Specific question: In an active system, if I pad down the spl from the roll off point up, say by 3db, then effectively, the f3 of the padded system will be equal to f6 of the system before padding, correct? For example: In an active 3 way system, I have a woofer having sensitivity of 100db @1w/1m with f3, f6, f10 tuned to 35, 30, 28 . If I sacrify a bit of sensitivity and pad it down to 97db above the roll off point, then the new F3, f6, f10 would be 30,28, 25.
Would love to hear some thoughts whether this is correct and the pros and cons of this approach, especially if you have tried this before.
Thanks.
Would love to hear some thoughts whether this is correct and the pros and cons of this approach, especially if you have tried this before.
Thanks.
Yes, correct. This is common practice in loudspeakers designed for home audio. The drawback offcoarse is lower sensitivity and lower power handling which is a concern in public adress system but to a much lesser extend in home audio.
There are a couple of ways you can achieve this. With passive crossovers it was common pratice to tune a bassreflex a bit lower to create a shelf below somewhere in between 60 to 100Hz (depending on the compromises you are willing to make) that is about 3dB down and than use a bigger inductor in series with the woofer, either with or without a resistor in parallel, to bring everything above that plateau to the same level
With an active crossover you can use the same method or you can use a shelving lowpass filter to create a boost towards the lower end, effectively with the same end result. You can do this in your existing design but better pratice is to take in into account in the design. You can start by simply making the reflex port a bit longer and start from there. That way you prevent over excursion by boosting to much below the basreflex tuning.
There are a couple of ways you can achieve this. With passive crossovers it was common pratice to tune a bassreflex a bit lower to create a shelf below somewhere in between 60 to 100Hz (depending on the compromises you are willing to make) that is about 3dB down and than use a bigger inductor in series with the woofer, either with or without a resistor in parallel, to bring everything above that plateau to the same level
With an active crossover you can use the same method or you can use a shelving lowpass filter to create a boost towards the lower end, effectively with the same end result. You can do this in your existing design but better pratice is to take in into account in the design. You can start by simply making the reflex port a bit longer and start from there. That way you prevent over excursion by boosting to much below the basreflex tuning.
@Sjef: Thanks a lot for sharing. It's pretty intuitive to me but I hardly ever seen this topic being discussed to I thought there may be some thing I missed. For more context, for my 3 way active, I'm debating between a 12" hifi woofer (22hz fs, 230L vas & 91db sensitivity) with a 15" PA woofer (35hz fs, 110L vas & 96db sensitivity). If I pad the PA woofer down 4-5db then in my simulation, the PA woofer actually have lower f3, but roll off more aggressively (higher f10). With room gain, they probably sound very similar and in that case, I'd prefer the PA just because it... looks better lol
There's no 'replacement for displacement', so bigger is better (BIB) rules down low! 😉 As Sjef noted, designing in the roll-off slope for better matching the room gain (extended bass shelf alignment/EBS) is the way (acoustic solutions to acoustic problems).
It is being discussed, only in different words. Some call it extended bass shelf alignment, some call it baffle baffle step (in essential the same process of correcting), others call it linkwitz transform (in essential almost the same process) or they simply call it tuning and build the matching crossovers/correction filters without knowing they are in fact extending the low freq roll-off.
Many PA manufacurers have an EBS alignment in their recommended enclosures for their midwoofers. It very common for drivers with very high BL factors, otherwise you won't get any bass out of them at all.
Personally I would go for the 15", depending on the drivers offcoarse. Yes bigger is often better.
@Sjef thanks. I did some reading on EBS and its exactly what I asked, I do see the shelf when playing with my Woofer Box tool as well.
Now that the 12" woofer (SB Acoustics SB34NRXL) is out of the equation, I have two candidates for the 15", the B&C 15TBX100-8 and B&C 15SW115-4. I have pretty much settle on the later since all the specs indicates better fit for low frequency (higher xmax, higher mms, slightly lower fs...) but what puzzles me is that B&C recommends 90L, 40Hz tuning for it while 115L, 35Hz for the TBX??? If I follow this recommendation then the 15TBX100 will fit like a glove to my design. Thoughts?
Heres more info on my project:
Here's the info on the drivers:
Now that the 12" woofer (SB Acoustics SB34NRXL) is out of the equation, I have two candidates for the 15", the B&C 15TBX100-8 and B&C 15SW115-4. I have pretty much settle on the later since all the specs indicates better fit for low frequency (higher xmax, higher mms, slightly lower fs...) but what puzzles me is that B&C recommends 90L, 40Hz tuning for it while 115L, 35Hz for the TBX??? If I follow this recommendation then the 15TBX100 will fit like a glove to my design. Thoughts?
Heres more info on my project:
- 3 way active, 100-110L internal vol, 6.5" mid + 1.2" tweet, crossover ~300 and ~2000
- Available power: 200w 8ohm, 350 4ohm - Im not sure if this is enough for the beasty 15SW115
- Room is smallish but opens to kitchen and dining room
Here's the info on the drivers:
My first thought is that you don't have to be slaved to manufacturer's recommendations, which are generated by someone (sometimes it's been ME, borrowed by the marketing department) with who knows what criteria in mind and what philosophy they have.
- 2nd, I'd want the more capable woofer.
- 3rd I like ports tuned low, to really cover the bottom end, and because tuning higher tends to worsen transient response (boominess).
- Now as for "padding the SPL" I do not understand how you are proposing to do this. If you're going to put passive components to knock down the highs that's a horrible idea for a multitude of reasons, killing efficiency being one. You should get a different driver that suits requirements in the first place.
- Presuming you're meaning to do it actively, if you try to EQ down the highs it's rather the same as boosting the lows which is a more direct method.
- All this is crap because unless you're crazy lucky the response in the ROOM will not be flat, and you really should have something like Audyssey/Dirac. Without that you're fooling yourself. (You could also fiddle a lot with measurement and parametric EQ as an alternative).
- A corollary to that is that F3 is not really important. Dick Small himself told me so when in a conversation in Indiana I asked about that and said that in a room with room gains, F6 or F10 seemed more important. He agreed and said F3 was just a mathematical convenience, and it's also kind of a standard electrical filter number which is what he and Thiele and Benson were modeling loudspeakers upon. https://en.wikipedia.org/wiki/Richard_H._Small
Oh if you're talking about subwoofers then all this is doubly crap, because the F3 simulated is relative to an ideally calculated midband figure that quite possibly doesn't apply.
- The semi-inductance in subwoofers generally chokes off response before you even really get to the midrange, though I have to caveat I was working in autosound and that woofers like big PA-type may differ.
- So if you're not using something like LEAP which uses real impedance curves of the woofer, your simulations are just a vague approximation of reality.
- IIRC the low-frequency sensitivity was really determined by the suspension+box stiffness and the moving mass. If I did simulations with more or less BL (magnet strength) the SPL curves all traced the same line at the lowest frequencies. The ones with stronger magnets just kept rising longer to a higher maximum.
There you said precisely what Diablogt was asking. Because of a stronger magnet, you'll create a rising response towards the midrange. Compared to the midrange your -3dB will shift up higher in frequency. Not only on paper but also for your ears. Diablogt called correction for this rising response 'padding down' which is maybe a bit confusing term to you but in essence what he is doing is just that, create a flatter response and by this lowering F3 on paper and to your ears.
Offcourse, this says nothing about in-room response but it does say a lot about the capability of producing low frequencies at the lowest possible distortion. Dirac does not correct for distortion. It is easy to get an in-room response down to 30Hz from a single 6.5inch woofer with Dirac but it will sound like crap, distortion all over.
👆
@head_unit you answered your own question. Yes I can boost bass with dsp but in my limited experience playing with equalization, you have much better result padding the spl curve to flat than raising it to flat. Of course you lose sensitivity, but thats why the highly sensitive pro woofers are being considered.
The more capable woofer is an easy choice if it didnt cost exactly 2 times as much than the cheaper one. Cost is not important until it is of course lol. Also what put me off is the really high mms and rms while having the same Bl. This + its much higher xmax indicate to me that the 15SW115 is built for more max spl than the 15tbx100, which is important in PA but useless for me since Ill never have enough room or power to play at its limit.
@head_unit you answered your own question. Yes I can boost bass with dsp but in my limited experience playing with equalization, you have much better result padding the spl curve to flat than raising it to flat. Of course you lose sensitivity, but thats why the highly sensitive pro woofers are being considered.
The more capable woofer is an easy choice if it didnt cost exactly 2 times as much than the cheaper one. Cost is not important until it is of course lol. Also what put me off is the really high mms and rms while having the same Bl. This + its much higher xmax indicate to me that the 15SW115 is built for more max spl than the 15tbx100, which is important in PA but useless for me since Ill never have enough room or power to play at its limit.
Xmax for a 15" woofer for home use only is an issue if you use it (baffle-less) dipole. In a closed box or reflex box you really don't need 7-8mm Xmax, 3-4 mm is already more than enough. I'm running double 15" woofers with only 2mm Xmax (older JBL's, not being made anymore) and haven't been able to get them running out of steam. I have chosen my woofers on relatively low mms of around 100grams and low rms of around 2. I have tried heavier coned woofers with higher rms but they need to be driven to much higher spl to get them going and therefore don't sound as good on lower volume levels. Drivers that are optimized for higher power handling often have higher rms because they use voicecoil formers primarely made for better heat dissipation. For home use you could get away with a paper voicecoil former but these are not made anymore these days.
Take a look at the Faital Pro 15PR400 for instance. Only 85 grams mms and a rms of a very reasonable 3.1. Qts is not to low, Xmax with 5.75mm plenty for home use and it has pretty good linearity without much cone breakup.
The B&C woofers you pointed out are real PA subwoofers. I'm using drivers like that for theaters with up to 800 seats. You will never reach their Xmax with normal home use unless you want to have rave parties with lots of friends.
Take a look at the Faital Pro 15PR400 for instance. Only 85 grams mms and a rms of a very reasonable 3.1. Qts is not to low, Xmax with 5.75mm plenty for home use and it has pretty good linearity without much cone breakup.
The B&C woofers you pointed out are real PA subwoofers. I'm using drivers like that for theaters with up to 800 seats. You will never reach their Xmax with normal home use unless you want to have rave parties with lots of friends.
Last edited:
Thats pretty much my thought. Im concerned my amp are too weak for them let a lone breaking lol. The Faital 15pr400 is probably the most popular option due to its low mass and Le. But its VAS is 230L so my cabinet seems too small for it. I modeled it and struggle to turn it to 30hz f3 in my box. It would be a great option for a full range guitar cab but I will only use it under 300hz so its capability to play more in the mid range is wasted. Now its brothers, the 15fh500 and 15fh510 seem very tempting.
Yes, these look nice, really low rms for a 15" driver. Anything around 2 is very good for details and dynamics at low spl level.
Faital drivers sound very good, at least the ones I have heard.
Faital drivers sound very good, at least the ones I have heard.
It isn't like this. Sensitivity is not affected by EQ. Going down or coming up is essentially the same thing.
Efficiency is only affected by passive crossovers. With actieve crossovers effiency as in how much amplifier power is transfered into acoustics energy, stays the same nomatter how much eq you apply. That's one of the main benefits of actieve systems
You still have the same capability of the driver and the same risk of excess excursion. No matter how you do it, the driver can only product so much at any given frequency. Use your CAD, ( I use WINISL) and map the excursion/frequency/power.
Padding the top is no different than boosting the bottom. Of course a boost in DSP requires an overall reduction so you do not exceed the dynamic range.
Padding the top is no different than boosting the bottom. Of course a boost in DSP requires an overall reduction so you do not exceed the dynamic range.
Sjef, not necessarily. Consider the case where the boosted bass represents the largest signal. The amp can only go to the same output overall. 3dB of the passband is wasted by way of missing headroom (not unlike before). It is also lost, only it happens at line level.
I have tested this in my system. Most of the energy in music is in the area of 80 to 120Hz, and second the area from 120Hz to around 250Hz. Way down low there is much less energy, especially below 60Hz. When you take a look at the spectrum of a kickdrum or a bass guitar for instance you will notice that most of the energy is centered around the first harmonic. That's why all these tiny loudspeakers with 5,5 inch drivers are often voiced with a bump in frequency response between 80 and 120Hz to make them sound bigger than they look. (but with false bass and lot's of distortion) When I made a correction by shelving down the area above +/- 60Hz compared it as passive system with passive speaker level eq (as in a big inductor with parallel resistor) and doing the same in dsp line level I get about 5 to 6dB more headroom from my amplifier with the active system. That is with normal music playback, not with pink noise or other test signals. So unless you listen to pipe organ or subwoofer tests all day there is a benefit.
Ok, but let's not lose sight of the standard argument. That comes from when a midrange is too sensitive you can either burn power in a passive crossover or turn the gain down in an active system. People read a lot into this, however an active system also burns relative power but at a different point in the system. Mostly you have a rearranging of the gain structure of the system. There are always benefits to optimising but it can work in different configurations.
In this thread, one of the limitations comes when you run out of driver excursion.
In this thread, one of the limitations comes when you run out of driver excursion.
Rearranging the gain system has no influence on sensitivity, you don't lose any amplifier power by lowering the signal on its input. Nothing is wasted, in fact, you gain headroom for an amp that drives the too sensitive midrange.
Yes, the excursion limit is a limiting factor in how much you want to compensate but with 15" woofers in a normal home situation it's most likely not an issue. 4mm displacement from a 15" woofers is the same volume displacement as 25mm diplacement of a typical 6,5 inch woofer. In reverse this means that 4mm displacement of a 6,5 inch woofer translates into a only 0,64mm displacement for a 15" woofer. So all in all, Xmax is not much of an issue with big woofers at home. Distortion figures are quite a lot lower as well. For a typical 6,5 woofer distorsion figures of 25% and higher is normal when trying to dive down in subbass area. Distortion rises almost parabolicy from 300Hz on down with these small drivers
Yes, the excursion limit is a limiting factor in how much you want to compensate but with 15" woofers in a normal home situation it's most likely not an issue. 4mm displacement from a 15" woofers is the same volume displacement as 25mm diplacement of a typical 6,5 inch woofer. In reverse this means that 4mm displacement of a 6,5 inch woofer translates into a only 0,64mm displacement for a 15" woofer. So all in all, Xmax is not much of an issue with big woofers at home. Distortion figures are quite a lot lower as well. For a typical 6,5 woofer distorsion figures of 25% and higher is normal when trying to dive down in subbass area. Distortion rises almost parabolicy from 300Hz on down with these small drivers
Is this said like it's a bad thing? Are you familiar with the efficiency of the average Class AB amp, or the reference efficiency of a speaker?