Hi,
I'm designing a box to use with my JBL GTO 1014D (fs = 30 Hz, Qts = 0,44, VAS 30 L) and, as all small subs, I´m having a lot of trouble to tune it.
The ideal box volume is about 40L and Fb around 33 Hz. The problem is to avoid port turbulence, I need to use a 4" port, which is too long for this enclosure (about 60 cm), do it´s a quite hard to fit into the box even with curves.
Using a 3" port, the length problem is gone, but I get a ~ 60 m/s air speed peak at 31.5 Hz, 34 m/s at 40 Hz and 17 m/s at 48Hz. It's a 100% sure it will get noisy.
I was thinking about using a even lower Fb. I know it doesn't solve the problem, but the air speed peak will be located in a lower frequency were we don't have much musical information. I've simulated a Fb at 25 Hz and I got: ~ 60 m/s at 24Hz, 34 m/s at 31.5 Hz and 17 m/s at 39 Hz.
I know the port will get longer with the lower fb, but it still fitting the box and the higher air speeds still located at lower frequencies.
The other possibility is go to a sealed box. I get a 0,707 Qtc with a small 20L box. But I loose 9 db at 30 Hz and 5 db at 40 Hz. Since my bookshelf goes down to 55 Hz, I thing that the sealed subwoofer gets a little useless.
What you guys think? Where should I go without replacing the driver?
I'm designing a box to use with my JBL GTO 1014D (fs = 30 Hz, Qts = 0,44, VAS 30 L) and, as all small subs, I´m having a lot of trouble to tune it.
The ideal box volume is about 40L and Fb around 33 Hz. The problem is to avoid port turbulence, I need to use a 4" port, which is too long for this enclosure (about 60 cm), do it´s a quite hard to fit into the box even with curves.
Using a 3" port, the length problem is gone, but I get a ~ 60 m/s air speed peak at 31.5 Hz, 34 m/s at 40 Hz and 17 m/s at 48Hz. It's a 100% sure it will get noisy.
I was thinking about using a even lower Fb. I know it doesn't solve the problem, but the air speed peak will be located in a lower frequency were we don't have much musical information. I've simulated a Fb at 25 Hz and I got: ~ 60 m/s at 24Hz, 34 m/s at 31.5 Hz and 17 m/s at 39 Hz.
I know the port will get longer with the lower fb, but it still fitting the box and the higher air speeds still located at lower frequencies.
The other possibility is go to a sealed box. I get a 0,707 Qtc with a small 20L box. But I loose 9 db at 30 Hz and 5 db at 40 Hz. Since my bookshelf goes down to 55 Hz, I thing that the sealed subwoofer gets a little useless.
What you guys think? Where should I go without replacing the driver?
Do you already have the enclosure built? If not then perhaps you could make a slot port and just incorporate it into the design and build a larger enclosure.
I am having a similar issue with a current build of a small subwoofer using 5.25" drivers. I thought a 1.5" port would be sufficient and went about building without checking port velocity which is in the 30's. A 2" port would solve the problem, but wont fit in my current enclosure design.
Another option for each of us is making a bigger box which results in a shorter port length for the same tuning, but I have modeled that in WinISD and wasn't really able to get the results I was looking for.
Hopefully some others chime in with more helpful advice.
I am having a similar issue with a current build of a small subwoofer using 5.25" drivers. I thought a 1.5" port would be sufficient and went about building without checking port velocity which is in the 30's. A 2" port would solve the problem, but wont fit in my current enclosure design.
Another option for each of us is making a bigger box which results in a shorter port length for the same tuning, but I have modeled that in WinISD and wasn't really able to get the results I was looking for.
Hopefully some others chime in with more helpful advice.
Are the port speeds given at maximum power, or at xmax ? Sometimes the driver hits xmax before full power is applied.
Maybe consider a high pass filter close to tuning frequency ?
Adding a radius to the ends of the port can reduce noise. There used to be a program called 'flare it' which could help you.
Rob.
Maybe consider a high pass filter close to tuning frequency ?
Adding a radius to the ends of the port can reduce noise. There used to be a program called 'flare it' which could help you.
Rob.
The air velocities quoted by the simulation programs will only occur when you apply a full-power sine tone at that frequency.
You'll find that if you go to 1/2 voltage (1/4 power, -6dB), the port velocity will also halve.
Will you be constantly close to the limits of this subwoofer, or is it for background music in a small space?
Chris
You'll find that if you go to 1/2 voltage (1/4 power, -6dB), the port velocity will also halve.
Will you be constantly close to the limits of this subwoofer, or is it for background music in a small space?
Chris
Velocities up to 35m/s are fine in my experience.
It is all about the flare rate and type. Bigger flares are of course better. Aim for at least the same diameter circle on the flare as the inner diameter of your port. 2xID is safer again.
Reason:
You have to allow a thin boundary layer between the dynamic air mass and walls to allow a sensible air velocity distribution / core 'size'. The idea form would actually be elliptical not circular, profiling the air best to keep it attached. I did model it up in CFD a while ago, I may try to find some images.
It is all about the flare rate and type. Bigger flares are of course better. Aim for at least the same diameter circle on the flare as the inner diameter of your port. 2xID is safer again.
Reason:
You have to allow a thin boundary layer between the dynamic air mass and walls to allow a sensible air velocity distribution / core 'size'. The idea form would actually be elliptical not circular, profiling the air best to keep it attached. I did model it up in CFD a while ago, I may try to find some images.
Last edited:
http://www.profblairandassociates.com/pdfs/RET_Bellmouth_Sept.pdf
Different scenario and only comparable from a visual point of view, but look at the "core" size on pages 36/37 with different inlet geometries.
Also, didn't realise this article covered spit back! (exciting!). Look at the vortices on fig. 14b for the air structure created by reverse flow - as you'll get in reflex ports.
Different scenario and only comparable from a visual point of view, but look at the "core" size on pages 36/37 with different inlet geometries.
Also, didn't realise this article covered spit back! (exciting!). Look at the vortices on fig. 14b for the air structure created by reverse flow - as you'll get in reflex ports.
Last edited:
Building a small sub is hard. Small boxes doesn't fit long ports. Maybe that´s why ported subs from SVS are that big. I didn't make the enclosure. But, the problem with slot port is that it is hard do adjust and it has a high surface area (see that thin slot port in the Klipsch sub) which has a high power compression.
It's not suitable for me since i have a small room.
I do have a flared port here. I bought a 4" and 3" flared ports. But, as I said, the 4" port will bother me with it´s long length.
I'm sure it does operate at lower volumes than simulated ( I did at 350W RMS). It's a small room (<1000 ft³) with a even small listening area (~700 ft³).
35 m/s ?!
Some people say that we have to keep it under 0.05c! I already have a flared port here which is close to the elliptical one. But, as I said, I got speeds near to .20 which is unacceptable.Thats nice!
What do you think about lowering Fb to bring the port peak down to lower frequencies?
I thought it may be a nice solution...
I might be stating the obvious but 90mm (3.5") is available.
90mm PVC Pipe Solvent Weld and Fittings; Tee, Elbow, Socket, Joiner, Cap etc. | eBay
90mm PVC Pipe Solvent Weld and Fittings; Tee, Elbow, Socket, Joiner, Cap etc. | eBay
It will help sure, but 60m/s on its own still seems high.
How are you processing? An LR 12dB/oct HPF at 25Hz should bring that in check to 30m/s.
EDIT: winISD pro can simulate this, may be worth a fiddle
Not available in Brazil
I'm using the filters available in my Crown XLS amp which only have crossover, not eq..
I simulated with crossover. It lowered the speed to 40 m/s at 25 Hz. Pretty high, anyway. But I'm doing the simulation at 350W. It it reasonable? I mean, does the companies assume that? Seems odd to me assume that the sub will operate with a 25Hz sine at 350 W.
Whether the port noise will be objectionable is a matter of signal to noise ratio.
Using graphs from
Noise Generated in Air Ducts
Sound Power
we see worst case (no rounding) at 40 m/s a 3" duct may generate roughly 70dB SPL noise at 1m, dropping to perhaps 60 dB at the listening position.
Although that level of noise may be audible if you are playing only low frequency sine wave tones, the SPL of music peaks when the sub is running at levels high enough to generate high port velocities will be so much higher that it will mask that noise.
If not, you know what to do
.Attachments
Last edited:
I have run WinISD perhaps hundreds of times in various drivers & scenarios and I note those velocities also, modeling to get those speeds down...then it occurred to me SINCE WHEN DOES THE SPEED OF SOUND CHANGE?
There needs to be much more clarification as to just what it is we are measuring here.
---------------------------------------------------------------Rick......
There needs to be much more clarification as to just what it is we are measuring here.
---------------------------------------------------------------Rick......
Rick,
When the slug of air in the port "spills out", it makes noise, the higher the speed it spills at, the more noise created. The speed of sound remains the same.
If vent velocity turbulence noise is objectionable, and the cabinet volume required for large ports can't be afforded, passive radiators can be substituted, for additional cost...
Whether the port noise will be objectionable is a matter of signal to noise ratio.
Using graphs from
Noise Generated in Air Ducts
Sound Power
we see worst case (no rounding) at 40 m/s a 3" duct may generate roughly 70dB SPL noise at 1m, dropping to perhaps 60 dB at the listening position.
Although that level of noise may be audible if you are playing only low frequency sine wave tones, the SPL of music peaks when the sub is running at levels high enough to generate high port velocities will be so much higher that it will mask that noise.
If not, you know what to do
Wow, what a useful graphs!
I was running some live tests here, with a 33 L prototype and 3" port and I saw that is no port audible noise with all acceptable listening levels.
On the other hand, I liked the smoothness of the closed box (I've filled the port with a lot of socks
). Even when I tried to adjust the sound level (closed boxes are less efficient) to sound even, I found that the bass reflex sound more aggressive, I don't know how to say that precisely. When I played the box closed, it feels like there was a soft spring dumping the speaker cone. May be it's time to think about a closed box. It sound more smooth and it will make thing a lot easier.
The filter I mentioned is a crossover. Not sure what types you have on your amplifier but you can substitute in any HPF - so long as you adjust the frequency and check the excursion / response.
Good point on 'chuff' SWL being compared to clean port output & cone output!
When you get to that much power lower down, the wise thing to do would be to add some filtering to reduce excursion or have a box that allows it. Since there are many ways to implement a driver - it's up to the user to protect it and not set it on fire
Maybe build it with a 3" internal. I suspect this will be fine - if not then make an external 4" duct, with an extra brick or two chucked inside to compensate for the change in volume.
Good point on 'chuff' SWL being compared to clean port output & cone output!
When you get to that much power lower down, the wise thing to do would be to add some filtering to reduce excursion or have a box that allows it. Since there are many ways to implement a driver - it's up to the user to protect it and not set it on fire
Maybe build it with a 3" internal. I suspect this will be fine - if not then make an external 4" duct, with an extra brick or two chucked inside to compensate for the change in volume.
Wow, what a useful graphs!
I was running some live tests here, with a 33 L prototype and 3" port and I saw that is no port audible noise with all acceptable listening levels.
On the other hand, I liked the smoothness of the closed box (I've filled the port with a lot of socks
When I played the box closed, it feels like there was a soft spring dumping the speaker cone. May be it's time to think about a closed box. It sound more smooth and it will make thing a lot easier.
Sounds like you may have a peak in what you hear at the port tuning frequency, maybe due to room interaction. Do you have measuring equipment?
Unfortunately no! Just my PC mic.
The sub is tunned at 30 Hz right now (checked with REW and a impedance measurement jig) and sealed version sounded more smooth.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.