WinIsd port length question?

[trevor995]

[Chris Roemer]

If you play around in WinISD, you'll see that unless a port is VERY short (like, maybe < 6") being off by +/-1" is not going to make THAT big of a difference.

If you model a pair of 3"id ports, and WinISD says they should be 14" long, that's per EACH.
If you find it hard to remember, just remember this, a PAIR of 3" ports will have to be as long (approx.) as a single port of DOUBLE the area. THAT would be a single port w/an id of 4.2" (which is 3.0 times the square root of 2, which is about 1.4).

Using WinISD, I'll typically NOT subtract any internal volume for the driver's displacement (unless the magnet is HUGE), or for internal bracing (but all I use are "sticks"); but I WILL subtract the volume used up by the port structure - ESP. when: the port is a large dia., the port is LONG (over 6-8" or so), OR the box is small-ish (like, under 1-1.5cf). For ME, my port length is the length of the straight portion inside the vent, that means that I don't include the extra length of any roundovers.

I've had a WT3/DATS for many years now, and I've verified my port tunings (almost always end up within 1 or 2 Hz of target).

[billfitzmaurice]

I have read when simulating a box and using a slot (square) port to deduct half the port height from the length that WinIsd proposes for a specific tune. I gather this is more of a “rule of thumb” than exact science.

It's more of an old wives tale, although there is some truth to it. BTW, a slot isn't square, and that's the source of the tale. All ports have a section of high friction between the port air mass and the port wall, extending perhaps a quarter inch or so from the port wall, which has the effect of making the port area act as if it's smaller than the actual dimension. If you make a very narrow slot port that area of high friction will be a much higher percentage of the overall port area than with a square, round or other shape port that's not extreme in its length to width ratio, making the de facto port area considerably smaller than the actual area. Since less area equals lower tuning the port needs to be shortened to compensate. There is no exact formula I know of to calculate the length reduction. Since the reduced effective area also leads to increased port velocity, and therefore chuffing, I just avoid high length to width ratio ports entirely.

[Brian Steele]

It's more of an old wives tale, although there is some truth to it. BTW, a slot isn't square, and that's the source of the tale.

I suspect that it has more to do with the slot vents normally being run along one side or more sides of the box. The proximity of the side to the vent requires its length to be adjusted.

[billfitzmaurice]

I've seen that said many times as well, but my testing hasn't born it out. On the last major vented cab project I did I used round ports mounted in the baffle, easily swapped out for different lengths to see the results. Once I determined the tuning I wanted I made the actual prototypes with corner ports. They ended up giving almost identical results when made the same area and length as round ports baffle mounted, and that's with proximity to two sides. I bet the result with a slot port across the middle of the baffle would give the same result at the side of the box. It might need to be shortened, but because it's a slot, not because of where it's located. But I won't be the one to test it, as I stated I don't do slot ports.

[trevor995]

Thanks everyone, that clears it up.

[Brian Steele]

I've seen that said many times as well, but my testing hasn't born it out.

Interesting, my experience has been quite different. My "Enigma" bandpass build didn't use slot vents, but it did use long vents that were in close proximity to the sides and the rear of the box. I had to trim them back significantly to achieve the target Fb.