I am comparing WinISD simulation for port tuning with UniBox.
If I use "theoretical" response the results are the same but if I use FRD file for Dayton DC300 in UniBox, the difference is huge.
I have made screenshots for port tunig in 138 l cabinet (4.87 cubic ft) with port diameter 7.50 cm. There is obvious that different
Fb should be used for the best results (the lowest F3). So which tuning should I use or consider to be better?
The "real" one with frd file seems more realistic but it´s a surprise that 12" woofer will get F3 only at about 33 Hz.
I do not have possibility to measure the speaker response in cabinet.
Theoretical response - port length 9,95 cm
FRD simulation - port length 4,85 cm
If I use "theoretical" response the results are the same but if I use FRD file for Dayton DC300 in UniBox, the difference is huge.
I have made screenshots for port tunig in 138 l cabinet (4.87 cubic ft) with port diameter 7.50 cm. There is obvious that different
Fb should be used for the best results (the lowest F3). So which tuning should I use or consider to be better?
The "real" one with frd file seems more realistic but it´s a surprise that 12" woofer will get F3 only at about 33 Hz.
I do not have possibility to measure the speaker response in cabinet.
Theoretical response - port length 9,95 cm
FRD simulation - port length 4,85 cm
I'm perhaps missing something here, especially as I'm not familiar with Unibox, but you've got a port approx half the length in the second sim - that is why the F3 is substantially higher.
Could you explain why you chose that shorter port?
If you want the best comparison (ie with/without the FRD) then keeping everything else the same, including port length would be the better methodology.
Could you explain why you chose that shorter port?
If you want the best comparison (ie with/without the FRD) then keeping everything else the same, including port length would be the better methodology.
I used UniBox just because of possibility to use real frequency response of the driver.
I did not choose shorter port, just increased Fb (which gives shorter port) because if I use Fb 24 Hz (like in theoretical simulation), F3 is much higher.
I have made a comparison with the same Fb so you can see the difference and much higher F3.
I did not choose shorter port, just increased Fb (which gives shorter port) because if I use Fb 24 Hz (like in theoretical simulation), F3 is much higher.
I have made a comparison with the same Fb so you can see the difference and much higher F3.
OK, I see where you're coming from now, thanks.
I think that if you're using the standard frd files from Dayton, they will already be including some rolloff because even an infinite baffle for testing will impose it's own natural LF slope.
Really, you'd need to undo that before using the frd for your box modelling - not sure what the best method for doing that is however, sorry.
I know VituixCad has a variety of calculators built in so that might be able to do it, at a guess.
Good luck,
David.
I think that if you're using the standard frd files from Dayton, they will already be including some rolloff because even an infinite baffle for testing will impose it's own natural LF slope.
Really, you'd need to undo that before using the frd for your box modelling - not sure what the best method for doing that is however, sorry.
I know VituixCad has a variety of calculators built in so that might be able to do it, at a guess.
Good luck,
David.
The point is that this is useless.
Bass response under 100-200Hz on a bass reflex box depends entirely from the tuning (*), and port tuning behaviour depends on the T/S parameters of the driver. The only problem in not measuring is in having faith in the published T/S parameters.
Another point: searching for the lowest F3 is almost always not the best tuning.
Ralf
(*) In reality it depends heavily on the room and placement, but this is another story
Thank you for your answer, it helped me a lot to understand some things I was not sure.
I have found measured T/S parameters. I know each driver can be a little bit different but should I use theese ones?
https://www.dibirama.altervista.org...n-audio-dc300-8-woofer-12-8-ohm-160-wmax.html
Please, could you help me with port tuning? What would you suggest? I was thinking flat response with as low F3 as possible would
be the best but probably not. 🙂 In crossover simulation I have included baflle step so frequency up to 400 Hz are a bit "louder" than the rest.
In another words woofer is louder than mid and tweeter. So what would be the best port tuning for 138 l cabinet?
I have found measured T/S parameters. I know each driver can be a little bit different but should I use theese ones?
https://www.dibirama.altervista.org...n-audio-dc300-8-woofer-12-8-ohm-160-wmax.html
Please, could you help me with port tuning? What would you suggest? I was thinking flat response with as low F3 as possible would
be the best but probably not. 🙂 In crossover simulation I have included baflle step so frequency up to 400 Hz are a bit "louder" than the rest.
In another words woofer is louder than mid and tweeter. So what would be the best port tuning for 138 l cabinet?
If you tune for the flattest possible response to the lowest frequency, the side effect is that you then get a rather abrupt fall-off in bass response below that point, meaning the Q of the system is somewhat high. In this case, you end up with poor transient response, and that results in "boomy" bass. (Notice how the "corner" of the frequency response gets sharper, and the bass falloff more abrupt, as you tune for flattest bass response above f3 - this is showing you the system Q is increasing.)
At the other extreme, if you mount the woofer on a big flat baffle instead of a box, you get good transient response, and a rather gradual fall off in bass response, which produces very "tight" sounding bass - but you also get very little bass, because open baffle mounting produces very poor bass frequency response. IMO the compromise is absolutely not worth it, and open baffles are a terrible way to design a speaker system, most particularly a woofer system. But some people love that tight transient response enough to overlook the severe problems elsewhere.
Other woofer alignments have in-between properties. All sealed-box systems have bass response like a second-order high pass filter, i.e. bass falls off at 12 dB/octave below the bass f3. If the sealed box is big enough, and Q at the bass f3 is low enough, this can give you good transient response and clean bass - but you will also get a lot less deep bass than with a properly ported enclosure.
(If the sealed box is too small, or driver Qts too high, you get a bass peak, and you get bass boom anyway, even with only 12 dB/octave bass rolloff below bass f3. Sealed boxes do not guarantee good bass - you still have to have the right driver in the right-size box.)
All ported woofer boxes act like a fourth-order high pass filter, i.e. bass falls off at 24 dB/octave well below bass f3. This steeper fall-off tends to produce worse transient response and more "boom", compared to a sealed-box woofer system.
This is a handicap all ported woofer systems suffer from. If no care is taken, you end up with lots of boomy one-note bass, and there have been a lot of bad ported speaker systems like this, especially many decades ago, before we had Thiele-Small's research paper, or home PCs and woofer box simulation software.
The problem can be managed by choosing port tuning to allow bass to fall away relatively gradually, i.e. don't tune for maximum flatness, but rather for a smooth, gradual fall-off in bass. This is the same thing as tuning for a lower system Q near bass f3.
All this explains where the old classic Bessel, Butterworth, and Chebyshev tuning responses came from. Of the three, Bessel alignment produced the least bass, but the best transient response. Butterworth was maximally flat, and produced more bass, but worse "boom", compared to Bessel. Chebyshev produced the most bass of all, with the downside of very boomy bass and a peaky bass response. These responses were calculated by hand, by mathematicians working with pencils and paper, long before there were any computers.
These days, with computer simulation software, you don't have to pick between Bessel and Butterworth, but choose a response anywhere in between. Somewhere in between might be a good compromise between deep bass extension, and poor transient response.
If the simulation software you're using models transient response, that will be a good guide.
I'm unfamiliar with the software tools you're using, so unfortunately I cannot help at all with your software-specific questions.
-Gnobuddy
Thank you very very much for comprehensive answer. I will try to tune the port in WinISD but your explanation helped me very much.
Once again thanks a lot. 👍
Once again thanks a lot. 👍
Is there a specific reason for a 138L cab? The T/S parameters of the driver published by the mfg suggest a smaller cab.
The driver can work also in a closed box, but bass is somewhat limited. A 30L closed box will yeld a F6 at around 43Hz. A vented box works better, and a 90L box tuned to 25Hz gives a better low extension (F6 at 24Hz), power handling (*) and room integration than the maximally flat "standard" design that every simulator suggests.
Ralf
(*) The driver has a relatively poor x-max. This will limit the power handling to a mere 30-ish W, before exceeding x-max. Badly chosen alignment can further reduce the power handling.
Just for clarity's sake if the OP goes back into Unibox, importing the driver's FR into the program the way you did is incorrect. The "Import ext. active filter" button is intended to show you how something like a Low Pass filter from an AVR crossing to a sub will affect the selected box response. Very useful in fact to see the excursion demands placed on a driver when you use a LP filter on it and to therefore can also see how loud it can play in its desired application.
Filters of many kinds are easy to create in Response Modeler and then import into Unibox if anyone should feel so inclined.
Filters of many kinds are easy to create in Response Modeler and then import into Unibox if anyone should feel so inclined.
Hi guys. First of all I want to thank all of you for your help, I appreciate it very much. All of you gave me very important information
and I am glad and very grateful that you are helping me and teaching me.
shadowplay62 - thank you for advice. I used VituixCAD for crossover modeling but did not know it has this feature. I will study and try.
To be honest I want lower bass response from such big woofer so I hope I will get it. 🙂 But this is great advice. Thank you.
giralfino - thanks for your suggestion. 138 l is existing cabinet so I would like to use it. WinISD suggests box volume around 90 l so I hoped in bigger cabinet driver could go lower. I will compare your suggestion about 90 l with my thoughts and will try to find a way. You are right, xmax is very poor and I was surprised with cone excursion graph which shows that 25 W could be the maximum power... And higher Fb makes it even worse. How can then Dayton declare it as an 80 W speaker???
jReave - thank you for explanation and Response modeler, I will try it.
and I am glad and very grateful that you are helping me and teaching me.
shadowplay62 - thank you for advice. I used VituixCAD for crossover modeling but did not know it has this feature. I will study and try.
To be honest I want lower bass response from such big woofer so I hope I will get it. 🙂 But this is great advice. Thank you.
giralfino - thanks for your suggestion. 138 l is existing cabinet so I would like to use it. WinISD suggests box volume around 90 l so I hoped in bigger cabinet driver could go lower. I will compare your suggestion about 90 l with my thoughts and will try to find a way. You are right, xmax is very poor and I was surprised with cone excursion graph which shows that 25 W could be the maximum power... And higher Fb makes it even worse. How can then Dayton declare it as an 80 W speaker???
jReave - thank you for explanation and Response modeler, I will try it.
To re-use the cabinet you already have, you can always reduce its internal volume if you want - build an internal partition to close off part of the box, or even simpler, just put some large solid object(s) inside! This can be a smaller box, or pieces of solid wood, or even chunks of Styrofoam (expanded polystrene foam, also called EPS).
In your case, if you want to reduce the internal volume from 138 litres to 90, you'd need to put 48 litres worth of objects inside. Solid wood would be too heavy and too expensive in this case, but blocks of EPS are lightweight and usually available from home building centers (used for wall insulation).
If you look around with a creative eye I'm sure you can find other suitable objects that would do the job, too. Even plain ordinary bricks would work, but are extremely heavy.
-Gnobuddy