Port air velocity calculator – using WinISD
How to determine the air velocity of your bass reflex port?
Wouldn’t it be awesome to have a port air velocity calculator to tell you if your port is gonna be noisy? Happily, there is a popular free enclosure design software that does exactly that (upon many other functions). You guessed it! I’m talking about WinISD. I don’t know why I don’t use this app more often. I kind of dislike the way you select your driver, but maybe I’m being too picky. Something else I don’t understand is why have a list of existing drivers to choose from. I can’t trust that those parameters entered by someone else are correct. Or maybe the driver has been updated and the parameters are not valid anymore. My OCD clearly disapproved with this. Nevertheless, it’s a good app. Thumbs up!
What’s up with this air velocity?
First of all, this article presumes that you want to build a bass reflex box. It also presumes that you already know how to design and build one. If not, check out this article or this one. In conclusion, our interest goes solely around the noise generated by the port. Chuffing, is an alternative way of calling this noise. Let’s cut to the chase! Consider these 4 aspects of your speaker design :
- Diameter of the speaker.
- Power handling of the speaker.
- Port diameter.
- Tuning frequency.
These factors contribute directly to air velocity through the port. While speaker size and power handling is a given, port diameter and tuning frequency is at your discretion. However, when you open up a port air velocity calculator, you mainly checking to see the velocity for different port diameters. As a result, tuning frequency is kinda fixed from the beginning, and only the port diameter remains the variable.
How much is too much?
People like to see numbers. So, when do we start to get concerned about port noise? When the air velocity through the port gets too high, of course. If you ask conservative people, they will say that anything above 10 m/s can result in unwanted port noise. While this might be true for a straight pipe with rough edges, it’s not so true for a flared port.
If you study fluid dynamics you will find your answer. Yes, any gas or liquid is a fluid (including air). However, any fluid dynamics study presumes that the liquid goes only in one way. According to these studies, there are certain things you can do to make the fluid flow seamlessly.
How to design loudspeakers - video courses
For example, let’s say, you have a container of water and a pipe attached to it. The water will flow easier if the connection between the container and the pipe is flared. And let’s also consider that the other end of the pipe is not connected to anything and the water drops on the floor. The other end of the pipe makes no difference if it has a flare or not. In conclusion, a flared intake makes a difference, but a flared exit does not.
However, this isn’t the case in audio. Since the air moves back and forth through the port, each end of the port is both an entry and an exit for the air. In conclusion, it’s best if both ends of the port are flared. By doing so, you can go up to 30 m/s of port air velocity. If you know your port could exhibit noise, you can place it on the back panel or bottom panel, so it’s not on axis with the listener (seen on many floorstanding or bookshelf speakers).
Device under test
Let’s pick a random subwoofer and design a bass reflex enclosure for it and see what the port air velocity calculator tells us. I’m picking a 12″ subwoofer : the Dayton Audio RSS315HFA-8 (Amazon affiliate paid link). So let’s open up WinISD.
Click Create new project -> Add new -> And add all the parameters from the specification sheet of the speaker. The green values are introduced and the blue ones are calculated automatically, as we introduce sufficient parameters.
Then, create new project again and select your newly created driver. Choose 1 driver in normal placement and hit next. Vented box. Finally QB3 alignment for maximal flat response. The volume and tuning frequency are calculated automatically for this type of alignment. Which is 139.10 liters and tuning frequency of 19.92 Hz. The next thing we want to check is the port shape and size.
So for a 12″ woofer, a 4″ port (Amazon affiliate paid link) size is adequate. Also, make sure you check the “flanged at both ends” option. This does not affect the air velocity calculation, it’s only for the port length calculation. Now all we have to do is switch to the port air velocity graph.
Port air velocity calculator
Let’s check out the port air velocity graph :
Switch to “Rear port – air velocity”. Why rear port? Because the front port is reserved for the front chamber of a 6th order bandpass (which has 2 chambers with a bass reflex port). For all other enclosures with one port, you will find the “rear” prefix.
If we analyze the graph, we can see that we maxim port velocity is near the resonant frequency of the box. Which is around 20 Hz. This is no surprise. However, a maximum air velocity of only 2 m/s is quite unreal. This is because the default signal is set to 1 W. Normally, you would crank up your sub, right?
Check out your subwoofer power handling. For this Dayton woofer it’s 400 W. Set the signal to 75% of that value. In my case, 300 W.
Now the port air velocity calculator spits out the number we are after : 42 m/s. This is on the high side. You can expect some audible chuffing at high volumes. If you are going to place the sub in the trunk of a car that is absolutely no issue. If you think this is going to be a problem, you could try and change the resonant frequency of the box. Keeping the same volume and increasing the resonant frequency will only make matters worse, so it’s a combination of both volume and tuning frequency tinkering.
Alternatively you can increase the port size.
As you can see, just by increasing the size by 1 inch, we are in the safe zone : 27 m/s. Good luck finding a 5″ port! I just wanted to get the point across though.
Conclusion
When you design a bass reflex enclosure, port size is important. I always recommend to go as large as the design permits. There are limitations, because larger diameter means longer port. Anyway, the port air velocity calculator found inside WinISD can help you out with numbers, to choose the correct port size. Alternatively, you can play around with box volume and tuning frequency to reach a lower port air velocity. It’s up to you how you do it.
References
- Image source : link.
13 comments
In this example, what size/length slot could take the place of the 5” port?
Thanks, Tony
The shape of the port doesn’t matter. So a 5″ diameter port has an area of 19.62 sq inches. Any slot port with the same area (19.62 sq inches) will have the same result (given that the length is the same also).
WiISD calculates the length of the port automatically if you input the shape and size.
Great article! Thank-you for sharing your knowledge. The tutorial part is especially appreciated.
Is there any reason why this article would not also apply to ported bandpass enclosures?
Thank-you!
You have the option of designing a bandpass enclosure in WinISD, so you can calculate the port air velocity of your exact enclosure.
While flares definitely help, don’t you still see output loss due to port compression?
Once above 20m/s, even with flares as well as the Polk power port, I could not get away from the loss in output. And by the time I hit the 30m/s, I measured 3db down. I spoke with John at AE speakers, who initially tried to explain and later after my measurements discussed the phenomenon.
Bare in mind, I did not actually measure the velocity but went by WinISD simulated speed for the port diameters I used.
Well, this article takes into consideration primarily port noise. If a port is noisy, it’s basically a ruined speaker. Of course, it’s better to have a larger port, but depending on your application, there are physical limitations. If I take as an example my M-Audio BX5 studio monitors. They have a port of about 1.5″. The air velocity is so high that it moves my curtains, but there is barely any noise (which is the most important factor, in my opinion, when you are forced to use small diameter ports). Of course, when you start to under-size your port you run into other issues as well. As you keep decreasing the port size, the air impedance through the port gets so high, you basically have a leaky sealed box, rather than a true bass reflex.
Thanks so much for shedding more light on this. I do want to say that my quick research shows many other places stating 17 m/s is the best fastest port velocity.
I’m doing the evl 15 inch single sub, was going to do a 4 cuft box, and the port length for an 8inch diameter aeroport would be too long, is a 6 inch diam port too small for a 15incb sub?
Probably going to be noisy at high volume. Hard to tell without numbers. But since an 8″ port is not feasible, you can go for slot port.
A little off topic though, WinISD was mentioned here and signal increase from 1 to 300 W. Just curious how did your Cone excursion graph looked like? Mine went wild below Fs, could use only 4W from 300 nominal, not to exceed max. Ported version. Thanks!
A speaker in a bass reflex enclosure will easily exceed its max excursion below the resonant frequency of the box. Some amps come with subsonic filters which eliminate this problem.
Once you have your port velocity from WinISD, you can use the following tool to see how large your flares need to be to avoid chuffing:
https://www.subwoofer-builder.com/flare-it.htm
1) The power handling listed on the spec sheet may be irrelevant. Quite often we’re displacement limited. To check that, make sure your xMax is in your speaker model and select Cone Excursion from the drop down at the top. Now redo your signal “System input power” until the resonant peak of the speaker is kissing the xMax line. Depending on xLim you may be able to exceed that at the cost of distortion.
Also bear in mind that this is all frequency dependent. Since I am typically designing for a bass guitar, getting terribly worked up over a maximum air velocity at 30 Hz doesn’t make sense.
2) The help file in WinISD has an article about entering driver parameters with what the creators consider best practices, basically letting WinISD calculate unknowns and the preferred order in which to enter them. I use that rather than relying on the spec sheet.