They're different things, really. Linkwitz transforms essentially change one second-order response to another second-order response, whereas a transconductance amplifier will affect the frequency response because the high output impedance will cause higher gain at higher speaker impedances, and conversely lower gain at lower impedances. The curves will tend to be different.
For example, take a sealed box woofer system with a resonance of 40 Hz at a system Q of 0.707. That means the output will be -3dB at 40 Hz, decreasing at 12 dB/octave below there. Now a specific LT designed for the task can shift the -3dB point from 40 Hz to 28 Hz, half an octave down, so long as one accepts a maximum boost of about 6 dB or so.
The transconductance amplifier will show a peak at the system resonance of 40 Hz, depending on how high the impedance rises above baseline. Say the impedance goes from 6 ohms to 24 ohms: power is defined by current squared times impedance, so increasing the impedance by a factor of four means the system will have +6 dB gain at resonance, changing the 3 dB drop to a 3 dB peak at not necessarily the most convenient place.
Transconductance amplification would shine with low Q systems, where the droop at resonance will be counteracted by the power rise at resonance. For example, making the above example's box bigger so the system Q is now 0.5 means the resonance gain is -6dB, which would be counteracted by the aforementioned power increase at resonance.
Note also that transconductance amplifiers are quite suboptimal for vented systems since the higher impedance peak occurs well within the passband so you get a massive boom right where you don't want it.
Now, if you're very very careful and also include a high pass filter so the cone doesn't go flying across the room, you can use a LT with a vented system. This is NOT a simple proposition, however! The possibility of error cannot be overstated, so unless you've years of experience with vented systems and filters I cannot recommend doing this.
Thanks for the response. I have to learn some more about the system Q, and I am using a transconductance amplifier with a vented subwoofer system!
Hi.Because of Waf i need to build stereo subwoofers as small as possible(will be sealed tower subwoofers.One of them SB34NRX75-6 and other is B&W spare part 12 inch).
As i think , when box gets smaller, there is less amount of air in the box .( m) mass of air is smaller (Q=m.c.deltaT)and this air will heat up faster than the air which was in bigger box and this hotter air forces the cone and makes the cone move less than a bigger box and because of less movement there will be decrease at lower frequencies right?
And we need more power for same spl for example at 20Hz in smaller box to move the cone as much as bigger box.
I have a subwoofer which have QTS: 0,685 VAS: 217L and FS: 20,5 Hz
At win ISD it needs 2700 liters of box.
Do you think i can get same sound with 50 liters of box and Linkwitz transform circuit?
Power is not a problem as i have powerful 500w class D amp.
What will be distortion figures compared to big box?
As i think , when box gets smaller, there is less amount of air in the box .( m) mass of air is smaller (Q=m.c.deltaT)and this air will heat up faster than the air which was in bigger box and this hotter air forces the cone and makes the cone move less than a bigger box and because of less movement there will be decrease at lower frequencies right?
And we need more power for same spl for example at 20Hz in smaller box to move the cone as much as bigger box.
I have a subwoofer which have QTS: 0,685 VAS: 217L and FS: 20,5 Hz
At win ISD it needs 2700 liters of box.
Do you think i can get same sound with 50 liters of box and Linkwitz transform circuit?
Power is not a problem as i have powerful 500w class D amp.
What will be distortion figures compared to big box?
Don´t worry about heat in a home einvironmet with WAF and neigbours present, as long as u are listening to "normal Music" or uses extremly small drivers... Of course, if you are into "dubstepp all night long until the glasses fall out of the shelf", it´s a different story 🙂
closes boxes with linkwitz transforms usually are much smaller than ported enclosures with the same driver.
Your limitations are xmax of the driver most of the times. Most "bass drivers" from two way speakers are built to work in ported enclosures and to deliver mids as well, so they don`t have too much excursion capability, compared to "real subwoofer drivers" or car-hifi subs..
Since it´s so easy to build a closed box - why not simply try it out... To check if the SPL is enough for you, you can simply take any EQ and boost a little where you miss bass... If that seems fine - you can go ahead and build a nice looking box and solder some Linkwitz circuits together 🙂
After all, noone can guess if "it´s enough" for you, since this is totally subjective.
closes boxes with linkwitz transforms usually are much smaller than ported enclosures with the same driver.
Your limitations are xmax of the driver most of the times. Most "bass drivers" from two way speakers are built to work in ported enclosures and to deliver mids as well, so they don`t have too much excursion capability, compared to "real subwoofer drivers" or car-hifi subs..
Since it´s so easy to build a closed box - why not simply try it out... To check if the SPL is enough for you, you can simply take any EQ and boost a little where you miss bass... If that seems fine - you can go ahead and build a nice looking box and solder some Linkwitz circuits together 🙂
After all, noone can guess if "it´s enough" for you, since this is totally subjective.
This is why I mostly add at least a 1st order filter, but mostly a 2nd order filter.
If the total Q of your LT and highpass is around 0.5, you even get a very nice groupdelay and response.
Another way of doing this is adding a parametric EQ, or a combination of both
Let's take the questions in order:
- Heat has little effect on cone movement. Compliance, however, does enter into it. A smaller closed box will have a smaller volume of air behind the speaker, which will be stiffer than a large volume of air and therefore pushes the resonant frequency of the system higher. It's the same thing as a spring: make it longer and the force needed to deflect it by a certain distance will decrease by the same ratio as the ratio of the spring's new length to the old length. Now put a weight on that and you'll see the oscillations will happen at a higher frequency, precisely the same as happens for a speaker in various sized boxes. More power being needed to get the same output from a smaller box is correct. Now compliance is called that because a more compliant system needs less force to compress it by a given volume, which is inversely proportional to stiffness. The resonant frequency goes up as the square root of the increased stiffness of the system and Q increases by the same amount, so if you have a speaker with VAS of 100 litres (let's call that stiffness 0.01) and Q = 0.6 in a 33.3 litre (stiffness 0.03), then the resonant frequency will increase by a factor of sqrt((0.01 + 0.03)/0.01), or 2.0, and Q will increase to 1.2.
You could get the same response with any size box; it's just a matter of how much amplifier power is used before blowing up the speaker. Note that WinISD tunes closed boxes to a Q of 0.707, but your specified speaker has a Q of 0.685. Since Q is a function of stiffness as we see above, it only takes a tiny bit of added stiffness to increase the system Q to 0.707 and that's why the recommended box is so huge. Setting the box size to 50L brings the resonant frequency to 48.4 Hz with a Q of 1.44, which results in a 4 dB boost around 55 Hz and a -3 dB point around 33 Hz. Doing the stiffness numbers as above shows the WinISD calculations check out. The generated Linkwitz transform (LT) will indeed give you -3 dB rolloff at 20 Hz, just as you specified.
However, that might not be the right thing for your place. Rooms tend to have more gain as you go further down in frequency, with small rigid rooms increasing the lowest frequencies by 10 dB or more. WinISD designs LTs at a Qp of 0.71, which will produce thick-sounding bass in an actual living space. For my own taste, I tend to favour a lower Q system, at a Q of about 0.4 with a resonant frequency around 20ish Hz. The resultant curve looks awful, starting to roll off around 80 Hz before being 8 dB down at 20 Hz, but since rooms increase in gain at just a bit less than 6 dB / octave below 80 or 100 Hz (note the incredible precision of these numbers! 😀 ) the resultant response at my ears is close to flat. A look at the transfer function magnitude for the EQ/filter shows the LT is far more speaker-friendly at lower Q as well. If you're feeling experimental try a Q of 0.3 or so, although that might be a bit dry for most tastes.
- Distortion is typically a factor of cone displacement, so producing the same level at a given frequency is more or less independent of how a system is tuned. However, for a same setting of the volume control the lower Q system will have less distortion since the system response is rolled off gently to allow for room gain, with the added advantage of asking the speaker to handle less power. You can see my bias by now 🙂, but it works for me.
whats the view on using a LT on a driver that you know you can get a FS of 55-60hz at a proper enclosure size (lets say 0.16cuft), but you want to halve the box, but lift the lower end back up to what you would get normally with the larger?
you guys think it would sound ok at lower, but distort at normal?
I'm thinking of attempting something with a 3.5in FR driver, but want to make something small that sounds bigger.
you guys think it would sound ok at lower, but distort at normal?
I'm thinking of attempting something with a 3.5in FR driver, but want to make something small that sounds bigger.
Hi.Thank you for your response.I have never been a normal listener 🙂 2pcs 12 inch subwoofers for 20m2 room may not called extremely small drivers IMHO(not extremely small for me and also for neighbours 🙂.I can listen to loud music because we also listen their fighting teen twins everyday so i have lots of credits 🙂 I do not know you,maybe you are listening 4x 18 inch subs at home lol )
My subwoofer towers will be 22x33x105cm dimensions. Makes about 53 liters inside volume and 66liters effective volume with stuffing.
I should build 2 different Linkwitz circuit because drivers are different(one B&W spare and other SB34NRX75-6.I hope different subwoofers may not be a problem because in diyaudio i read people told they are using different types of subwoofers in same room and they are happy with them and see no problem.
I think we should crowd-fund for a Behringer DCX2496 DSP, cheap on eBay. That way any curve you can think of (and more) could be explored.
There's no magic to LT except for anechoic chambers, if there. The disturbances to any theoretical solution are absolutely monumental, as fine-tuning always proves. So nearly zero benefit being picky about LT or any other rationally chosen bass boost.
So just get a DSP and dial-in whatever you need for fine tuning.
Special Added Bonus: the Behringer has elaborate limiting and compression routines which can be set in seconds according to a whole bunch of parameters of your choosing. Therefore, you can protect your sub driver from harm - which can easily be caused by LT.
B.
There's no magic to LT except for anechoic chambers, if there. The disturbances to any theoretical solution are absolutely monumental, as fine-tuning always proves. So nearly zero benefit being picky about LT or any other rationally chosen bass boost.
So just get a DSP and dial-in whatever you need for fine tuning.
Special Added Bonus: the Behringer has elaborate limiting and compression routines which can be set in seconds according to a whole bunch of parameters of your choosing. Therefore, you can protect your sub driver from harm - which can easily be caused by LT.
B.
Last edited:
Hi.I read this text and he says air becomes stiffer because of heat is caused by cone movement.
Sub Box Polyester Fiberfill
And when i make Qp 0,5 k becomes negative and R2 is negative
Enter the following:
f(0) = 43,5 Hz
Q(0) = 1,2900
f(p) = 20,0 Hz
Q(p) = 0,5000
f(hpf) = 10,000000 Hz
k = -0,19 (k>0 required)
Is it too bad to make Qp 0,707 at 20-25m2 room?
Thank you for your response too.
Hi! Its old thrrad,but maybe i can find someone who can help me. First i want to build a bass horn,but after many months calculating horns,w bins,ported enclosure,i realize that the best option for me is to make sealed enclosure cca 200l and use linkwitz transform to get nice low response.
What is bodering me is my driver. I have Eminence Kappa pro 18lf-8.
I try to sell but nothing for now. Its big an powerful speaker,very sensitive 98dB.
But,it said in spec that resonant frequency is 32Hz,and usable frequency 38-700 Hz.
What that means?
That i can not go low as 25 or 20Hz with that driver,or is posible to hit so low using linkwitz transform?
Any advice is wellcom.
I need it for woofer for my HIFI,so i dond need extra big SPL,just for home use.
What is bodering me is my driver. I have Eminence Kappa pro 18lf-8.
I try to sell but nothing for now. Its big an powerful speaker,very sensitive 98dB.
But,it said in spec that resonant frequency is 32Hz,and usable frequency 38-700 Hz.
What that means?
That i can not go low as 25 or 20Hz with that driver,or is posible to hit so low using linkwitz transform?
Any advice is wellcom.
I need it for woofer for my HIFI,so i dond need extra big SPL,just for home use.
Last edited:
If your driver can handle it, you can EQ any way you want. LT is a pretty radical EQ.
Linkwitz did the math to plot the boost needed for open-back cabs.... in the abstract. But rooms, driver specs, where you sit in the room and where you locate the speaker will over-whelm the abstract math, as is often the case with that kind of abstraction.
B.
Linkwitz did the math to plot the boost needed for open-back cabs.... in the abstract. But rooms, driver specs, where you sit in the room and where you locate the speaker will over-whelm the abstract math, as is often the case with that kind of abstraction.
B.
So if i understand corectly,this driver can go low as 20 Hz with linkwitz transform and i can ignore information from spec. list that this driver is only usable from 38Hz?
You can input any freq to any driver at any power level. But - esp for the bass - lot more to know about how much sound comes out and how much driver is destroyed.
It is routine to push the woofer with EQ but you need to understand how. The steep slope of LT is liked by people with open-baffles, at least as a starting point. But other cabs need other EQs and now-a-days, DSP to tailor the output to your room, not brute-force LT.
You can Linkwitz Transform any sealed-box or open-baffle driver to 20 Hz. But you will quickly run into problems with (a) power handling, or (b) excursion limits. So you cannot ignore those parts of the specification list.
Technically you can also Linkwitz Transform a ported-system.
You can also make up a LT filter of a bunch of param EQ's.
One has to keep an eye on the max power handling and excursion limit with any system.
Depending on the context and application, but in general I advice having at least a 1st order HP filter for some protection.
Woodnes,
Use Basta for simulations. And you can get LT, max excursion etc...
https://www.tolvan.com/index.php?page=/basta/basta.php
Use Basta for simulations. And you can get LT, max excursion etc...
https://www.tolvan.com/index.php?page=/basta/basta.php
Yes. It's quite a bit more complicated, and 4th-order instead of 2nd-order.
Only approximately, because a PEQ doesn't change the critical frequency like a LT does.
although i have limited experiences in this area, i would advised against driving a speaker below its resonant frequency, in this case 32Hz (post #52 data)