I'm using tubes but the output transformer secondaries are sectioned and can be connected to suit the driver. Though , as this progresses, I feel my resistance to using a plate amp getting less . . . . . . . but I need to keep the learning curve manageable so would like to continue looking at the basic system before added compensation and power assist.
"I would narrow down the woofer selection not for their sensitivity rating but their Fs, pick highest you think you are okay with . . . . "
Ok , this is the kind of place where my lack of knowledge is limiting. I don't really know how forcibly the Fs sets the lower limit. I know it's not a brick wall, but I have yet to understand how wide the margins are.
"I would narrow down the woofer selection not for their sensitivity rating but their Fs, pick highest you think you are okay with . . . . "
Ok , this is the kind of place where my lack of knowledge is limiting. I don't really know how forcibly the Fs sets the lower limit. I know it's not a brick wall, but I have yet to understand how wide the margins are.
^ I edited that sentence out but too late 😀 Rest of the sentence you left out is important. In general it seems the higher the Fs, the smaller the system with given sensitivity (number). Since you have the sensitivity requirement = low power, then all there is left is to balance between the system size and low extension. It is always balancing out the three, one has to go to achieve the other two: high sensitivity, low extension, small size.
Since there are drivers with ~95dB rated sensitivity in database it seems logical number to look at it first to get first narrow down of drivers. Then the Fs and size kind of balance out. You won't find too low Fs amongst the ~95db sensitive drivers, and the lowest Fs ones are bigger than the higher Fs ones. You could use multiple drivers, with lower sensitivity rating, in which case you'd see lower Fs drivers but the system size would stays about the same.
You should look simulate the candidates, because for example the Beyma 15k200 mentioned earlier is kind of oddball with very low Fs and high sensitivity at 100Hz, and seems to have nice mellow slope below 100Hz and could balance out nicely with help of room. Unfortunately this is old driver, but I guess there would be similar modern drivers as well. Check out BMS 18n862. The bigger you can go the better.
I like to think there is not too much difference, because the room can make +10dB peaks / dips to the response depending where you and the speaker are at (check out the multisub).
Since there are drivers with ~95dB rated sensitivity in database it seems logical number to look at it first to get first narrow down of drivers. Then the Fs and size kind of balance out. You won't find too low Fs amongst the ~95db sensitive drivers, and the lowest Fs ones are bigger than the higher Fs ones. You could use multiple drivers, with lower sensitivity rating, in which case you'd see lower Fs drivers but the system size would stays about the same.
You should look simulate the candidates, because for example the Beyma 15k200 mentioned earlier is kind of oddball with very low Fs and high sensitivity at 100Hz, and seems to have nice mellow slope below 100Hz and could balance out nicely with help of room. Unfortunately this is old driver, but I guess there would be similar modern drivers as well. Check out BMS 18n862. The bigger you can go the better.
I like to think there is not too much difference, because the room can make +10dB peaks / dips to the response depending where you and the speaker are at (check out the multisub).
Last edited:
Got it, thanks.
An earlier comment you made was about the 6dB to 12dB rise from sealed to vented. When using Small's chart is it safe to assume 6dB for starters?
An earlier comment you made was about the 6dB to 12dB rise from sealed to vented. When using Small's chart is it safe to assume 6dB for starters?
Depends on the driver and frequency, and the box size can change too between alignments. But yeah if it was for porting alone about 6dB would be safe to assume I think, but that is only at the tuning frequency, less above and below making a steeper slope. Since this is for sub woofer use with low power, I would make reflex yeah, anything to get the extension and sensitivity required. You can test this and all in simulator for various drivers to get some hunch on the parameters and how much cheap amplifier Power changes things. I think the Small charts are very handy to see how the displacement requirement varies with the extension and SPL though.
You could ask at the subwoofer forum for more accurate answers what is the difference between various box alignments. Low power is kind of special requirement and there might be more suitable alignment than the standard 4th order reflex? Horn would get sensitivity up like no other, but the size goes up like no other as well. More info on bandpass and the more exotic variety available there as well.
You could ask at the subwoofer forum for more accurate answers what is the difference between various box alignments. Low power is kind of special requirement and there might be more suitable alignment than the standard 4th order reflex? Horn would get sensitivity up like no other, but the size goes up like no other as well. More info on bandpass and the more exotic variety available there as well.
Last edited:
OK.
Would you say something about how you use the relationship of tuning frequency to Fs? My understanding is that there's an output peak at Fs after which roll-off is sharp. Is it correct to think in terms (generally) of placing the tuning frequency just below the Fs ?
Also, a question of terminology. The posts I have read have some referring to box tuning and some referring to vent or port tuning. Are they identical?
Would you say something about how you use the relationship of tuning frequency to Fs? My understanding is that there's an output peak at Fs after which roll-off is sharp. Is it correct to think in terms (generally) of placing the tuning frequency just below the Fs ?
Also, a question of terminology. The posts I have read have some referring to box tuning and some referring to vent or port tuning. Are they identical?
Yeah the driver output has natural rolloff, below the Fs. I've observed from the simulations that if the port tuning is higher than Fs then there appears a peak to the response above. Also if the tuning is too low it is no much of use (dip on the response above) but it depends on the driver Qtc for example, and box size. With some particular Q drivers there is opportunity to get various useful shapes on the response, and I think Extended Bass Shelf is one that could perhaps be useful for you, reasoning from the name 😀 I'm not too pro on this, but it is all playing with the holy three: size, sensitivity and extension, shifting them around. One benefit tuning on drivers Fs is that the Helmholtz resonance at the port replaces the driver resonance, which makes less excursion = more output capability than closed box before the voice coil burns, very useful for high power applications. Tune the port even higher and use high pass filter = even more output capability, but no low extension. Again, one was traded off for the other. I guess you could tune below Fs with your low power application, trade off some output capability to inch more extension, but don't expect miracles. I would consult the subwoofer forum for detailed rundown of the various options.
On the terminology: I'm not sure what is the correct one. Port tuning depends on the box size and on the other hand box would be tuned with the driver. What?😀 In general, driver and the box make one system. Port tunes the box... I guess main goal is to tune the box/driver as whole and port tuning is part of the process. Well,.. I'm not used to talk about bass boxes so easily use wrong terminology, sorry about that.
ps. then comes the room and laughs on the alignment names and the carefully tweaked box and implies the nasty dips and peaks on top of it not caring tiniest bit on the woofer TS parameters etc. 😀 So we should be actually concentrate tuning the room instead.. have fun!
On the terminology: I'm not sure what is the correct one. Port tuning depends on the box size and on the other hand box would be tuned with the driver. What?😀 In general, driver and the box make one system. Port tunes the box... I guess main goal is to tune the box/driver as whole and port tuning is part of the process. Well,.. I'm not used to talk about bass boxes so easily use wrong terminology, sorry about that.
ps. then comes the room and laughs on the alignment names and the carefully tweaked box and implies the nasty dips and peaks on top of it not caring tiniest bit on the woofer TS parameters etc. 😀 So we should be actually concentrate tuning the room instead.. have fun!
Last edited:
No need to be sorry about anything at all. Absolutely ! The ability to back and forth is very, very helpful. Books don't offer the opportunity to question what the author assumed everybody knew, and I usually get stuck on those points. This is exactly the kind of conversation that helps. I really do appreciate it!
Time to unplug and re-read what has been said this far. Back in a day or two. Thanks again.
Time to unplug and re-read what has been said this far. Back in a day or two. Thanks again.
Glad if it helped! Writing helps me too, thinking it through while typing, hence large possibility for errors and equally large possibility to learn!
OK, hardly started and back with question already.:
"Eyeballing from the charts behind the link you'd need displacement of ~200cm3 for ~105dB at ~40Hz. Quadruple the displacement to get what is require for ~20Hz. This would be roughly 800cm3, so a ~15" driver with SD of 800cm2 excursing +/-0.5cm would do it? In which case a single 15" driver can do the ~105dB at ~40Hz with quarter of the excursion of that, so +/-0.125cm."
I need to check my understanding of the above statement. Is the displacement linear with cone travel? ie. where Sd = 800cm^2, will 1cm of cone travel produce displacement of 800cm^3 ?
"Eyeballing from the charts behind the link you'd need displacement of ~200cm3 for ~105dB at ~40Hz. Quadruple the displacement to get what is require for ~20Hz. This would be roughly 800cm3, so a ~15" driver with SD of 800cm2 excursing +/-0.5cm would do it? In which case a single 15" driver can do the ~105dB at ~40Hz with quarter of the excursion of that, so +/-0.125cm."
I need to check my understanding of the above statement. Is the displacement linear with cone travel? ie. where Sd = 800cm^2, will 1cm of cone travel produce displacement of 800cm^3 ?
Reading .
"Since there are drivers with ~95dB rated sensitivity in database it seems logical number to look at it first to get first narrow down of drivers. Then the Fs and size kind of balance out. You won't find too low Fs amongst the ~95db sensitive drivers, and the lowest Fs ones are bigger than the higher Fs ones. You could use multiple drivers, with lower sensitivity rating, in which case you'd see lower Fs drivers but the system size would stays about the same."
This is where it starts to get less simplistic (or more interesting) and clearly has something to do with other T/S parameters. You mentioned the 18N862 , (which I assume is because you like it,) so I started looking at some 18" drivers. Thinking to look at some of the highly regarded names that I've heard of , I found a sheet for the old TAD TL-1801, which has an Fs of 26Hz yet simmed needs a huge box to go as low as some 15"s with a higher Fs.
Wanting to make connections it's easy to just draw a line straight to Vas. (The TAD's was 500 Litres!) Not sure yet if the Q's have a big part to play in frequency response or whether their influence is mostly on other parts of the picture, like distortion for one example.
"Since there are drivers with ~95dB rated sensitivity in database it seems logical number to look at it first to get first narrow down of drivers. Then the Fs and size kind of balance out. You won't find too low Fs amongst the ~95db sensitive drivers, and the lowest Fs ones are bigger than the higher Fs ones. You could use multiple drivers, with lower sensitivity rating, in which case you'd see lower Fs drivers but the system size would stays about the same."
This is where it starts to get less simplistic (or more interesting) and clearly has something to do with other T/S parameters. You mentioned the 18N862 , (which I assume is because you like it,) so I started looking at some 18" drivers. Thinking to look at some of the highly regarded names that I've heard of , I found a sheet for the old TAD TL-1801, which has an Fs of 26Hz yet simmed needs a huge box to go as low as some 15"s with a higher Fs.
Wanting to make connections it's easy to just draw a line straight to Vas. (The TAD's was 500 Litres!) Not sure yet if the Q's have a big part to play in frequency response or whether their influence is mostly on other parts of the picture, like distortion for one example.
Yeah the 15" would have more excursion than 18" for same volume displacement, and in general higher excursion means higher distortion. Although some drivers have less distortion on more excursion than some others with less excursion, better suited for purpose, also more expensive. It is always a compromise between multiple things, some property gets traded off for some other. The 15" would probably require more power to reach same SPL than the 18" big box, on some low frequency where the displacement makes the rules of the game. Some modern drivers can take few thousand Watts, huge SPL capability, but that is very rarely needed at home application. Always have to keep the application in mind to find most suitable solution for (any) problem.
I'm not too versed on T/S parameters either, but at quick look the Vas is air volume that has equivalent compliance to the driver suspension Cms. Bigger Vas = looser suspension = lower free air resonant frequency, but the moving mass is also at play etc. If Vas is small, you need more moving mass, which takes away the sensitivity for example. See, the size / sensitivity / low extension link is everywhere. All the small signal T/S parameters are more or less interlinked and describe the driver capabilities from multiple perspectives size, extension and sensitivity being the most obvious ones, perhaps there are more perspectives but not sure if they are important for subwoofers other than suitability for given amplifier. Checkout for example this resource, https://mynewmicrophone.com/full-list-thiele-small-speaker-parameters-w-descriptions/ or some other, lots of easy to understand information on the various T/S parameters. I guess there are dozen threads on the forum as well, expect this kind of discussion popup at least yearly.
You can go very deep on the stuff and while it is very interesting I never went too deep, not yet at least 🙂 Because, the very long wavelengths of the sub frequencies are kind of special because their size. Basically, any audio problem with subwoofers boil down to the low extension vs. SPL combat and both can be increased just by adding more woofers, and bigger. The long wavelengths allow multiple boxes combine constructively because they easily fit close enough each other, closer than 1/4 wavelength. In fact 100Hz barely fits in typical livingroom being ~3.4m long and two subs closer than ~80cm from each other would combine constructively 100Hz and down. An easy fit, even the biggest 24" woofers could make it easily side by side. And thats it. Kind of boring, but audio related problems of sub frequencies are solved by increasing cost and complexity, by increasing cone area, the volume displacement.
See the multisub stuff, how multiple sources positioned and and aligned can average out the room modes statistically and yield smooth bass response on many listening spots / big sweet spot / or through out the whole room, while there is almost no other way to achieve that. If one assumes multiple subwoofers, say four, the size of each box can be smaller because now the cone area is 4x compared to single subwoofer box. Or if the size is no problem then you'll get more SPL and low extension while smoothing out the response. There really is not many downsides other than cost and complexity, mo woofers and multiple amplifiers with some signal processing, cabling. But, all of it could fit inside furniture for example.
So yeah, basically look for big stuff, big cones with big boxes to get low frequencies with low Watts (high sensitivity). The bigger you can go, the mo bass you get with the Watts you have. There you go, all TS parameters simplified to one sentence, in sub woofers all problems and solutions relate to the size 😀
In reality, size and money are limited so in the end it is just about finding balance between the extension and SPL requirements. With low power application sensitivity needs to be high (set in stone) so the low extension will be limited by the size. Trying to say that not limiting yourself to low Watt amplifier without EQ possibility will impose limit to the system low frequency extension for manageable sized system. Unlimited power and EQ would cure problems in home use, this would allow your sensitivity requirement go down and allow you to use almost any driver you get your hands on, small or big.
Forgot to mention, checkout the small bluetooth speakers. Ever wonder how they manage to have any bass with so small form factor? They cheat by adding distortion, trick hearing system to hear the fundamental which is not there!
I'm not too versed on T/S parameters either, but at quick look the Vas is air volume that has equivalent compliance to the driver suspension Cms. Bigger Vas = looser suspension = lower free air resonant frequency, but the moving mass is also at play etc. If Vas is small, you need more moving mass, which takes away the sensitivity for example. See, the size / sensitivity / low extension link is everywhere. All the small signal T/S parameters are more or less interlinked and describe the driver capabilities from multiple perspectives size, extension and sensitivity being the most obvious ones, perhaps there are more perspectives but not sure if they are important for subwoofers other than suitability for given amplifier. Checkout for example this resource, https://mynewmicrophone.com/full-list-thiele-small-speaker-parameters-w-descriptions/ or some other, lots of easy to understand information on the various T/S parameters. I guess there are dozen threads on the forum as well, expect this kind of discussion popup at least yearly.
You can go very deep on the stuff and while it is very interesting I never went too deep, not yet at least 🙂 Because, the very long wavelengths of the sub frequencies are kind of special because their size. Basically, any audio problem with subwoofers boil down to the low extension vs. SPL combat and both can be increased just by adding more woofers, and bigger. The long wavelengths allow multiple boxes combine constructively because they easily fit close enough each other, closer than 1/4 wavelength. In fact 100Hz barely fits in typical livingroom being ~3.4m long and two subs closer than ~80cm from each other would combine constructively 100Hz and down. An easy fit, even the biggest 24" woofers could make it easily side by side. And thats it. Kind of boring, but audio related problems of sub frequencies are solved by increasing cost and complexity, by increasing cone area, the volume displacement.
See the multisub stuff, how multiple sources positioned and and aligned can average out the room modes statistically and yield smooth bass response on many listening spots / big sweet spot / or through out the whole room, while there is almost no other way to achieve that. If one assumes multiple subwoofers, say four, the size of each box can be smaller because now the cone area is 4x compared to single subwoofer box. Or if the size is no problem then you'll get more SPL and low extension while smoothing out the response. There really is not many downsides other than cost and complexity, mo woofers and multiple amplifiers with some signal processing, cabling. But, all of it could fit inside furniture for example.
So yeah, basically look for big stuff, big cones with big boxes to get low frequencies with low Watts (high sensitivity). The bigger you can go, the mo bass you get with the Watts you have. There you go, all TS parameters simplified to one sentence, in sub woofers all problems and solutions relate to the size 😀
In reality, size and money are limited so in the end it is just about finding balance between the extension and SPL requirements. With low power application sensitivity needs to be high (set in stone) so the low extension will be limited by the size. Trying to say that not limiting yourself to low Watt amplifier without EQ possibility will impose limit to the system low frequency extension for manageable sized system. Unlimited power and EQ would cure problems in home use, this would allow your sensitivity requirement go down and allow you to use almost any driver you get your hands on, small or big.
Forgot to mention, checkout the small bluetooth speakers. Ever wonder how they manage to have any bass with so small form factor? They cheat by adding distortion, trick hearing system to hear the fundamental which is not there!
Last edited:
I'm not altogether sure this is the best use of forum bandwidth but you were kind enough to ask for updates and so here is where I've got to in my search so far.
Very nice that VituixCad has the ability to click on the tab for parameter you're interested in and so be able to prioritize that in a scrolling search .
The BMS was the first one I found and I thought I'd found a good pick, but I really take issue with the way they laid out the three graphs on the right side of the data sheet. Placing the vented box FR between the other two that start at 10Hz so you'll make the assumption that it does too. So BMS is out.
The B&C drivers all look pretty good to me, what do you think?
I have a question about ribbed cones vs smooth ones. My experience with water tells me that these shapes matter in the way they get more grip on the material they're trying to move, in this case air. So I have to ask why an engineer might go with one over the other.
Very nice that VituixCad has the ability to click on the tab for parameter you're interested in and so be able to prioritize that in a scrolling search .
The BMS was the first one I found and I thought I'd found a good pick, but I really take issue with the way they laid out the three graphs on the right side of the data sheet. Placing the vented box FR between the other two that start at 10Hz so you'll make the assumption that it does too. So BMS is out.
The B&C drivers all look pretty good to me, what do you think?
I have a question about ribbed cones vs smooth ones. My experience with water tells me that these shapes matter in the way they get more grip on the material they're trying to move, in this case air. So I have to ask why an engineer might go with one over the other.
Spoken like the proverbial Zen master, a mountain of experience summed up in a few words. It was enough to help me do a "smooth cone vs ribbed, polar response" search though.
The first page I read didn't really talk polars so much as tone.
The second one makes only brief mention of polar response but has a lot of other interesting info.
I guess it will take more than a few minutes online to get a few decades worth of education. : )
Thanks
The first page I read didn't really talk polars so much as tone.
The second one makes only brief mention of polar response but has a lot of other interesting info.
I guess it will take more than a few minutes online to get a few decades worth of education. : )
Thanks
Hearinspace, it's good to see the use of coloured fonts. I'd just remind you that now the forum has different themes 🙂
AllenB, I hear what you're sayin' but I don't know what it means. Can you give an example of theme use?
(I just had the thought that maybe it's not clear the coloured words are links. Maybe they should have been underlined or something.)
(I just had the thought that maybe it's not clear the coloured words are links. Maybe they should have been underlined or something.)
I just wanted you to know so you didn't make any the same as the background so we couldn't see them 😉