Hello everyone, I'm in doubt about a question: how to simulate response of a loudspeaker using two (or more) different drivers sharing the same chamber with an unique vent. Simulator I know, in fact, permits only to simulate same drivers in compound (Winisd) or different drivers but in different defined chamber (Boxsim).
So using this last, it would be enough for me to know how the total mount of cabinet's liters divides spontaneously between two drivers based on their different T/S parameters, can you help me estimating this? This wouls work for both closed and vented systems?
Thank you very much.
So using this last, it would be enough for me to know how the total mount of cabinet's liters divides spontaneously between two drivers based on their different T/S parameters, can you help me estimating this? This wouls work for both closed and vented systems?
Thank you very much.
Thank you, very useful article, but I wish to find a different method, able to give to each driver to be simulated with its own filter circuitry, I refer in particular to 2 1/2 way systems, in which often woofer and midwoofer share an unique inner space but are wired through different filter stage.
For this I wish to calculate how much of space each drivers uses of the total amount and divide these values in a generic simulator.
What do you mean exactly for weakness and strenght of a driver? By what T/S parameters is this indexed?
from Thorsten's method, the Vas of the two woofers are added, so the space 'used' will be proportional to the Vas's. I think that simulating the box required is a different problem to dealing with the crossover, they're not really linked....
(FWIW, I've used Thorsten's method, & it works very well with drivers that are reasonably close in parameters)
(FWIW, I've used Thorsten's method, & it works very well with drivers that are reasonably close in parameters)
I don't think it works for vented boxes, all the early Japanese speakers I have seen/repaired/used/dismantled over the years that had dissimilar drivers sharing a box volume have been sealed. But I could be wrong there.
One of the better sounding 70's Japanese boxes had a 12" + a 10" in a shared heavily stuffed box with very simple closed back cone tweeters with a single cap in line on each 3.3uF and 1uF respectively. I remember this because I couldn't believe how good something this simple was. No brand name that I can remember but they were white cone on the 10" and a ridged black cone on the 12" May even have been home made using generic Japanese drivers, definitely Japanese characters on the backs of the magnets tho with a 8R stamp
Not Corals, I would have remembered if the drivers were from Coral but reversed surrounds on both
One of the better sounding 70's Japanese boxes had a 12" + a 10" in a shared heavily stuffed box with very simple closed back cone tweeters with a single cap in line on each 3.3uF and 1uF respectively. I remember this because I couldn't believe how good something this simple was. No brand name that I can remember but they were white cone on the 10" and a ridged black cone on the 12" May even have been home made using generic Japanese drivers, definitely Japanese characters on the backs of the magnets tho with a 8R stamp
Not Corals, I would have remembered if the drivers were from Coral but reversed surrounds on both
2 1/2 way speakers usually use 2 of the same model of woofer/midwoofer, so having them share the same space would have no effect on their acoustic response. That means you can model them simply as 2 woofers in parallel in any modelling software. The difference is only in the low-pass filtering, which can be dealt with individually in crossover simulators.
A couple of examples: Troels' CA18RNX and CNO-25 and Zaph's WaveguideTMM.
Hope this helps,
David.
Well, however let's trying to understand better. I think that other than Vas, the cabinet space used by two different woofers sharing it depends on Qts and Sd too.
About Qts, given as a value indicating a driver's ability or strenght to compress the loading air and given that, if I don't go wrong, a low Qts indicates a low ability (in fact a low Qts driver's motor tends to brake itself if loaded with the yielding air mass of a big cabinet, so low Qts needs small cabinets) and a high Qts a high ability (for the opposite reason respect the previous chance), when different Qts woofers work together, on equal terms of Vas and Sd, the high Qts one will take more space and the low Qts less.
About Sd, it's simpler imagine that a, on equal terms of Vas and Qts, a bigger Sd woofer will take more space and a smaller Sd woofer less.
What do you think about?
About Qts, given as a value indicating a driver's ability or strenght to compress the loading air and given that, if I don't go wrong, a low Qts indicates a low ability (in fact a low Qts driver's motor tends to brake itself if loaded with the yielding air mass of a big cabinet, so low Qts needs small cabinets) and a high Qts a high ability (for the opposite reason respect the previous chance), when different Qts woofers work together, on equal terms of Vas and Sd, the high Qts one will take more space and the low Qts less.
About Sd, it's simpler imagine that a, on equal terms of Vas and Qts, a bigger Sd woofer will take more space and a smaller Sd woofer less.
What do you think about?
I've read and found it very useful; I'm just looking for an alternative method based on leaving drivers as they are and divide loading volume instead, so I can use classical simulator like Boxsim: I fear it can't be possible.
However, the thing I don't understand of 2 1/2 way in a single chamber is the following: one woofer (1) is cut very low to compensate baffle step and another (2) higher to cross with midrange or tweeter; in the mid-low band where the woofer 1 doesn't emit, why woofer 2 doesn't lose SPL through woofer 1? Maybe does it work like a passive radiator tuned so low for 2 that it simply remains steady in that frequency range?
Thank you very much.
You can use boxsim & unibox etc, with the combined parameters, after all, that's the whole point of combining them...
Re:"why woofer 2 doesn't lose SPL through woofer 1?" - why would it? assuming they're in phase they're working together i.e. pushing in the same direction, no matter what the crossover is doing. I suspect you're overthinking this...
Re:"why woofer 2 doesn't lose SPL through woofer 1?" - why would it? assuming they're in phase they're working together i.e. pushing in the same direction, no matter what the crossover is doing. I suspect you're overthinking this...
Can you be clearer about what, exactly, you are talking about and trying to do?
When you say 'different woofers' do you simply mean two drivers of the same type / model, sharing the same enclosure volume? Or do you mean two drivers of different types, sharing the same enclosure volume?
Some basics:
-Sealed and vented enclosure volume is dominated by effective Qts' (Qts + any series R in the circuit), Fs and Vas. As far as I am aware, there have been no changes to the laws of physics and the popular electrical analogies as applied to moving coil drive units.
-Other than some possible interactions with the impedance peak of a large series inductor, a crossover filter per se is neither here nor there as far as box volume and tuning is concerned.
-If you have two drivers of the same type, sharing the same enclosure volume, then in essence they each simply 'see' 1/2 the total volume available. There will be some minor differences due to production tolerances, but for practical purposes that's what happens.
-If you have two different types of drivers sharing the same enclosure volume (it doesn't matter if it's a sealed or vented box from the point of view of modelling) then your simplest way of establishing suitable cabinet volume and tuning frequency is to use the compound method Thorsten suggests in the link I provided you. This does not involve any change to the drivers whatsoever. It is simply a convenient way of allowing you to model a combined enclosure volume and tuning frequency for your drivers in your choice of software. If you want to make life harder for yourself than it really needs to be, then you can model the drivers individually in a shared volume with something like AkaBak, ABEC3 or equivalent combined lumped element with BEM package, or finite element software with acoustic module, e.g. COMSOL.
When you say 'different woofers' do you simply mean two drivers of the same type / model, sharing the same enclosure volume? Or do you mean two drivers of different types, sharing the same enclosure volume?
Some basics:
-Sealed and vented enclosure volume is dominated by effective Qts' (Qts + any series R in the circuit), Fs and Vas. As far as I am aware, there have been no changes to the laws of physics and the popular electrical analogies as applied to moving coil drive units.
-Other than some possible interactions with the impedance peak of a large series inductor, a crossover filter per se is neither here nor there as far as box volume and tuning is concerned.
-If you have two drivers of the same type, sharing the same enclosure volume, then in essence they each simply 'see' 1/2 the total volume available. There will be some minor differences due to production tolerances, but for practical purposes that's what happens.
-If you have two different types of drivers sharing the same enclosure volume (it doesn't matter if it's a sealed or vented box from the point of view of modelling) then your simplest way of establishing suitable cabinet volume and tuning frequency is to use the compound method Thorsten suggests in the link I provided you. This does not involve any change to the drivers whatsoever. It is simply a convenient way of allowing you to model a combined enclosure volume and tuning frequency for your drivers in your choice of software. If you want to make life harder for yourself than it really needs to be, then you can model the drivers individually in a shared volume with something like AkaBak, ABEC3 or equivalent combined lumped element with BEM package, or finite element software with acoustic module, e.g. COMSOL.
Last edited:
Of course, I was intending two different kind of woofers, with different T/S parameters. I understood well that article and put it among my bookmarks, but I'd want a method to leave in my simulation different woofers as they are because I'll need to drive them with different crossover filters, as normally is made in 2 1/2 way systems: if I use an unique compound driver how can I simulate this?
However, my aim is to improve the bass quality output of my brand towers, 32 liters 2 1/2 way using two 6'' woofers and 1'' tweeter in an unique vented chamber accorded at 40 Hz; woofers are driven with two different filters, one of which is cut very low, compensating baffle step.
Bass is not so engaging and punchy as I would like and I'd try to substitute the low woofer. I took a T/S measurement of both drivers, they are slightly similar, the low has a Fs of 65 Hz, the "high" 70 Hz, both have a high Qts of 0.6 and Vas around 11 liters; the only commercial driver I found similar to them is Faital Pro 6FE100.
Response is quite extended but not often as I would like.
In your opinion, enlarging cabinet would improve bass quality? In terms of extension or in punchiness? It's very difficult find a driver able to make better in that space.
I was very surprised to discover a so high Fs drivers in my towers, never known similar projects, given their quite low F3.
Thank you
Simulating two different drivers in a common volume is NOT something that typical software can do. The problem becomes more complex if the drivers are connected to different crossovers. First consider two identical drivers in a common volume driven by the same signal. In this case both drivers response identically to the drive signal, move in phase, and the result is that each driver behaves as if it were a single driver in a box of 1/2 the volume. Each driver sees the loading (compliance) of the shared box as if it were an isolated box of 1/2 the volume. Now, consider that the two drivers are different; different Qts, Fs, Vas...,etc. As the frequency of the drive signal approaches the resonant frequency of the driver with the higher Fs it will start to move differently that the other driver. It will have a phase shift relative to the lower Fs driver. As a result, one driver may be moving in while the other driver moves out. This, in effect, makes the box volume look frequency dependent and effects the motion of both drivers.
Getting more technical, if you examine the equation governing the motion of the drivers, for a single driver in a box there is a term that looks like
....... X * (1 + Vas/Vbox)
This term is the compliance term which represents the force of the driver due to suspension compliance and the compliance of the air in the box. When two driver are placed in the same box there are two such equations and this term becomes
...... X1 * (1 + Vas1/Vbox) + X2 * Vas2/Vbox for driver 1 and
...... X2 * (1 + Vas2/Vbox) + X1 * Vas1/Vbox for driver 2.
where X1, and X2 are the varying positions of the drivers as they move in and out.
You can see that if the drivers are identical, Vas1 = Vas2, and X1 = X2 and these terms become
...... X * (1 + 2 Vas1/Vbox) from which we conclude that for identical drivers in a common volume, Vbox must double if the drivers are to move the same as in isolated single boxes.
Now, if you introduce different filters for each driver the result becomes more complex because the signal to each driver will also be different as well.
Hope this helps understand the complexity of placing two different drivers in a shared volume with different filters.
Getting more technical, if you examine the equation governing the motion of the drivers, for a single driver in a box there is a term that looks like
....... X * (1 + Vas/Vbox)
This term is the compliance term which represents the force of the driver due to suspension compliance and the compliance of the air in the box. When two driver are placed in the same box there are two such equations and this term becomes
...... X1 * (1 + Vas1/Vbox) + X2 * Vas2/Vbox for driver 1 and
...... X2 * (1 + Vas2/Vbox) + X1 * Vas1/Vbox for driver 2.
where X1, and X2 are the varying positions of the drivers as they move in and out.
You can see that if the drivers are identical, Vas1 = Vas2, and X1 = X2 and these terms become
...... X * (1 + 2 Vas1/Vbox) from which we conclude that for identical drivers in a common volume, Vbox must double if the drivers are to move the same as in isolated single boxes.
Now, if you introduce different filters for each driver the result becomes more complex because the signal to each driver will also be different as well.
Hope this helps understand the complexity of placing two different drivers in a shared volume with different filters.
Many thanks, really glad to read your exhaustive explanation.
What do you think about sonority of high Qts / high Fs / Vas woofer systems?
What gives a punchier bass output:
- high Qts / high Fs / low Vas in large cabinet (mine)?
- low Qts / low Fs / high Vas in small cabinet?
- ......??
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.