Box and Port Tuning Frequency

[Martyn]

I have a few rudimentary questions that I'd like to ask. In each case, I thought I'd try to do it such that they are easily searchable and provide a useful resource for non-expert speaker builders. Here's the first question:

Let's start by assuming that you are looking at woofers for a ported design, and that you are using Unibox. How do you choose Fb, the system tuning frequency?

My understanding is that you want the two peaks on the impedance chart to be about the same height (and that Fb will be the low point between them), and for Fb and F3 to be close to each other. How close is close enough?

Your design criteria might impose a limit on the physical size of the box, so you try reducing Vb, but this pushes F3 up. Since a low F3 is another of your design criteria, what do you look for in arriving at a happy compromise?

You decide that you can tolerate a significantly larger box after all and find that this gives you the F3 that you're looking for, but the F3 has now drifted further away below Fb. Again, how do you arrive at a happy compromise?

What else do you consider when deciding on system tuning?

[Evil Twin]

What you describe will get you into a fairly safe place in most cases, but there are more options and considerations to it than that.

One issue is the general driver T/S parameters, and how they influence the selection of the enclosure size- a simple ROT is that high sensitivity drivers will generally need larger boxes, but it depends on their VAS - if the driver Fs is 70 Hz, you're not going to be able to do the same things with it as one that is 35Hz. There's no getting around Hoffman's Iron law, regarding efficiency and low frequency extension for a given enclosure size. (well, you can sometimes cheat a little bit with EBS alignments, but that's usually only a consideration with subwoofers).

Another point is the general shape of the ported response curve from just above the Fb to below Fb. Basically, this is just the parameters for the tuning alignment, and will NOT necessarily produce the same curves when applied to theoretical optimal alignments for different drivers. The most popular are QB3, B4, and C4. (Quasi-Butterworth 3rd order; Butterworth 4th order, and Chebyshev 4th order). Those are roughly in order of increasing Q... which may be necessary, but is not necessarily a good thing, as that means more ringing or overhang in the transient response.

If you're experimenting with different drivers and want to play around with nominally optimal alignments as well as screw them up on your own I suggest using Kimmo's VituixCAD, the enclosure tool. There's a good database, there are many analytical options, and it interfaces well with the rest of his design tools.

https://kimmosaunisto.net/Software/Software.html

[Martyn]

Mr Hoffman certainly was a spoil sport. At the moment, I'm committed to a specific set of drivers, but I keep wondering whether I could have done a better job on my enclosure. Then, of course, there's always the next design...

Getting that ported response curve seems a little tricky - presumably one measures with the mic at the entrance to the port. I haven't thought much about using any kind of alignment for shaping the bass roll-off - other than to protect the woofer from damage or to reduce any excessive bass reinforcement from the room.

Actually determining and optimising the box tuning was going to be my next question, but I think I'm already in too deep! So far, I've measured the impedance which gives me Fb. It happens to be conveniently close to my Unibox target, albeit with a significantly shorter port (just happened to be a length I already had cut).

If anyone has any reading recommendations on this subject for non-electrical guys, I would much appreciate a link.

I'll check out that software!