Size and position of the port in a ML TQWT

[teodorom]

Given that, in a ML TQWT, the port is the only opening of the (shaped) box, my simulations show that the port size is very critical (that is small variations lead to huge differencies), and that the ratio radius/lenght doesn't follow the standard vented rules (ML TQWT is not a vented box, as stated by MJK).
So I would like to understand if even the position of the port is critical and if there are some rules to follow.
Is there some formula relating radius and lenght ?
Thanks

[joensd]

You can use ported box sheet for simulating port position which surely works for a straight ML-TL.
Not sure though if works for more complex TQWT´s (like folded and tapered ones).

Help with effective port length in ML-TQWTs

[Bob Brines]

Since everything interacts, changing one parameter will require changing others. That said:

You will want the shortest, fattest pipe that will meet your bass cut-off goal. This will reduce the amplitude of higher harmonics.

You want to position the driver so as to suppress the 3rd harmonic (TQWT) or 5th harmonic (TL). This reduces, but will never eliminate the dip 150-300Hz.

You want the smallest diameter, shortest port that will meet your bass cut-off goal. This will reduce the amplitude of higher harmonics. You can check the minimum size required by modeling your driver in a program like Unibox. Make sure your driver Xmax is not exceded. However, a 2" port (with a 6" driver) will often model shorter than the material thickness, so I find that a 3" port usually works for a 6" driver and a 4" port for an 8" driver.

On the port SPL plot, you will notice a group of harmonics between 500-1000Hz. You want to position the port to suppress the middle of this group. This will drastically reduce the ripple in the combined SPL plot.

All of this moleling must be done with no stuffing. If you need more than 0.25lb/ft^3 stuffing to reduce ripple, you need to seriously reconsider your design. Stuffing sucks the live out of a TL. Stuffing is not a substitute for proper design.

(The frequency ranges quoted above assume a cut-off ~40Hz.

Bob

[teodorom]

Thank you Bob. Thank you for the N-th time ! You are so kind and passionate. And patient with me. I alway try to ask questions that are the most general as possible. I hope that you (and the Forum) appreciate that.
Thank you joensd. I didn't notice that thread. On this forum there is so much material ! I did what it's suggested and I have seen the results (very similar to my old ones).
Being now convincef of the quality of I could (I shall) build, I have a last (I hope) question: in ML TQWT the Velocity at the Terminus is one order of magnitude (epsilon=60 at the resonance) larger than in a "pure" TL.
Are there any problems related to that ?
Thanks