I acquired 4 of these drivers a few years ago. I am planning to build 1 subwoofer for 2 channel fed by a commercial amp. Later down the road I plan on building a second and incorporate it into a 2 channel/ HT setup. Since I already have 4 drivers, I was thinking of 2 drivers in an isobaric set up in a sealed box to reduce the size of the enclosure. The only negatives I can think of are cost and a more complex build. I bought the drivers years ago (used) so cost is not a consideration and my woodworking skills are good enough that the box construction would not be difficult.
Can anybody think of other reasons Isobaric is not a good choice and which Isobaric configuration is favored?
Thanks
Clarke
Can anybody think of other reasons Isobaric is not a good choice and which Isobaric configuration is favored?
Thanks
Clarke
If you plan on running them loud, then isobaric + sealed box could be a bad idea, as there will be very little opportunity for the drivers to keep cool.
Often the outer volume savings from Isorarik are much lower than theoretical possible. Tipical only 1/3.
Depending on the woofers parameters, a passive high pass (a capacitor in series with the woofer) will often be a better solution, with only one chassis needed. This principle is not well known and even more often misunderstood.
Putting two woofer on opposing sides of a cabinet has huge sonic advantages, by the way, compensating the impulse of the moving cones. So a passive high pass with two chassis would sure be a nice alternative with compareable size. The impulse response does NOT suffer from such a construction and you can NOT achieve this active or with a DSP. It is no Linkwitz transformation.
Depending on the woofers parameters, a passive high pass (a capacitor in series with the woofer) will often be a better solution, with only one chassis needed. This principle is not well known and even more often misunderstood.
Putting two woofer on opposing sides of a cabinet has huge sonic advantages, by the way, compensating the impulse of the moving cones. So a passive high pass with two chassis would sure be a nice alternative with compareable size. The impulse response does NOT suffer from such a construction and you can NOT achieve this active or with a DSP. It is no Linkwitz transformation.
... and huge moving mass of two added 15" membranes shaking your enclosure. What i mean is: havin two drivers i would always go for force cancelling layout!
I have the RS390 HF too... one in 110 liter closed and one in 130 liter closed - in a 4 subwoofer setup - multisub - the two others are 12".
One of my friends have 4 of them like you, and put them in the corners - closed 90 liters - and lacks nothing. He pointed them downwards and made the boxes with legs, so that they can be used as a table for a flower and painted discreetly, so that you actually forget them a bit 👍
But please give them some power and EQ.... especially EQ... cause that will matter - big time!
I'm not saying isobaric is bad... it definitely works.... but 3-4 subwoofers instead of two, works better at leveling out the response in the room, before EQ is applied.
Usually a bigger driver in a smaller box, gives you a higher Q, making a bump in the response, before the room even joins the party. BUT - you can even that out with EQ, and the driver is pretty powerful and handle quite a lot - especially with 4 of them.
But again.... what is big, elegant, small - for you?
One of my friends have 4 of them like you, and put them in the corners - closed 90 liters - and lacks nothing. He pointed them downwards and made the boxes with legs, so that they can be used as a table for a flower and painted discreetly, so that you actually forget them a bit 👍
But please give them some power and EQ.... especially EQ... cause that will matter - big time!
I'm not saying isobaric is bad... it definitely works.... but 3-4 subwoofers instead of two, works better at leveling out the response in the room, before EQ is applied.
Usually a bigger driver in a smaller box, gives you a higher Q, making a bump in the response, before the room even joins the party. BUT - you can even that out with EQ, and the driver is pretty powerful and handle quite a lot - especially with 4 of them.
But again.... what is big, elegant, small - for you?
Last edited:
So force cancelling layout is one driver on front facing outwards and the second driver on back facing outwards. Wired in phase and either in series or parallel. Is the box size doubled, because of a second driver? Or halved? Parts Express recommends 1 driver in a 2.88 cubic ft sealed box.
A passive high pass is simply a capacitor in series with the woofer of a (usually) closed box with a theoretically too small volume for the drivers parameter.
This concept is somehow complicated to understand, as most people see this as a typical way to connect a tweeter in a loudspeaker.
The size is in most cases 500-1200 micro Farad and can be calculated. These can be special audio capacitors, but two average, cheap 63V electrolitic's connected in series with positive to positive, do just as well. A too smal enclosure results in a high Q, the capacitor corrects the resulting "boomy" bottom end and extends the low frequency respose. As an additional bonus the driver is protected from subsonic frequency down to DC, like from a failing amp,
The impulse of a single driver shakes the cabinet, in extreme cases, subwoofer move in the room. If you have a driver at both ends of the cabinet, these forces cancel themselve out, resulting in a better bass. This effect is quite audible, the more the larger the driver gets.
As you have 4 drivers, two such sub's, placed in the right room positions, should be very impressive.
This concept is somehow complicated to understand, as most people see this as a typical way to connect a tweeter in a loudspeaker.
The size is in most cases 500-1200 micro Farad and can be calculated. These can be special audio capacitors, but two average, cheap 63V electrolitic's connected in series with positive to positive, do just as well. A too smal enclosure results in a high Q, the capacitor corrects the resulting "boomy" bottom end and extends the low frequency respose. As an additional bonus the driver is protected from subsonic frequency down to DC, like from a failing amp,
The impulse of a single driver shakes the cabinet, in extreme cases, subwoofer move in the room. If you have a driver at both ends of the cabinet, these forces cancel themselve out, resulting in a better bass. This effect is quite audible, the more the larger the driver gets.
As you have 4 drivers, two such sub's, placed in the right room positions, should be very impressive.
Exactly. Sorry for my sloppy post ..
Either front/back or on the sides.
Up/downwards could result in cone sag, i would not recommend this.
I have a thread with a passively-assisted sealed alignment here, complete with an example...
https://www.diyaudio.com/community/threads/passively-assisted-sealed-alignments.360786/
I try to explain it in a simple way:
The construction works by enabeling the woofer to receive more current in some (usefull) regions and limiting it in those that are not wanted (the bump). Interestingly, distortion drops in the low frequency, even as the woofer makes higher excursion. This can be explained from the steep low pass the driver and capacitor forms below the usabel, mostly size dependend frequency. Also the smaller cabinet size gives the cone better damping, limiting it from wobling around. Theoretically you have to feed more power to the woofer, as it takes up more current. Practically this has been proven to be untrue. As the amplifier does not deliver frequency too low to be produced by the driver and also reduces the "too small enclosure" bump, it actually saves this (power supply) energy an can use it where it is needed. In fact you should be able to drive it even louder than in a large, closed box.
The only downside, I have found, the prediction you get from a simulation does in most cases not give the exact capacitor value. So you may start with 1200 micro Farad end end up with 800 for example. Driver tollerance and the actual behavior of the cabinet, speaker connection etc. may add up.
The easy way of getting it spot on is to place a microphone in front of the woofer and change the capacitor size while you measure.
Linearity of the microphones frequency response is absolutely unimportant, as long as it goes somewhat low enough. We only compare and do not need absolute measurements!
This should be done in about 10 minutes, if you have some capacitors handy. You can easily hear the result A to B if you shorten the caps. It should be quite similar to turning in 3 dB bass at your preamp or pushing the loudness button.
Capacitors:
You can use any cap that fits the voltage to be expected by the driver. Most people use bipolar, electrolitc, often in combiination with some (1-10% of capacitance) foil film caps, intended for "fast" response.
Because these are expensive in these "high" values, as a cheap alternative, you can take two identical electrolitic capacitors (like 63volt types) and connect them with their negative poles to form a bipolar cap. The resulting bipolar cap has 1/2 the capacity of one of the singles. It is most important to connect capacitor negative to capacitor negative, has to do with the electrolitcs inner construction. So two 2.000 micro Farad caps form a 1.000 micro bipolar.
If someone might think this priciple or constructed bipolar caps have sonic disadvantages, the German loudspeaker manufacturer Isophon used both in its large high end loudsoeakers. They even where so brazen to patent this priciple that is in use for about 50 years, mostly in car audio.
Why do we not see it more often? Commercially seen, customers pay more for a lot of air in a large box, while the smaller box seems worth less.
In hobby audio, many have idiologic views, sometimes as tollerant as the Taliban of Afganistan. See: A/B, valve and Class D amps for example.
A cap in the signal path is considered bad, smearing impulses. The fact that the often preferred vented cabinet has a far worse groub delay, is just as any fact today, from some considered fake news.
Last, as I pointed out, you need to measure. Many DIYS builders still are afraid of connecting a measuring microphone to a laptop and install REW or some other free software. Learning something new is not appreciated by anyone. Also, if you start to measure, you may have to admitt your own faults...
The construction works by enabeling the woofer to receive more current in some (usefull) regions and limiting it in those that are not wanted (the bump). Interestingly, distortion drops in the low frequency, even as the woofer makes higher excursion. This can be explained from the steep low pass the driver and capacitor forms below the usabel, mostly size dependend frequency. Also the smaller cabinet size gives the cone better damping, limiting it from wobling around. Theoretically you have to feed more power to the woofer, as it takes up more current. Practically this has been proven to be untrue. As the amplifier does not deliver frequency too low to be produced by the driver and also reduces the "too small enclosure" bump, it actually saves this (power supply) energy an can use it where it is needed. In fact you should be able to drive it even louder than in a large, closed box.
The only downside, I have found, the prediction you get from a simulation does in most cases not give the exact capacitor value. So you may start with 1200 micro Farad end end up with 800 for example. Driver tollerance and the actual behavior of the cabinet, speaker connection etc. may add up.
The easy way of getting it spot on is to place a microphone in front of the woofer and change the capacitor size while you measure.
Linearity of the microphones frequency response is absolutely unimportant, as long as it goes somewhat low enough. We only compare and do not need absolute measurements!
This should be done in about 10 minutes, if you have some capacitors handy. You can easily hear the result A to B if you shorten the caps. It should be quite similar to turning in 3 dB bass at your preamp or pushing the loudness button.
Capacitors:
You can use any cap that fits the voltage to be expected by the driver. Most people use bipolar, electrolitc, often in combiination with some (1-10% of capacitance) foil film caps, intended for "fast" response.
Because these are expensive in these "high" values, as a cheap alternative, you can take two identical electrolitic capacitors (like 63volt types) and connect them with their negative poles to form a bipolar cap. The resulting bipolar cap has 1/2 the capacity of one of the singles. It is most important to connect capacitor negative to capacitor negative, has to do with the electrolitcs inner construction. So two 2.000 micro Farad caps form a 1.000 micro bipolar.
If someone might think this priciple or constructed bipolar caps have sonic disadvantages, the German loudspeaker manufacturer Isophon used both in its large high end loudsoeakers. They even where so brazen to patent this priciple that is in use for about 50 years, mostly in car audio.
Why do we not see it more often? Commercially seen, customers pay more for a lot of air in a large box, while the smaller box seems worth less.
In hobby audio, many have idiologic views, sometimes as tollerant as the Taliban of Afganistan. See: A/B, valve and Class D amps for example.
A cap in the signal path is considered bad, smearing impulses. The fact that the often preferred vented cabinet has a far worse groub delay, is just as any fact today, from some considered fake news.
Last, as I pointed out, you need to measure. Many DIYS builders still are afraid of connecting a measuring microphone to a laptop and install REW or some other free software. Learning something new is not appreciated by anyone. Also, if you start to measure, you may have to admitt your own faults...
Last edited:
I figured out how to wire 2 single electrolytic capacitors into bipolar capacitor at 1/2 value of a single. I don't know what voltage my driver would expect or what size electric capacitors to choose to start the experimentation process of the determining the correct capacitors to stick with. And am I correct this high pass filter goes on the positive line out of the amp before the first driver. And both drivers wired in parallel and box volume the same as sealed for 1 driver for a 2 driver forced cancelling layout.
Thanks for all your patience,
2 caps in series also should result in double the voltage rating. I am not sure wether this is also the case with polarized electrolytics used as NP ...
You can calculate maximum peak voltage:
Peak voltage = sqrt(peak Power*impedance)
You can connect the caps (in series) whereever you like. The result will always be the same.
You can calculate maximum peak voltage:
Peak voltage = sqrt(peak Power*impedance)
You can connect the caps (in series) whereever you like. The result will always be the same.
Last edited:
There is a very good simmulation program called AJHorn, many call it the best there is, as its predictions are 98% spot on. In fact, if your build dose not measure as simulated, you will allways find a fault in your construction or input data. It is not free (quite inexpensive at 110 €/$ for what it does!), but you can download a Demo version. It looks more complicated than it is and you find good tutorials.
The demo version is limited, but you can simulte closed boxes perfectly.
The great advantage: It has an x-over simulator, that works just perfect. If you only put a capacitor in front of the driver, you will exactly see what it does to a certain driver in a closed box of different volume. With a few clicks you know if your driver/ cabinet combination makes sense and what frequency extention you get with the high pass. As I said, the simulated capacitor size may not be the optimum, but not because of the programs limitation, but the unpredictable properties of the actual driver, box and wireing. So if the drivers TSP are perfectly measured and the cabinets data is also entered 100% right, the capacitor will be spot on too. Remember that the usual cabinet, if not made from concrete, is somehow elastic and stores and takes up energy.
So if you simulate in AJHorn and then fine tune the capacitor by measureing, you will get what you have simulated. This makes speaker building much simpler, as you do not waste time and money with test cabinets.
The demo version is limited, but you can simulte closed boxes perfectly.
The great advantage: It has an x-over simulator, that works just perfect. If you only put a capacitor in front of the driver, you will exactly see what it does to a certain driver in a closed box of different volume. With a few clicks you know if your driver/ cabinet combination makes sense and what frequency extention you get with the high pass. As I said, the simulated capacitor size may not be the optimum, but not because of the programs limitation, but the unpredictable properties of the actual driver, box and wireing. So if the drivers TSP are perfectly measured and the cabinets data is also entered 100% right, the capacitor will be spot on too. Remember that the usual cabinet, if not made from concrete, is somehow elastic and stores and takes up energy.
So if you simulate in AJHorn and then fine tune the capacitor by measureing, you will get what you have simulated. This makes speaker building much simpler, as you do not waste time and money with test cabinets.
FWIW, Hornresp (free) can simulate passively-assisted sealed alignments as well.