I see a massive problem with this design. The constriction is way too small.
I've been kinda following along on and off so I'm not sure if anyone has mentioned it yet, but it looks like a huge issue.
Using the Akabak script in post 167 (and assuming it's correct - which is hard for me to do without checking everything but I don't have time for that), I turned off diffraction and boundary reflections, set Input Voltage to 36 to achieve 10 mm xmax, and checked Velocity at both ends of the constriction, which are Duct (1) D8 and Duct (2) D8. I always do this when any part of an enclosure seems too narrow.
Without the hpf, velocity at Duct (2) D8 is right around 30 m/s, with the hpf it's still up around 26 m/s.
Next it's off to Flare It to see how bad this issue is. But first some math. Reading the script it looks like the constriction area is 12 x 1.5 = 18 inches. My conversion program says that is equal to 116.1288 sq. cm. Changing that to a circular area suggests a diameter of approximately 12.2 cm. The details are important so please correct me if there are any mistakes here.
Flare It results for a 12.2 cm diameter port with no end correction:
And that's a big problem. The core velocity is fine but since the constriction is basically a port with sharp ends inside the main enclosure, chuffing at either end of the constriction appears to start at about 9 m/s at 35 hz. At xmax, with a velocity of 26 m/s, the blue line indicating onset of chuffing is exceeded by almost 3x.
A lot of that can be hidden inside Flare It and in real life by including allowances for seating position and masking by content (and further masking by the fact that it's happening inside the enclosure), but exceeding the lines in Flare It means distortion is present no matter how successfully you mask it.
Post 178 mentions a similar constricted design for the Lab 15 with a 150 sq cm constriction. That is going to be FAR worse than the one I've analyzed here wrt the velocity and chuffing issue.
I've been kinda following along on and off so I'm not sure if anyone has mentioned it yet, but it looks like a huge issue.
Using the Akabak script in post 167 (and assuming it's correct - which is hard for me to do without checking everything but I don't have time for that), I turned off diffraction and boundary reflections, set Input Voltage to 36 to achieve 10 mm xmax, and checked Velocity at both ends of the constriction, which are Duct (1) D8 and Duct (2) D8. I always do this when any part of an enclosure seems too narrow.
Without the hpf, velocity at Duct (2) D8 is right around 30 m/s, with the hpf it's still up around 26 m/s.
Next it's off to Flare It to see how bad this issue is. But first some math. Reading the script it looks like the constriction area is 12 x 1.5 = 18 inches. My conversion program says that is equal to 116.1288 sq. cm. Changing that to a circular area suggests a diameter of approximately 12.2 cm. The details are important so please correct me if there are any mistakes here.
Flare It results for a 12.2 cm diameter port with no end correction:
An externally hosted image should be here but it was not working when we last tested it.
And that's a big problem. The core velocity is fine but since the constriction is basically a port with sharp ends inside the main enclosure, chuffing at either end of the constriction appears to start at about 9 m/s at 35 hz. At xmax, with a velocity of 26 m/s, the blue line indicating onset of chuffing is exceeded by almost 3x.
A lot of that can be hidden inside Flare It and in real life by including allowances for seating position and masking by content (and further masking by the fact that it's happening inside the enclosure), but exceeding the lines in Flare It means distortion is present no matter how successfully you mask it.
Post 178 mentions a similar constricted design for the Lab 15 with a 150 sq cm constriction. That is going to be FAR worse than the one I've analyzed here wrt the velocity and chuffing issue.
I've heard chuffing issues before with my ported subwoofer and in practice I can't get this subwoofer to audibly chuff. Most likely that is because I'm in a room that has way more vibrations at that kind of volume than said chuffing, but still it is good to point out. Matthew what would it be like if I made the ports larger? I'm still going to build the second one identical to the first for consistency (both going to same person) but for future builds I'd like to get the width necessary to keep chuffing to a minimum straightened out. I'll likely do an outdoors test listening for this chuffing in a few weeks before I hand them off.
Speed of sound has nothing to do with velocity and chuffing issues.
MJK and a lot of others (including myself) like to limit velocity to 10 m/s or less as a general guideline. Some people say 17 m/s is ok, some are happy with a number a lot higher than that, some people never even bother to check on velocity. But these are all just numbers. Situations can be analyzed in much more depth.
Check out this website. Collo's DIY Subwoofer Enclosures
This is the home of the Flare It software too. But look all around and you will see that this guy did actual experimentation to find data and make the Flare It software.
Technically, since this port isn't round the software isn't going to be completely accurate but I use it as a common sense tool to properly size small ducts. When there are no sharp edges at the duct ends only the Core Velocity matters, but in cases like this with sharp edges at both ends you have to look at the Chuffing line too.
Even rounding over the duct ends would help a lot (use the software to see how much - and again you'll have to do some fudging because the port isn't round) but I don't think you can get within safe limits without a complete redesign.
As I mentioned a lot of the turbulence can be masked by listening postion (sitting a long way away), content (the music), and the fact that it's occurring inside the box. But even if it doesn't sound like farting it doesn't mean there isn't gross distortion present. There are several ways to test for this with or without fancy equipment. Start by playing the sub only (no mains) at xmax and then listen close to the mouth or measure for distortion and/or do low and high power response measurements and compare the response shape to see how port compression is affecting things at xmax.
Last edited:
Saba,
Since the magnet will be in the mouth you can actually make the mouth cm sq area larger than the pipe area (824cm sq or only about 7-5/8") ... The magnet and basket take up space there creating some additional mass loading and in order to compensate you can simply open up the mouth some more... So in the end it all works out because you can likely make the mouth opening 9 to 13" tall (maybe a little more) without shifting the FB of the box up too much above the target FB ..... An impedance measurement of the box can be used to determine where your fundamental ends up...... In the worst case scenario, if you have opened up the mouth as much as possible (don't want to detune the box by opening up the mouth any further) and still find it awkward to access the screws then you could just add a small access panel above the mouth..
Ok so make the back piece a little shorter until I can reach, got it. Most likely I will be able to finagle with it into submission.
Also, what are your thoughts on "what a guys" and his concerns for port velocity? I will have to run some power tests outside to see how much it will hinder low end output at high volume. I will be thoroughly disappointed if it is a big issue as I like this design otherwise.
What does it mean by flare radius? It seems that if I use a flare radius of 26 mm it turns out more or less to be right at the onset of chuffing, in other words -- much better.
Basically what I'm saying is what would it look like for me to modify the design to flare the port? I kind of have an idea of what that would look like but the shape of the mouth exit may make it tricky.
Basically what I'm saying is what would it look like for me to modify the design to flare the port? I kind of have an idea of what that would look like but the shape of the mouth exit may make it tricky.
Also I need to mention that these will be used in a club/portable DJ environment - distortion tolerance will not be an issue, however the ones I build for myself and or any that are used outside will definitely benefit from a larger constriction.
Could I use some flexible material in a flare formation used in the corners of each part in order to essentially make it one big hemi-cylindrical port? Not sure how to CAD that but I'm pretty sure you know what I'm getting at. Any way to salvage the design without making it hugely large?
Could I use some flexible material in a flare formation used in the corners of each part in order to essentially make it one big hemi-cylindrical port? Not sure how to CAD that but I'm pretty sure you know what I'm getting at. Any way to salvage the design without making it hugely large?
Last edited:
I would prefer not to. This isn't my script and all scriptwriters do things differently so messing with somebody else's script is asking for trouble unless you want to read through it line by line and completely understand everything they did. And I also prefer to do my own measurements to make sure everything was done correctly too. I don't have time for any of that.
I mentioned this so other people could check on it. The Akabak script is already done (or better yet you can do your own to make sure it's all perfect) and I gave you the line to the Flare It website and told you step by step how I came to my conclusions so you (or anyone else) can do exactly what I did.
Flare radius is described on the Flare It website. But here's an example in picture form.
Follow the link for more info. Round-over bits basics
In Flare It, a circular port is assumed and the flare radius goes all the way around the circular port entrance and exit. Since there's only one edge you can flare in this design (it's not a circular port) you can't get much flare in there unless you redesign a bit. All you can really do with this design is round over the entrance and exit of the constriction.
Port compression is always an issue ESPECIALLY at high excursion such as in a club or dj situation. Look up port compression and you will see that when it gets really bad you might as well just plug the port right up for all the good it does.
I'd really like to see the input hornresp params for this so I can play with increasing the port area.
First of all I don't know how bad the situation is, it needs to be measured before you start redesigning. Second, it's as simple as flaring the ends of the constriction. Like this.
An externally hosted image should be here but it was not working when we last tested it.
Same idea as a round flared port but not round. Incorporate a flare into the beginning and end of the constriction and if it's flared enough you won't have a problem.
You need akabak to check velocity in the middle of the design. The akabak script is already posted. Hornresp can't help you with this issue.
I would prefer not to. This isn't my script and all scriptwriters do things differently so messing with somebody else's script is asking for trouble unless you want to read through it line by line and completely understand everything they did. And I also prefer to do my own measurements to make sure everything was done correctly too. I don't have time for any of that.
I mentioned this so other people could check on it. The Akabak script is already done (or better yet you can do your own to make sure it's all perfect) and I gave you the line to the Flare It website and told you step by step how I came to my conclusions so you (or anyone else) can do exactly what I did.
Flare radius is described on the Flare It website. But here's an example in picture form.
Follow the link for more info. Round-over bits basics
In Flare It, a circular port is assumed and the flare radius goes all the way around the circular port entrance and exit. Since there's only one edge you can flare in this design (it's not a circular port) you can't get much flare in there unless you redesign a bit. All you can really do with this design is round over the entrance and exit of the constriction.
Port compression is always an issue ESPECIALLY at high excursion such as in a club or dj situation. Look up port compression and you will see that when it gets really bad you might as well just plug the port right up for all the good it does.
Not being a true ported design it will be interesting to see the level of compression it sees at high volume, I appreciate the advice, and eventually I'll be able to run akabak (ancient program requires me to get a virtual machine installed). Bigger port should take it down drastically but that will require extending the length, debating with self over how much bigger is tolerable.
This is merely for finding the combination of sizes to plug into akabak. I need to know how much longer the port needs to be to maintain the tune - given a larger cross sectional area.
It's not a helmholtz resonator if that's what you mean by "a true ported design" but look at the plans. It's clearly a port inside the box.
As I mentioned a few times the core velocity is fine so the port doesn't have to be bigger, it needs a flare (so technically it would have to be a bit longer to maintain the tuning but it doesn't have to get wider in the middle, just flared on the ends). You could make it bigger if you really wanted to but you don't have to.
That's not much of a flare, it's not going to do much of anything. It's better than nothing but it isn't much.
JustAGuy,
Your software that you are using , is it asking for "Pe" (watts input) or is it asking for something like "Acoustic Power" ?? It can be easy to mistake the difference ..
I am using other software to calculate port velocity and it is telling me that we are well below the DANGERZONE (yeah, thats an Archer reference
) even at full power ...Your port area calculations are good , ending up with a port area of around 120 cm sq (from the Akabak script for Saba's Alpine box) which is about 5 inches or almost 12-3/4 centimeters ....
I will post some screenshots so you can see what i did ..
You can open up the constriction a bit , but the port would have to be longer , taken too far and it would negate the design , you would essentially end up with a regular full length tapped pipe .....
Nevertheless, changes may not be necessary if there is no actual chuffing .... Software is telling me that we are well within safe limits for velocity .. . I will post some screenshots here in a moment ...
Last edited:
It finally clicked that I would basically turn the outer two port pieces into two piece segments making the flare and then gluing the pieces together and going from there, would need angled cuts and it would be tricky to clamp but would work.