What happens with different port tunning

[andy2]

Given the same driver, what would happens if the port are tuned at different frequency?

For example, how would the bass affected if the port is tuned at 30hz, 35hz, 40hz?

If you want tight and accurate bass, how would you tune the port?

[Pan]

Let´s say the 35Hz tuning results in the flattest response, then the 30Hz tuning will sound tighter and 40Hz a little loose/full/round.

Possibly there could be exceptions but in general this is the way it is.

Try to model a driver, build a box and put the vent tube outside the box. Use short pieces of tubes that can be added on eachother with some sticky tape and play with the length. Fun and educational (is that a word?).

/Peter

[5th element]

Indeed, educational is a word.

[andy2]

I have another question,

should you tune the port at lower than the driver fs or higher than the driver's fs?

[Pan]

Normally slightly above... or at Fs. It depends on the Q of the driver and what you want it to do.

A low tuned box/driver will begin to roll off earlier than a box tuned for max flat response, however this can have the advantage of giving tighter bass, partly due to the tuning itself but also becasue the gradual roll off can fit the room better with a better result overall.

I´d go so far as saying much of the bad rep reflex speakers have as compared to IB/closed boxes is becasue the (often much more) flat response excites room resonances/modes much more. This boominess that results is mostly the room. Take that "boomy" speaker and listen to it outdoors or in a car and chances are you will find it relatively tight and punchy.

Or to put it like this; remove the room and the discussion of dipoles vs. closed box. vs. reflex loading and so on is not needed anymore... sort of.

/Peter

[Ron E]

Originally posted by andy2
I have another question, should you tune the port at lower than the driver fs or higher than the driver's fs?

Fb=Fs*0.42*Qts^-0.96
Vb=Vas*20*Qts^3.3

If you play with these formulas a bit, you will find that generally if Vb>Vas then Fb<Fs and vice versa. The dividing line between the two regions is about Qts~=0.4. In general it is not a good idea to use a driver with Qts>0.5 in a ported box.....

[cjd]

Bass is an interesting thing. The best I've ever heard from a ported enclosure only hinted at tight bass. Sealed or dipole, yes - very clean, very accurate - the hallmarks of "tight" bass.

Altering the tuning frequency of a port does two things - it introduces a measure of control over cone movement, and it resonates out of phase with the driver and generally sums with the driver output.

Tune high and you introduce a rise in bass levels which isn't "loose", merely excessive and unbalanced. Tune for a "flat" response and you still get this rise due to room gain - not as badly, but it's there. Tune a bit low and you can get flat response in-room. Tune too low and you get a broad saddle with a peak at the tuning frequency. Leaning out the bass is often misinterpreted as "tight" bass.

Tune at Fs and you can significantly increase the output capability of the driver at that frequency - in a subwoofer, I have seen instances where it practically doubles its power handling. Of course, that's a few dB only but every little bit counts. In this case, response is only slightly stronger than in an infinite baffle. Otherwise, tune where it gives you the response you want. I have done a few projects with a driver that has an Fs at 116hz - If I port, I usually tune to ~75hz. Tune to Fs and it has a significant rise in response then falls like a rock.

Finally, because every room is different it may be difficult to guarantee results one way or another unless you consider room treatments and EQ. A speaker that sounds perfect in one room may be horrible in the next.

C

[Pan]

"Sealed or dipole, yes - very clean, very accurate - the hallmarks of "tight" bass."

Which I suggest is very much a speaker/room interface issue, you do not agree with this?

"Altering the tuning frequency of a port does two things - it introduces a measure of control over cone movement, and it resonates out of phase with the driver and generally sums with the driver output. "

The phase turns so in some places the port is IN phase and some places OUT of phase, and therefore add or subtracts to the cone direct radiating output.

"Tune high and you introduce a rise in bass levels which isn't "loose", merely excessive and unbalanced."

If you tune high you introduce higher groupdelay which is a time factor that together with the bump makes the speaker less cxontroled, worsens the transient response and makes it loose sounding. Also the room interaction becomes worse normall due to the fact that tuing often takes place wheree the room is plauged with standing waves. A high Q resonance (under damped) will sound loose becasue that is what it is.. loosely contrelod by the amp.. it follow the input signal worse than a response without a resonance or a lower Q resonance.

"Tune a bit low and you can get flat response in-room."

No tuning will give you flat in room response. The typical response is like +/-10dB or worse in the low range.

"Tune too low and you get a broad saddle with a peak at the tuning frequency."

This is mostly true for high Qtc drivers and not so with low Qtc drivers.

"Leaning out the bass is often misinterpreted as "tight" bass."

No, it often gives tighter bass. Partly due to better transient response from the box itself and also becasue the room interactin is better. Room contribution greatly affects the transient response = "tightness".

Cheers!

/Peter

[Svante]

Originally posted by Ron E


Fb=Fs*0.42*Qts^-0.96
Vb=Vas*20*Qts^3.3

If you play with these formulas a bit, you will find that generally if Vb>Vas then Fb<Fs and vice versa. The dividing line between the two regions is about Qts~=0.4. In general it is not a good idea to use a driver with Qts>0.5 in a ported box.....

Are those formulas based on theory or are they empirical?

[Ron E]

Originally posted by Svante
Are those formulas based on theory or are they empirical?

They are from Small-Margolis 1981 JAES - an article which condensed theory down into simple calculator programs for designing boxes.

They are curve fits of actual theoretical results. Basically the formulas can be duplicated by curve fitting the QB3 alignment for QL=7 for Qts<0.4048 and the C4 alignment for Qts>0.4048. Qts=0.4048 is the famous B4 alignment (which occurs at Qts=0.383 for Ql=infinity).

Depending on what range you use for Qts when curve fitting, you come up with different coefficients and exponents. Small-Margolis came up with the ones I just posted - which are used in WinISD for suggested alignment BTW D.B. Keele Jr. (former reviewer for Audio Magazine in the states and now a reviewer for the Audio Critic magazine - and holder of many patents in horn design, as well as writer of a few articles in JAES on 6th order alignments and parameter sensitivity in bass alignments, IIRC)

Small Margolis (1981)
Fb=Fs*0.42*Qts^-0.96
Vb=Vas*20*Qts^3.3
F3=Fs*0.28^Qts^-1.4

Keele (?)
Fb=Fs*0.42*Qts^-0.9
Vb=Vas*15*Qts^2.87
F3=Fs*0.26^Qts^-1.4

Hope all this helps someone