Moble Enclosurs 10+ year member
Box Designer/Builder
Well, what can happen is this:
The normal way that people follow a multi tuning enclosure for say, daily and spl use combined, is by creating two different ports with different characteristics, without changing the compression area much other than added displacement. So, when you have a port area that is constant, such as a round or aero port, you have a specific tuning that takes place. In other more complex designs, there are other effects that are much more important than tuning. Those are coupling and control of the response curve within a passband rather than a reliable tuning with a general cutoff.
SO, take this for example......you have a conventional ported design with two round ports, as stated by you above. With the port area of both combined and the length of those as another constant, you create a specific response curve that is not changed. Therefore you think that by covering a port, you will only merely change the tuning and passband of that previous response, but you can create problems. The main problem is that when a lot of designers create enclosures, they pay too much attention to the only curve they present to the clients, which is the response curve. But less attention is payed to other factors of the design, and some designers do not even pay attention to them at all. Those factors are either known or unknown. I will not go over them if you are unaware of them all.
Now, regarding that and covering the port, those changes will occur and they will either help or hurt the situation. When those other factors of the enclosure are considered, which they have to be, then they present changes with the response curve that are not shown in a normal calculated response. THESE HAVE TO BE CHANGED MANUALLY, which a computer program is not yet capable of. Not even hornresp. So, these changes have to be considered. I use 3 different programs that I combined to create the response I show the clients and it includes the changes that occur with the other factors.
So, by covering up a port, you can change one of those major factors, which is in fact the phase response. That one I will mention. ANd this may end up giving other designers the ability to create better or more accurate products, because the phase response is VERY IMPORTANT in the overall output of the design. If you have a design that is tuned to say 38Hz, and the response curve shows it goes down with a -3dB response to about 34Hz, for example, you will want to say that this design is capable of going down to 34Hz efficiently, but can go further below that. And that depends on the phase. If you have a normal conventional design with those 2 ports and cover one up, that changes the phase and not only that, will also change the port characteristics by cutting the volume in half, but not effecting the length. SO, for this, most designers will create ports with different lengths to create a different tuning. But this is not the case with your mentioned situation, but it definitely should be because that length change will and can balance out the phase response that is needed to keep the design operating within its calculated response curve efficiently, WITHOUT having to turn up the volume to compensate.
So, what you can create, and is not always true but as mentioned, a possibility if not done correctly, is an out of phase response just by closing a port and changing nothing else. You do have the chance to mess things up, but you can also have one that only helps the phase IF you designed it originally using a higher tuning and then created a lower tuning, which covering one port is capable of doing. SO, granted, it can work, but you still have to know how to figure for it, or else you just get lucky. There are no other options to it. either luck, or know what you are doing to the design. Pretty simple .
So, say you have a phase with a usable response from 1-242Hz. The driver specifications will provide a more broadband output along with subsonic filters, and low pass filters, but calculated you can have a HUGE response range, even above the nominal 1.5 octaves of a normal sub frequency system. So in other words, you can have a design that goes up to 120Hz or 80Hz, and still goes subsonic, just with lost output with a given slope, which is not natural....its based on the setting of the system. naturally, with a given 0dB capability with no losses, it is possible to get more out of that response, but naturally sub amps do not allow it due to excursion limitations as well, which is a major factor is lost dB levels below tuning. Above tuning, lost dB has a major factor of distance from the source.
Now, from that 1-242Hz range, you get a response phase that allows that to be used by the sub with the given mechanical limitations and resonacnes that get in the way. So, you end up with an efficient response from say 27-60Hz or whatever the design allows. This can be calculated and is not shown on a response curve.
NOW, say you cover a port up. You change the characteristics of that port. Thus changing the phase and output passband and essentially everything by its own factor of dominance and importance. Thus creating a different "tuning" and a different usable range. BUT, now you have to look at that range within the phase correlation of the entire response curve and then figure for what is usable then.
You can have, in a worse case scenario, and out of phase band that starts in the mid 30s and smooths out as the frequency increases. So, this is usually caused by covering up a port without accounting for specific factors of phase and tuning effects, and just relying on the lower tuning shown in the respond curve and saying, "I have two different tunings for my box". You can then narrowband the response range to something that has a higher rolloff rate at the lower end of the response curve, around the mid 30s, and lose more output below that based on covering the port ALONE. Not by any other change. ANd then have a higher efficiency as the frequency increases EVEN THOUGH by having the port area smaller, it should lower tuning, because tuning has no major effect on the response that you hear. And then that higher efficiency from the higher frequencies will still be faded because of the natural bandpass of the amplifier, so you have a constant output level there. SO, what happens, or what could happen, is you just lost lower end output by lowering the tuning! Sounds backwards, BUT completely possible.
Now, if you get into some other factors, which I said I would not mention, but one is already well known as well as phase, which is excursion limits and control. This changes as well. You can effect the output by changing that as well. So, where you once has a tuning with a low xmax when loaded, you now have a lower tuning with a peak at the previous tuning with an excursion factor that can exceed the maximum excursion of the drivers limitations! So, though you can tune and play lower(as long as you don't mess the phase up) you then have the possibility of blowing the sub mechanically due to overexcursion at a higher frequency! Again, one always thinks, " as frequency increases, xmax decreases". Generally yes, but in a conventional design you always have a peak excursion above tuning before it drops again! This should be known by anyone that figures for that as well.
I am sorry if this may be jumping around a lot, but it is easy to get confused when trying to explain everything that can occur without actually getting too detailed. because everything is effected. And if you do not understand what happens just by simple covering a port, then you have the change of things not working right or even causing damage if you are using max power levels all the time, which a lot of people do instead of getting a more efficient setup.
Now, I do not know if this explains a better way to do it, but it give you an insight on what to look for when doing it, which in turn makes it more accurate and safer to do. Keep in mind that this does not always occur! It can occur more often than not, and damage may not even be an issue in some cases due to proper power leeks being utilized, but still why take the chance when you can do it right the first time?
There was someone that came to me recently with a design from a very reputable designer, which I will respectfully remain anonymous, an=bout just this concept. I looked over the idea, and behold!....it had many issues with one of the ports not having the capability to provide the output within its "calculated" range of operation because no consideration was made to make sure it was correct. This is the exact reason why people say calculated is not as good as real world, when it comes to things like bandpasses and such...because the knowledge of this and how to control it is limited in the audio community. Even with a knowledge of its existence, meaning the factors mentioned, you still have to know how to account for them correctly.
And tell you the truth, by making a design with two ports and covering one up, you have a higher possibility to create a design that will be less optimized for the subs and the environment, though it can still play well, and be acceptable, it can be much improved at the same time.
I hope that helps you understand more of what I meant and I hope it helps the other designers to open up their capabilities of what can be accomplished and what it can take to make it right. This is not saying they do not already, but it can be shown in the work done if you pay attention to it.
It is not hard to make a design sound great. That is actually the simple part. Because since there are so many types of sounds that can be accomplished, the range of getting it right is great and the ear is very lenient to something that makes noise vs something that makes sound.
