Hi!
I would like to ask you for your advice about simulating bass reflex speakers.
My wife and me are about to build four 15 inch ported speakers for the local Reggae sound system. We are new to building loudspeakers. We are insecure what to build because simulators seem to disagree.
We would like to use the speakers above four 18 inch scoops between 80 Hz and somewhere near 400 Hz. The speakers will be close together in one stack and run in mono.
We did our simulations with Hornresp and with WinISD. We wish to thank all people who write and maintain these programs.
Hornresp shows a Helmholtz frequency of 55.1 Hz and a frequency response that falls by 1 dB between 70 Hz and 200 Hz. Falloff is very gradual. WinISD says the same cabinet is tuned to 71 Hz with a narrow 3 dB resonance near 88 Hz. Impedance minimum in each program is close to the frequency that the program states.
For this simulation we used the JBL 2226H datasheet, four cabinets with 87 litres each in half-space, cabinet resonances are masked. Temperature 20°C, inductance is included. Did we get something mixed up? Here are our input screens for simulation:
Should we worry about the peaky response from WinISD? The fall-off with frequency we get from Hornresp would be welcome but we're afraid of the peakiness in the WinISD result.
We are planning to tune the ports after assembling the cabinets but as it is, it's difficult for us to find a usable range for port length.
We imagine it's difficult to simulate the right kind of coupling between speakers in a specific case. We are aware there will always be some difference between simulations and measurements. Can Hornresp predict a peaky or flat response with multiple drivers? There have been some similar threads before but they seem to be asking different questions. We would be grateful for your help.
Best wishes
Axel
I would like to ask you for your advice about simulating bass reflex speakers.
My wife and me are about to build four 15 inch ported speakers for the local Reggae sound system. We are new to building loudspeakers. We are insecure what to build because simulators seem to disagree.
We would like to use the speakers above four 18 inch scoops between 80 Hz and somewhere near 400 Hz. The speakers will be close together in one stack and run in mono.
We did our simulations with Hornresp and with WinISD. We wish to thank all people who write and maintain these programs.
Hornresp shows a Helmholtz frequency of 55.1 Hz and a frequency response that falls by 1 dB between 70 Hz and 200 Hz. Falloff is very gradual. WinISD says the same cabinet is tuned to 71 Hz with a narrow 3 dB resonance near 88 Hz. Impedance minimum in each program is close to the frequency that the program states.
For this simulation we used the JBL 2226H datasheet, four cabinets with 87 litres each in half-space, cabinet resonances are masked. Temperature 20°C, inductance is included. Did we get something mixed up? Here are our input screens for simulation:
Should we worry about the peaky response from WinISD? The fall-off with frequency we get from Hornresp would be welcome but we're afraid of the peakiness in the WinISD result.
We are planning to tune the ports after assembling the cabinets but as it is, it's difficult for us to find a usable range for port length.
We imagine it's difficult to simulate the right kind of coupling between speakers in a specific case. We are aware there will always be some difference between simulations and measurements. Can Hornresp predict a peaky or flat response with multiple drivers? There have been some similar threads before but they seem to be asking different questions. We would be grateful for your help.
Best wishes
Axel
Last edited:
Do you want to use the JBL speaker you mention? If not, first pick a speaker that can do what you want. For this you must first know what you need.
Maybe describe what you have and why you want to add someting. Just throwing in a simulation and asking why the simulator is wrong (it isn't!) may not earn you a lot of answers.
Simulators work very simple: garbage in gives garbage out. Also known as GIGO.
So if you get the input wrong, the output will be wrong too.
If you are new to speaker building, stick to WinISD for a while. Hornresp is much more complicated and not really needed for a simple vented design. If used right both programs give identical results.
There is no need to tune your speakers when they are build. A simulation is more exact than any tuning you can do.
Maybe describe what you have and why you want to add someting. Just throwing in a simulation and asking why the simulator is wrong (it isn't!) may not earn you a lot of answers.
Simulators work very simple: garbage in gives garbage out. Also known as GIGO.
So if you get the input wrong, the output will be wrong too.
If you are new to speaker building, stick to WinISD for a while. Hornresp is much more complicated and not really needed for a simple vented design. If used right both programs give identical results.
There is no need to tune your speakers when they are build. A simulation is more exact than any tuning you can do.
Hi Axel, & welcome to the learning curve!
As Turbowatch2 says, if you’re careful to keep the inputs the same, you’ll get close agreement between Hornresp & WinISD.
In your case, your WinISD model is tuned to 71 Hz, not the 55Hz you’re looking at in Hornresp.
If you reduce the tuning frequency in WinISD to 55Hz, you’ll see the response curve is pretty much the same as in Hornresp. What that will do however is increase the required length of each vent.
The difference arises in how each program handles (or doesn’t) multiple vents. Hornresp simply takes the total cross sectional area of the vents, and does not pay any attention to whether that is made up of 1 large vent or multiple small ones. WinISD (and many other programs) does take account of this. If you model one larger vent in WinISD, tuned to the same 55Hz as you have Hornresp set to, you’ll get a vent length that is back down closer to the length in Hornresp. There is still a slight difference which I think may be due to end correction, but I’m not 100% sure on that.
HTH,
David.
As Turbowatch2 says, if you’re careful to keep the inputs the same, you’ll get close agreement between Hornresp & WinISD.
In your case, your WinISD model is tuned to 71 Hz, not the 55Hz you’re looking at in Hornresp.
If you reduce the tuning frequency in WinISD to 55Hz, you’ll see the response curve is pretty much the same as in Hornresp. What that will do however is increase the required length of each vent.
The difference arises in how each program handles (or doesn’t) multiple vents. Hornresp simply takes the total cross sectional area of the vents, and does not pay any attention to whether that is made up of 1 large vent or multiple small ones. WinISD (and many other programs) does take account of this. If you model one larger vent in WinISD, tuned to the same 55Hz as you have Hornresp set to, you’ll get a vent length that is back down closer to the length in Hornresp. There is still a slight difference which I think may be due to end correction, but I’m not 100% sure on that.
HTH,
David.
Given that you should consider making them sealed for simplicity.
Why does your Winisd sim show a cabinet volume of 348L? I did a quick sim at 80l tuned to 48hz that produces a perfectly flat response for that driver.
I'm happy that so many of you have answered my post. Thank you very much.
So the difference between simulations results is 1 port versus 8 ports?
This is a simulation I did with one set of parameters in Hornresp and two sets in WinISD. WinISD1 has the same port length as the Hornresp simulation. WinISD2 is tuned to the same frequency as Hornresp but has a longer port.
Hornresp WinISD1 WinISD2
total volume 348 l 348 l 348 l
total port area 231.5 cm2 231.5 cm2 1231.5 cm2
port length 10.8 cm 10.76 cm 24.75 cm
frequency 55.1 Hz 71 Hz 55 Hz
In Hornresp, we used a single port, in WinISD eight ports.
So going from one port to 8 ports will double the port length for a given total port area and tuning frequency? Does the difference between one larger port and more smaller ports really make such a big difference?
Here's another simulation we did in Hornresp. We did one simulation with one large box with one port and four drivers. In the other we divided area and volume by four and put in "4 Parallel Speakers" in the Multiple Speakers tool. There should be four ports involved now. This is the result:
Grey is the single cab, black is the four cab simulation.
Results are almost identical. At least between four boxes with four ports in total and one box with one single port there's not a lot of difference in simulations.
How come the difference between eight ports in WinISD and four ports in Hornresp is so large when the difference between a single port and four ports is barely noticable? Are we doing something wrong?
I attached the files we used for simulation to this post. I If you want to have a go, just rename the WinISD files to *.wpr.
We chose the JBL 2226 because it has very high sensitivity around 80 Hz. The B&C 15NW76 and the RCF MB15N give very similar results in simulation if we adjust the cabinet volume slightly. They also give us the same problems when switching between Hornresp and WinISD. All three drivers have a reputation for reproducing low mids smoothly.
Best wishes
Axel
I hope you don't mind if I ask this question in the Subwoofer forum. I believe that people in this forum know a lot about simulation, that's why I ask.
Best wishes
Axel
So the difference between simulations results is 1 port versus 8 ports?
This is a simulation I did with one set of parameters in Hornresp and two sets in WinISD. WinISD1 has the same port length as the Hornresp simulation. WinISD2 is tuned to the same frequency as Hornresp but has a longer port.
Hornresp WinISD1 WinISD2
total volume 348 l 348 l 348 l
total port area 231.5 cm2 231.5 cm2 1231.5 cm2
port length 10.8 cm 10.76 cm 24.75 cm
frequency 55.1 Hz 71 Hz 55 Hz
In Hornresp, we used a single port, in WinISD eight ports.
So going from one port to 8 ports will double the port length for a given total port area and tuning frequency? Does the difference between one larger port and more smaller ports really make such a big difference?
Here's another simulation we did in Hornresp. We did one simulation with one large box with one port and four drivers. In the other we divided area and volume by four and put in "4 Parallel Speakers" in the Multiple Speakers tool. There should be four ports involved now. This is the result:
Grey is the single cab, black is the four cab simulation.
Results are almost identical. At least between four boxes with four ports in total and one box with one single port there's not a lot of difference in simulations.
How come the difference between eight ports in WinISD and four ports in Hornresp is so large when the difference between a single port and four ports is barely noticable? Are we doing something wrong?
I attached the files we used for simulation to this post. I If you want to have a go, just rename the WinISD files to *.wpr.
We chose the JBL 2226 because it has very high sensitivity around 80 Hz. The B&C 15NW76 and the RCF MB15N give very similar results in simulation if we adjust the cabinet volume slightly. They also give us the same problems when switching between Hornresp and WinISD. All three drivers have a reputation for reproducing low mids smoothly.
Best wishes
Axel
I hope you don't mind if I ask this question in the Subwoofer forum. I believe that people in this forum know a lot about simulation, that's why I ask.
Best wishes
Axel
Sorry, but it look's as if you are trying to do a professional installation, while erasing the professional from the project.
You found out about speaker simulation tools and thought " Hey. any idiot can design a speaker with this program, so I'm perfectly qualified for this!"
You put weight on the wrong aspects, as you have basically no clue about pro audio. This is a safe way to an expensive disaster.
A way around this would be to buy some finished pro audio speakers, best even powered. You may find at Thomann or the like. Just talk to them, they got German speaking professionals that know their job and are used to any kind of funny people. That would reduce your job to installing the speakers and wiring them up. Then your sound technician takes over and dials everything in..
Do you know that a speaker system that is operated in a commercial location has to meet certain safety standards?
Who is going to certify your build?
Or, the other way round: What is your profession? Bar man, carpenter, artist? The only thing you seem to know is that JBL is an expensive name in audio stuff. You ignore that the loudspeaker chassis you buy are only a very small part of the complete sound system.
You don't just buy JBL speakers from a discount source and have a perfect sound system... "The JBL Sound" or any good sounding PA is a little more than this.
You found out about speaker simulation tools and thought " Hey. any idiot can design a speaker with this program, so I'm perfectly qualified for this!"
You put weight on the wrong aspects, as you have basically no clue about pro audio. This is a safe way to an expensive disaster.
A way around this would be to buy some finished pro audio speakers, best even powered. You may find at Thomann or the like. Just talk to them, they got German speaking professionals that know their job and are used to any kind of funny people. That would reduce your job to installing the speakers and wiring them up. Then your sound technician takes over and dials everything in..
Do you know that a speaker system that is operated in a commercial location has to meet certain safety standards?
Who is going to certify your build?
Or, the other way round: What is your profession? Bar man, carpenter, artist? The only thing you seem to know is that JBL is an expensive name in audio stuff. You ignore that the loudspeaker chassis you buy are only a very small part of the complete sound system.
You don't just buy JBL speakers from a discount source and have a perfect sound system... "The JBL Sound" or any good sounding PA is a little more than this.
Sorry, I got the table in my last post garbled. I simulated one cab geometry in Hornresp. It should read:
This is a simulation I did with one set of parameters in Hornresp and two sets in WinISD. WinISD1 has the same port length as the Hornresp simulation. WinISD2 is tuned to the same frequency as Hornresp but has a longer port.
Ther was only one port in the Hornresp simulation file but there were eight ports in WinISD. Does the difference between one larger port and eight ports with the same total area really change the tuning by so much?
Best wishes
Axel
This is a simulation I did with one set of parameters in Hornresp and two sets in WinISD. WinISD1 has the same port length as the Hornresp simulation. WinISD2 is tuned to the same frequency as Hornresp but has a longer port.
Ther was only one port in the Hornresp simulation file but there were eight ports in WinISD. Does the difference between one larger port and eight ports with the same total area really change the tuning by so much?
Best wishes
Axel
| Hornresp | WinISD1 | WinISD2 | |
| total volume | 348 l | 348 l | 348 l |
| total port area | 1231.5 cm2 | 1231.5 cm2 | 1231.5 cm2 |
| port length | 10.8 cm | 10.76 cm | 24.75 cm |
| frequency | 55.1 Hz | 71 Hz | 55 Hz |
OK., but he's not going to build a 4 driver box.. it would be hugh. I know it can be done that way but I like to sim drivers in singles.
Mellow Mood, when you are doing sims the tuning frequency has to remain the same for all of them otherwise you are comparing apples to oranges, if the software defaults to a different tuning you may have done something wrong but you can also just change it back to the tuning you want.
Hornresp is calculating the end corrections assuming a single port tube with a cross-sectional area of 1231.5 cm2. Not sure what assumptions WinISD is making when calculating the end corrections for multiple port tubes, but they obviously produce a different outcome. I would have thought that the two sets of results should be a lot closer than they are.
Hi all,
I have to admit I don't fully understand the underlying physics behind this, but the formulae I've seen for calculating vent length are usually some variation on this:
where Lv = Length of vent, Dv = Diameter of vent, Np = number of vents, k = end correction factor, Fb = net box volume and Fb = vent tuning frequency, and 23542 is a reduction of a couple of constants involving the speed of sound and Pi.
What that shows is that if we take multiple smaller vents, they get less end correction (due to having smaller diameter to start with) than a single larger vent would get, all else being equal.
Every ported box simulator I've played with has certainly followed this pattern, WinISD included.
By only having a single field for total vent area, Hornresp seems to miss this adjustment. How much it matters in real life, I don't know, as I've never built a test box to check.
HTH,
David.
I have to admit I don't fully understand the underlying physics behind this, but the formulae I've seen for calculating vent length are usually some variation on this:
where Lv = Length of vent, Dv = Diameter of vent, Np = number of vents, k = end correction factor, Fb = net box volume and Fb = vent tuning frequency, and 23542 is a reduction of a couple of constants involving the speed of sound and Pi.
What that shows is that if we take multiple smaller vents, they get less end correction (due to having smaller diameter to start with) than a single larger vent would get, all else being equal.
Every ported box simulator I've played with has certainly followed this pattern, WinISD included.
By only having a single field for total vent area, Hornresp seems to miss this adjustment. How much it matters in real life, I don't know, as I've never built a test box to check.
HTH,
David.
Hello David McBean,
I'm very happy you joined the conversation. I love Hornresp. I would like to thank you for sharing it with us.
I'm very interested in how simulation programs model coupling between speakers and ports. I'm expecting that this is critical when multiple cabs play together.
My wife and me are planning to build four cabs. It would be disappointing if we built one cab that sounded perfectly sweet, then we built three more and the stack sounded all boomy because of mutual coupling.
David Morison, thank you for your discussion of vent length correction.
I tried if Hornresp and WinISD will give me similar results if I put in the same number of ports. This way ports will have the same size in both programs. I went for the maximum area that WinISD allows for four ports.
If I understand things right, I simulated four cabs with one speaker and one port each in Hornresp. Volume in Hornresp was 87 litres for each cab, port area 154 cm^2 for each port.
Here's my result:
So although both programs now simulate the same port size, results differ by 7.7 Hz.
In WinISD I used "Num. of drivers = 4" and "Number = 4" on the vent page.
In Hornresp I set "Ang = 2.0 Pi" and used the "Multiple Speakers" tool "4 Speakers in Parallel" for plotting acoustical power.
Then I switched to "Ang = 0.5 Pi", to see the Helmholtz frequency. I plotted acoustical power again, this time for a single speaker, and compared the results. As expected, the plots were virtually identical.
In both programs, Q was 7.
I checked and checked again and again and I hope I didn't make some trivial mistake.
I'm using the newest version of Hornresp and WinISD 0.7.0.950 on Windows 10.
Does WinISD assume that there's less coupling between ports than in Hornresp?
If we put our ports close enough together - will they behave less like separate ports and more like a single one?
Best wishes
Axel
I'm very happy you joined the conversation. I love Hornresp. I would like to thank you for sharing it with us.
I'm very interested in how simulation programs model coupling between speakers and ports. I'm expecting that this is critical when multiple cabs play together.
My wife and me are planning to build four cabs. It would be disappointing if we built one cab that sounded perfectly sweet, then we built three more and the stack sounded all boomy because of mutual coupling.
David Morison, thank you for your discussion of vent length correction.
I tried if Hornresp and WinISD will give me similar results if I put in the same number of ports. This way ports will have the same size in both programs. I went for the maximum area that WinISD allows for four ports.
If I understand things right, I simulated four cabs with one speaker and one port each in Hornresp. Volume in Hornresp was 87 litres for each cab, port area 154 cm^2 for each port.
Here's my result:
| Hornresp | WinISD | |
| number of drivers | 4 | 4 |
| number of ports | 4 | 4 |
| total volume | 348 l | 348 l |
| total port area | 616 cm2 | 615.8 cm2 |
| port length | 7.2 cm | 7.19 cm |
| frequency | 47.4 Hz | 55.1 Hz |
So although both programs now simulate the same port size, results differ by 7.7 Hz.
In WinISD I used "Num. of drivers = 4" and "Number = 4" on the vent page.
In Hornresp I set "Ang = 2.0 Pi" and used the "Multiple Speakers" tool "4 Speakers in Parallel" for plotting acoustical power.
Then I switched to "Ang = 0.5 Pi", to see the Helmholtz frequency. I plotted acoustical power again, this time for a single speaker, and compared the results. As expected, the plots were virtually identical.
In both programs, Q was 7.
I checked and checked again and again and I hope I didn't make some trivial mistake.
I'm using the newest version of Hornresp and WinISD 0.7.0.950 on Windows 10.
Does WinISD assume that there's less coupling between ports than in Hornresp?
If we put our ports close enough together - will they behave less like separate ports and more like a single one?
Best wishes
Axel
Last edited:
Turbowatch2, I don't understand why you are getting so worked up. You are making a lot of negative assumptions.
I've been designing analog circuits for a living for the last twenty years.
I did a lot of reading on loudspeakers but there's still loads for me to learn. That's why I'm asking.
We are a bunch of enthusiasts with a small sound system that we run for the love of Reggae. Now we're wishing to build some speakers to share our music with others. We take care of the safety issues you're worried about.
Relax, and cool down, man. Things in life are not always as bad as they appear.
Peace
Axel
I've been designing analog circuits for a living for the last twenty years.
I did a lot of reading on loudspeakers but there's still loads for me to learn. That's why I'm asking.
We are a bunch of enthusiasts with a small sound system that we run for the love of Reggae. Now we're wishing to build some speakers to share our music with others. We take care of the safety issues you're worried about.
Relax, and cool down, man. Things in life are not always as bad as they appear.
Peace
Axel
The results will be exactly the same if Rg is set to zero and QL is set to Lossless, as shown in the attachments.
Not sure what WinISD assumes.
Yes.
Attachments
Sorry, somehow we got a bad start.
You may not realize, but you are degrading the work of any audio professional to something you can do and learn in half a day by getting a free simulation program and sending your wife to the "Baumarkt".
I often hear such things like "any idiot can build a box and screw in a speaker", mostly from idiots by the way. Then, later, I see the results...
If you want to build a speaker that is able to compete with a descend commercial product, there is more to do than buying the most expensive JBL chassis and building some random cabinet for it.
A speaker starts with the cause you need it for, the music, than the room comes in play, existing components and your budget, but most important in DIYS is what you can do yourself.
If you are a designing analog circuits, you should get along with WinISD and Hornresp. You seem to have problems with that and blame the free software for your inability to feed it the right input and interpret the output. Why don't you use just one, the simpler and more suited, as you were told?
Soldering some analog stuff does not make you a good carpenter.
If you are a good carpenter, you did not learn anything about building a speaker cabinet. The problem is, this professsion knows "any idiot can build a speaker cabinet." Heard that before? The worst functioning and most expensive speaker housings are build by very good carpenters. I have seen to many of those.
Now comes the next problem: You need an objective description of what you have. Objective is the problem. How can someone who never had anything to do with PA find out what he has and what would be necessary to make it better? Probably a measuring microphone, a DSP and a new amp would suit you better with your existing stuff than 4 new problems (=say: speakers), but who knows. Not you...
Four of these chassis you mentioned in the beginning, will be enough to make more than 250 people dance, used as mid drivers. They are not made for 5 people, easily listening to some music in the evening.
You ask stupid questions, but do not answer any or give the needed additional information. Maybe you think this forum is a service and members have to answer? Some news for you, no one gets paid here. If you want this thread to turn out something productive, maybe try a new start.
Peace, love and happiness!
PS Usually I would give you some names by PM, so you could ask a pro on the telephone, but I fear you would only waste their costly time, as you don't know or don't want to tell what you really want.
You may not realize, but you are degrading the work of any audio professional to something you can do and learn in half a day by getting a free simulation program and sending your wife to the "Baumarkt".
I often hear such things like "any idiot can build a box and screw in a speaker", mostly from idiots by the way. Then, later, I see the results...
If you want to build a speaker that is able to compete with a descend commercial product, there is more to do than buying the most expensive JBL chassis and building some random cabinet for it.
A speaker starts with the cause you need it for, the music, than the room comes in play, existing components and your budget, but most important in DIYS is what you can do yourself.
If you are a designing analog circuits, you should get along with WinISD and Hornresp. You seem to have problems with that and blame the free software for your inability to feed it the right input and interpret the output. Why don't you use just one, the simpler and more suited, as you were told?
Soldering some analog stuff does not make you a good carpenter.
If you are a good carpenter, you did not learn anything about building a speaker cabinet. The problem is, this professsion knows "any idiot can build a speaker cabinet." Heard that before? The worst functioning and most expensive speaker housings are build by very good carpenters. I have seen to many of those.
Now comes the next problem: You need an objective description of what you have. Objective is the problem. How can someone who never had anything to do with PA find out what he has and what would be necessary to make it better? Probably a measuring microphone, a DSP and a new amp would suit you better with your existing stuff than 4 new problems (=say: speakers), but who knows. Not you...
Four of these chassis you mentioned in the beginning, will be enough to make more than 250 people dance, used as mid drivers. They are not made for 5 people, easily listening to some music in the evening.
You ask stupid questions, but do not answer any or give the needed additional information. Maybe you think this forum is a service and members have to answer? Some news for you, no one gets paid here. If you want this thread to turn out something productive, maybe try a new start.
Peace, love and happiness!
PS Usually I would give you some names by PM, so you could ask a pro on the telephone, but I fear you would only waste their costly time, as you don't know or don't want to tell what you really want.
David, thank you very much for clearing that up.
I think I found the reason why I got different results in WinISD than in Hornresp.
a) If I put in one cab with one driver and one vent, Helmholtz frequency in both programs is the same.
b) If I simulate four cabs, each with the same individual size as before, each with one driver and one vent, the tuning frequency drops in Hornresp. In WinISD it stays the same.
I guess that's why my results from the two programs didn't agree.
WinISD didn't allow me to put in four cabs so I put in one larger cab with four vents and four drivers.
case a)
1 driver, one port, volume 87 litres, port length 7.2 cm, port area 154 cm^2, radiation angle 2 pi, frequency 55 Hz.
case b)
WinISD:
4 drivers, 4 ports, 348 litres, port area = 615.8 cm^2, 55.1 Hz. The program calculated port length = 7.19 cm, same as before.
Hornresp:
same values as in case a) except for radiation angle 0.5 pi.
In WinISD, both impedance curves were identical, in Hornresp they shifted to the left.
I'm new to this but I would guess that at bass frequencies a 90 litre cab won't have narrow directivity. And I would expect that bass cabs will play deeper when several are placed in a stack, playing the same input signal. I know this is the case with horns. Is the effect similar in strength with bass reflex speakers?
I wish you a happy Easter
Axel
Last edited:
No at these frequencies the output is omni directional.
No the response of a cluster of reflex cabs is the same as a single... just higher in level.
As for the ports issue, for packaging purposes you are probably not going to want to use a single large port.. particularly with modern high performance drivers as it will make for a very large baffle... not that that is inherently bad but it can make the enclosure unwieldly. Instead consider a design with 2 ports/driver and maybe non circular in shape, have a look at current mass produced designs for some ideas.
No the response of a cluster of reflex cabs is the same as a single... just higher in level.
As for the ports issue, for packaging purposes you are probably not going to want to use a single large port.. particularly with modern high performance drivers as it will make for a very large baffle... not that that is inherently bad but it can make the enclosure unwieldly. Instead consider a design with 2 ports/driver and maybe non circular in shape, have a look at current mass produced designs for some ideas.
Correct.
In theory yes, provided that the cabinets are close together.
See attached comparison. The black trace show the power response for 4 bass reflex cabinets connected in parallel and the grey trace shows the power response for a single cabinet. The input voltages have been adjusted to normalise the levels so that the bass extension can be more readily seen.
Attachments
As previously advised, they will.
It is interesting that the single large port and the equivalent-area cluster of four smaller ports shown in the attachment will be treated as being acoustically the same in Hornresp, but not so in WinISD, because of the way that the two programs calculate end corrections. In practice I would expect the two port configurations to perform the same.
