plot as any indication of the quality of the TL, I look directly at the SPL plots. I would recommend a visit to Bob Brines' site (linked from mine) where he has an excellent study of how various parameters impact a TL. He also has recommendations on starting geometries. He has done a much better job then I could ever do in this reply.
You do not neet to touch any of the entries on the second page of most of my worksheets. They are pre-formatted with defaults. The only exception is the "TL Sections" worksheet which is 100% user input. The only time you need to input data on page two is when you are tweaking the last little bit of a design or trying to make your own custom enclosure.
Dave (Kneadle) wrote :
"When does shortening cause a T-Line to be something else, like a rear-ported BR?"
(MJK) I call my methods "quarter wavelength loudspeaker design" because I cannot draw a sharp line between the labels typically assigned to a speaker enclosure. If I give a definition of BR there are people that would argue on both sides of the definition and nobody would be 100% correct. The line between a TL, a ML TL, and a BR is not clear to me. All of the worksheets on my site include standing waves in the long direction of the enclosure and not the traditional lumped parameter modeling used in most T/S programs. This makes definitions confusing and in my opinion not too important.
Hope that helps,
Raka wrote :
"It looks like if a checkpoint for a good design is a flat impedance curve in the low end, isn't it?
I'm starting with a lenght equal to the 1/4 wavelenght of the free air Fs, 1.25Sd to 1Sd and constant stuffing. Are those good figures to start with?
I'm still reading and understanding your text, so be indulgent with my questions...
What is n_open?
In the "transmission line definition", I don't have to introduce any value, do I?"
(MJK) I do not use the impedance plot as any indication of the quality of the TL, I look directly at the SPL plots. I would recommend a visit to Bob Brines' site (linked from mine) where he has an excellent study of how various parameters impact a TL. He also has recommendations on starting geometries. He has done a much better job then I could ever do in this reply.
You do not neet to touch any of the entries on the second page of most of my worksheets. They are pre-formatted with defaults. The only exception is the "TL Sections" worksheet which is 100% user input. The only time you need to input data on page two is when you are tweaking the last little bit of a design or trying to make your own custom enclosure.
Dave (Kneadle) wrote :
"When does shortening cause a T-Line to be something else, like a rear-ported BR?"
(MJK) I call my methods "quarter wavelength loudspeaker design" because I cannot draw a sharp line between the labels typically assigned to a speaker enclosure. If I give a definition of BR there are people that would argue on both sides of the definition and nobody would be 100% correct. The line between a TL, a ML TL, and a BR is not clear to me. All of the worksheets on my site include standing waves in the long direction of the enclosure and not the traditional lumped parameter modeling used in most T/S programs. This makes definitions confusing and in my opinion not too important.
Hope that helps,
Helps a lot. Every little bit is like the last few pieces of the puzzle you just can't place correctly.
Dave
Dave
Hi,
I've been around with Mr. King's software, TL open end, and my first conclusions are:
-there is a point where increasing the lenght of the line gives no much more extension (in my case a 130cm line gives a f5=45Hz)
-The higher the tapering ratio (5Sd to 1Sd) and the stronger the filling the smoother the ripple (+1-2dB)
About the ripple, is the king's model accurate until 1000Hz? Is a real problem this ripple?
Comparing this result with a BR box, I get deeper Hz with BR. Is there any other standard advantage using a TL, i.e. clearer midrange, impulse response, etc?
Now a question adressed to Mr. King: I'm trying to use the TL section, to give a shape to my project. I change the lenght of the sections, but no change in the drawings, why? What should I do?
Thanks
I've been around with Mr. King's software, TL open end, and my first conclusions are:
-there is a point where increasing the lenght of the line gives no much more extension (in my case a 130cm line gives a f5=45Hz)
-The higher the tapering ratio (5Sd to 1Sd) and the stronger the filling the smoother the ripple (+1-2dB)
About the ripple, is the king's model accurate until 1000Hz? Is a real problem this ripple?
Comparing this result with a BR box, I get deeper Hz with BR. Is there any other standard advantage using a TL, i.e. clearer midrange, impulse response, etc?
Now a question adressed to Mr. King: I'm trying to use the TL section, to give a shape to my project. I change the lenght of the sections, but no change in the drawings, why? What should I do?
Thanks
The first thing I notice about your post is that you used the TL Open End. Try your numbers with the closed end worksheet so you can compare and contrast.
As far as I understand it, a TL is supposed to give you a smoother rolloff at the low end, not that F3 "cliff" where the bass response simply disappears.
As far as I understand it, a TL is supposed to give you a smoother rolloff at the low end, not that F3 "cliff" where the bass response simply disappears.
Hi Raka,
The worksheets are probably accurate to around 500 Hz (rule of thumb) where standing waves across the line start to occur. I use FEM analysis as a last step in all of my designs to try and determine where to stop believing the "simple" MathCad models. The usual upper frequency of believable results is typically 500 Hz. To estimate at what point you should stop worying about the ripples calculate the frequency of half sine waves using the larger internal dimensions of your enclosure.
As for the "TL Sections" worksheet, the results and graphs should change when you change some of the dimensions in the input area. Make a change and then scroll down to see if the little light bulb starts to flash indicating the computer is calculating. If it does not flash then look under the math menu at the top and recalculate the entire worksheet.
Please let me know if you a still having problems,
The worksheets are probably accurate to around 500 Hz (rule of thumb) where standing waves across the line start to occur. I use FEM analysis as a last step in all of my designs to try and determine where to stop believing the "simple" MathCad models. The usual upper frequency of believable results is typically 500 Hz. To estimate at what point you should stop worying about the ripples calculate the frequency of half sine waves using the larger internal dimensions of your enclosure.
As for the "TL Sections" worksheet, the results and graphs should change when you change some of the dimensions in the input area. Make a change and then scroll down to see if the little light bulb starts to flash indicating the computer is calculating. If it does not flash then look under the math menu at the top and recalculate the entire worksheet.
Please let me know if you a still having problems,
Hi,
This is the most interesting solution I've found: A ML TQWT!
Could you please give your comments? Any suggestions?
Seems a good one, doesn't it?
This is the most interesting solution I've found: A ML TQWT!
Could you please give your comments? Any suggestions?
Seems a good one, doesn't it?
Hi Raka,
Looks very good. I am not sure I could offer much to improve the design. Build it and let us know how it turns out.
Looks very good. I am not sure I could offer much to improve the design. Build it and let us know how it turns out.
- Status
- Not open for further replies.