PDA

View Full Version : Port Resonance



allen mueller
03-05-2008, 05:20 PM
I have been working out my box calculations for my Le14H-3 project. I am planning build a 3.5 ft^3 box tuned to 30Hz. I was planning to use a single 4" port that is flared at both ends. When I calculate the length in WinISD I see that the first resonance falls below the crossover point I would like to use of 1000-1200 Hz.

I could step down to a 3" port which would raise the resonance, but than my vent mach goes up. :banghead:

I have also tried a few multi port variations but run into the same problems.

I read some where in my google searches, that I don't have to worry about resonance, until my port length reaches several times the diameter. Is this true? In my case my port length will be around 8" for the 4" dia port.

I know a few of you have built 2 way projects based on the LE14 variants. Did anyone have to make a similar compromise in box design? Or did you find the resonance or vent mach was not significant enough to become a factor? Which would be the better direction to go, step down the diameter and raise vent mach or keep 4" and potentially have resonance?



Thanks,

Allen

whizzer
03-06-2008, 09:48 AM
Others may disagree, but I find noise caused by high wavefront velocities through the port more of an annoyance than a resonance falling within the driver's pass band. Since the resonance will be of a relatively high frequency, compared to the port's principal output, and since the damping material inside the enclosure is most effective at absorbing these shorter wavelengths, it should pose less of an actually audible problem. If, however, you believe you are hearing evidence of that resonance, then the solution is to prevent midrange frequencies from reaching the port at an audible level, which may mean repositioning damping material closer to the backside of the driver cone--the damping is basically transparent to the frequencies you want radiated by the port anyhow--or by actually adding a bit more damping directly behind the driver. These measures will usually work.

Hoerninger
03-06-2008, 11:09 AM
Do you mean that the tube itself has a resonance defined by its length?

Once I read about it that there can be made some rows of holes around the middle of the tube. This reduces the tube resonance.

The idea behind is that at the low bass reflex alignment (defined by the mass of air in the tube and the box volume) these holes are "unvisible" but at the high tube resonance they are open.
I think there may not be too many holes.
____________
Peter

jerv
03-06-2008, 11:41 AM
Do you mean that the tube itself has a resonance defined by its length?

Once I read about it that there can be made some rows of holes around the middle of the tube. This reduces the tube resonance.


Yes. The tube's resonance frequency is (mostly) a function of pipe length and width. The resonance can be quite substantial, and is often at an inconvinient frequency. I have also had this problem in one of my LE14A cabinets: it was at 760 Hz, only 6-7dB below max port output at 34 Hz and clearly audible.

It is possible to damp the resonance by stuffing the tube (old socks are frequently suggested:)), but the resonance frequency and Q of the enclosure will change.

Some have suggested another solution: make the port about 1.4 times longer, then drill about 4-6 8mm holes inthe port "wall" at half length. This helps killing standing waves in the port.

I have used this solution with good effect. For optimum tuning you will have to measure actual enclosure resonance and maybe experiment somewhat with the tube length.

Espen

Ian Mackenzie
03-06-2008, 12:22 PM
In the Loudspeaker Cook Book there are some ideas. You need to read the whole subject to understand the graphs and simulations. Even if you don't have anold edition it will always come in handy.

duaneage
03-06-2008, 07:44 PM
I would build and test. Make a box from cheap material and start testing it. Multiple ports can equal one large port, to tune you would cut only one.

Another idea is to use curved tubes, they present a changing radius through the length. Still another idea is to mount the port at an angle, this might prevent resonances from developing.

I've built a few dozen speakers and never had the problem your concerned with, I would go for 4 inches and leave it at that.

jerv
03-07-2008, 02:02 AM
II know a few of you have built 2 way projects based on the LE14 variants. Did anyone have to make a similar compromise in box design? Or did you find the resonance or vent mach was not significant enough to become a factor? Which would be the better direction to go, step down the diameter and raise vent mach or keep 4" and potentially have resonance?

For me, I would make the holes as described in my previous post above, and put the port at the back of the cabinet. Various methods of damping/stuffing can help some. Since tube resonance is a function of tube length and width, angling ports does in my experience not help much.

I also agree that chuffing air noises from too small ports is even more objectionable than port resonance. Stay with the 4" port.

allen mueller
03-07-2008, 03:55 AM
Thanks everyone, I will proceed with a test box and the 4" dia port.

If I encounter any problems I'll have to try some of the methods above.

Would it be possible to use an RTA and a mic placed very close to the port to see if i have any significant unwanted resonance coming from the port?


Allen

duaneage
03-08-2008, 05:23 PM
Any resonance recorded at the port will be very small and probably a single note only. In actual practice I doubt very much your going to notice any problems.

jerv
03-10-2008, 09:30 AM
IMO, a port resonance can certainly be large enough to be objectionable.

Below is a port measurement of a 2235H in a 144 ltr cabinet, tuned to 26 Hz. The single port is 11 cm (4 1/4 inch) diameter, length 21 cm. The cabinet has 5-10 cm glass wool damping on all inner panels. No attempt has (yet) been done to damp the resonance in the port. The microphone is placed in the nearfield - about 1 cm from the port.

As you can see, the first port resonance is at 616 Hz (as predicted by Unibox). It is only 2-3 dB below the level of maximum port output. The next harmonic resonance is at 1230 Hz, some additional 8-10 dB down.

Other woofer/cabinet/ports combinations exhibit similar behaviour, though the resonance frequency of course varies. IMO, this resonance is often audible, if untreated and if the resonance lies within the operating range of the woofer.

allen mueller
03-10-2008, 09:51 AM
Thanks Jerv,

I have the necessary equipment to RTA the box using a demo version of Smaart.

I was hesitant about measuring the port response because I know that in improper measurement would meaningless. If checking the ports response is possible by placing the mic in close proximity to the port, than that sounds within my ability.

If I could measure a response close to what winisd predicts than my measurements could be verified to an extent. From that point that I could try some of the techniques mentioned in this thread to help get rid of the resonance. I just have to be sure my measurement technique is accurate, to be able to do use the data.

Allen

duaneage
03-10-2008, 09:13 PM
Using a single 4 inch tube more than twice it's length with a driver like the 2235 is a mistake. That driver needs at least 2 ports to effectively damp that driver. For a 12 in a medium box you can get away with it, especially rear vented, but it is recommended that port area closely approximate driver area for extreme duty, and at least 50% for high levels. IF the Sd matches the port area you get best results.

Port resonance and power compression occur more severely in systems that have ports undersized for the driver.

Look at the ports used in JBL systems like the 4350, 460B etc. Lots of port area there. I would go with at least 2 ports, adjust only one to find final tuning. I have built sub systems with 15 inch drivers that used 4 each 4 inch tubes and the air was pretty tight across them all at resonance.

Jan Daugaard
03-11-2008, 01:41 AM
When a 2235H in a 144 litre cabinet is equipped with a port the same size as the woofer, i.e. 896 cm², the port needs to be over 3 metres long to achieve a tuning of about 25 Hz ...

Ian Mackenzie
03-11-2008, 05:20 AM
You guys need to look at JBL's 4430 design and figure it out from there.

jerv
03-11-2008, 12:10 PM
Using a single 4 inch tube more than twice it's length with a driver like the 2235 is a mistake. That driver needs at least 2 ports to effectively damp that driver.


I can agree on that one. But the original posters issue was about a single 4-inch port.

Any port size is a result of compromises. Bigger (or more) ports is of course desirable from a perspective of air noise, etc - but bigger ports is also longer (for a given tuning), and has the pipe resonance at a lower and often more unconvinient frequency.

In my opinion, the undesirable pipe resonance in bass reflex tubes is a factor that should be taken into consideration in any design. If the pipe resonance lies within the woofer's operational range (seldom with a subwoofer, often the case with bass units that goes up to 1 kHz) it can be treated with damping, moving the port to the back, holes or slots in the pipe "wall", etc.

A9X
03-13-2008, 10:14 AM
If still in the design stage, and you have carpeted floors, consider placing the port on the bottom of the enclosure and having it raised 2" or so. This will greatly reduce audibility of higher freq noises from the port directly.

Damping and retuning the enclosure innards could be very effective at 600Hz to minimise problems issuing through the cone.

allen mueller
03-14-2008, 06:15 PM
I have begun the build, and things are moving smoothly. I decided to make the front baffle removable, so that if things turn out poorly I can try different port diameters if need be.

WinISD predicts a 1st resonance around 900Hz. I can go down lower with my horn below that frequency so that looks to be my best option to avoid unwanted resonance.

I'd like to thank again everyone who has helped me with this.

Allen

Ian Mackenzie
03-14-2008, 10:42 PM
The answer may lie in using a curve (elbow) pvc tube.

allen mueller
03-15-2008, 01:53 PM
The answer may lie in using a curve (elbow) pvc tube.

Thats one of the things I will try if I find that i have a resonace. I can cross over at 800 hz with out a problem for the driver (BMS coax) but i dont tend to like the sound of the horn that i'm using crossed that low (westlake style smith horn).