It is basis for the min allowed STR formula given in post #844, the formula is used to adjust the port size to match the min allowed STROuhal number.
I plan to do a quick research into strouhal number vs. port velocity comparison and I'll publish results here.
For the compression behaviour of ports you could have a look at my test here, including behaviour of PRs (scroll down for compression graphs).
During these test the turbulent noise started at about 1.5 dB of compression. Around this compression level also H3 distortion started to raise.
I would also think that it is not a coincidence that tapped horns and other quarter wave resonators are predominant in the high SPL PA range. Those levels require very big ports up to a point where the (theoretical) distinction between enclosure volume=spring and port=weigth gets questionable.
(It is also not entirelly correct for low spl boxes, but the model works well).
Thanks for your time stv! I'm sorry but the links do not work, #844 is easy to find but can't figure out the post of the second link.
edit. Found it!
Post #844 https://www.diyaudio.com/community/...rbers-and-port-geometries.388264/post-7935493
And the second link:
edit. Found it!
Post #844 https://www.diyaudio.com/community/...rbers-and-port-geometries.388264/post-7935493
And the second link:
Last edited:
Sorry about this! It maybe has to do with copying links and inserting on a mobile phone ... glad you found the posts anyway!
I would have to learn how to measure and interpret IMD results first ... 🙂
IMD of a narrow bandwidth resonator (such as the helmholtz one) would only refer to multi-tones very near to the resonance, correct?
@tmuikku
I used your parameters given in post #778 (the higher tuned hi spl variant) and made a comparison of an "optimized" geometry port using my tool and a straight, but rounded over port, using an updated version of the var. NFR calculation spreadsheet, observing the min allowed Strohual number as estimated according to formula in post #844.
This confirmed what I alreads suspected - for rather short ports the port length and volume occupation advantage of a flared port vanishes, at least using my rudimentary simulation tools:
Input parameters:
Vb = 250 liters
Fb = 33 Hz
SPL = 125 dB
Just for reference I added the corresponding approximated max air velocities.
The port volume occupation values are just rough estimations, without roundovers.
The comparison eventually changes with two instead of one port ...
I suppose for your speaker project it may be convenient to simply use straight tube ports, but including generous roundovers!
Schematic sketch of the variants given above:
I will post more comparisons of different cases next.
Generally smaller tuned low enclosures with high SPL requirements profit most from an optimized port, resulting in very much shorter and much more feasible ports.
I used your parameters given in post #778 (the higher tuned hi spl variant) and made a comparison of an "optimized" geometry port using my tool and a straight, but rounded over port, using an updated version of the var. NFR calculation spreadsheet, observing the min allowed Strohual number as estimated according to formula in post #844.
This confirmed what I alreads suspected - for rather short ports the port length and volume occupation advantage of a flared port vanishes, at least using my rudimentary simulation tools:
Input parameters:
Vb = 250 liters
Fb = 33 Hz
SPL = 125 dB
| Geometry ... | resulting from the "port optimizer" | using a nearly straight tube (NFR = 0,01) |
| Lact | 72,4 cm | 80 cm |
| Dext | 42,9 cm | 30 cm |
| Dmin | 28,6 cm | 29,6 cm, thus, still nearly 30 cm |
| Lact/Dext ratio | 1,69 (very low, because of big enclosure volume) | 2,67 ( higher, thus in theory better controlling the laminar flow) |
| roundover radius | 5,7 cm | 6 cm |
| volume occupied by port | appr. 65 000 cm3 | appr. 55 000 cm3 (smaller!) |
| Vmax (exit) | ~ 4,5 % of mach 1 | ~ 9 % of mach 1 |
| Vmax (mid) | ~ 10 % of mach 1 | ~ 9 % of mach 1 |
Just for reference I added the corresponding approximated max air velocities.
The port volume occupation values are just rough estimations, without roundovers.
The comparison eventually changes with two instead of one port ...
I suppose for your speaker project it may be convenient to simply use straight tube ports, but including generous roundovers!
Schematic sketch of the variants given above:
I will post more comparisons of different cases next.
Generally smaller tuned low enclosures with high SPL requirements profit most from an optimized port, resulting in very much shorter and much more feasible ports.
Last edited:
That's a good question.
Obviously it acts as a lowpass filter, but mostly for the sound going through.
Besides IMD can sometimes result in higher harmonics.
I am inclined to say that for just laminar flows it isn't that bad (very different story for sudden non-linear bursts)
But that's just based on feel and still doesn't give us any numbers. (aka vague)
Putting it all in the right context can be tricky, since variables will be added and change depending on the type of system (and drivers).
So to some extend it's also just learn to live with the disadvantages from a particular system.
Law of the preservation of misery 😉
For example, a PR is more predictable but will have a non-linear Cms/Kms (=Vas = Fs), while a BR-system seems to be a bit better in that regards, but is more difficult to acoustically predict beforehand.
But again all depends, in a massive box, all of this is less of an issue, while in a small cabinet moving things a little will change things drastically.
A bit bulky, but hopefully I can find a way to fit this in here without it being too close to sidewalls. I think I'm encountering the same concern that @tmuikku has regarding the ports ending up quite large compared to their straight-tube counterparts.
However the SPL target is pretty high for this size of enclosure, so lowering the target SPL from 108dB > 102dB would probably result in a much more reasonably sized port. I'll probably print one with the lower target SPL and see if chuffing is within an acceptable tolerance.
However the SPL target is pretty high for this size of enclosure, so lowering the target SPL from 108dB > 102dB would probably result in a much more reasonably sized port. I'll probably print one with the lower target SPL and see if chuffing is within an acceptable tolerance.
Dayton Pro Sound 15 subwoofer. Not been available for a while.
Factory specs I haven't measured it yet.
Still weighing my options. I'll make this. Right now I'm battling Arthritis and the end of winter which in the Ottawa Canada area is hard on a person with Arthritis. That and I have 4 cabinets that I'm veneering for a client. I still do custom work. Just not time sensitive custom work.
The flared ports have an advantage (or characteristic) that due to the gradual transitioning from high velocity/high loading middle port section to low velocity/lower loading at the port ends they are less influenced by wall proximity.
Simplified: the port already includes most (some) of the transition from moving air to air spring.
More present when you have a longer port length. This is well investigated in organ pipe investigations. On reasonable lengths the port resonances are there for sure, but they are much higher up. Those resonances are audible. The argument that they are out of the passband is incorrect. Those harmonics are produced as part of the functioning of the port while it is correctly functioning. One of the reasons why I never place a port on the front of an enclosure that I design.
What is the best practise to use the excel sheet? Do I need to calculate an optimal port in Vituixcad using the enclosure tool, then read the max spl from this graph and enter it in the excel?
If not, can someone guide me in how to do this?
I'm trying out what port I could use in a very small enclosure of a Q100 driver, aiming at an f3 of 80hz.
If not, can someone guide me in how to do this?
I'm trying out what port I could use in a very small enclosure of a Q100 driver, aiming at an f3 of 80hz.
I would do it exactly like that, yes.
Make sure you read the vent output SPL (101 dB in your case).
Also use net enclosure volume and tuning frequency as simulated with VituixCad (I am not sure if vituixcad provides enclosure vol with or without vent volume, so read that up in the manual).
Not sure if the vent will fit in such a small enclosure, thou. Maybe let it protrude outwards?
Thank you, good to know this is the way.
Unfortunately a port like this wouldn't fit in such small enclosure, it seems like. The port size in the Excel is 5 cm longer than the port from Vituixcad. Thus unfortunately I would be forced to use a bigger/longer enclosure. I was hoping a 17x17x17 enclsoure with a short port would fit 🙂
Edit: it seems, if I would accept a lower max SPL, a shorter port would still be viable. Is this wise, when aiming for a 80hz crossover to a subwoofer?
Last edited:
Depends on your SPL expectations, of course!
I'm not sure if you really can expect 105 dB around 80 Hz from that small 5" driver ... what xmax does it have?
You could also try to make the enclosure slightly bigger, that reduces port length.
I thought it was 4,2 mm based on these numbers, but looking at what Erin did using the poles, 3 mm seems more realistic. And then 100 db max at 80hz seems more reasonable too.
Last edited:
Standing waves of the port, I wasn't talking about those.
But yes, those are inherent physical part of a port (or two walls, it's basically all the same)
Even in that case you won't get the same harmonics from a smaller vs bigger pipe.
In organ pipes they are being used on a different way compared to a mass-loading device = BR-port.
Hence the name "mass-loading", it therefor MUST have a lowpass filter effect, since that's what mass does.
You'll also see and measure this when you use different sized ports with the same system tuning frequency.
Ports with a higher first mode will produce a standing-wave with less energy (not talking about speaker frequency here that has an effect on it)
Downside is obviously more air-flow resistance as well as more compression.
So there is a very clear trade-off here.
But to be fair and transparant, since a standing wave doesn't need much to go out of control, I guess practically it doesn't change much.
As long as you stay well below the first mode of a port you can place a port anywhere you want (from that point of view).
One of the big disadvantages of a 2-way system.
I was more talking about broadband "noise" (audio) here, not talking standing waves.
All you typed is very true. Things that are not often discussed on these forums. A situation, like this thread of investigating and finding solutions. Or investigating what has been done for decades and finding that there is a better way to do it!
Agreed. A to way is indeed a compromise in many ways. Sadly most three ways are not much of an improvement. In realistic, as is copying live performance dynamics, a 6.5 inch or even and 8 inch driver simply cannot move enough air to recreate the dynamics of what is possible in music.
But that is an entirely different topic.
On this topic. I found two cabinets that I thought I had turned into smoke! so, I have a test bed for my 15 inch driver. The central problem is the cabinet size and the length of port. SO this should be an interesting workout for this spreadsheet and the creation of a low noise port!
Add the fact that it also has to do low-end, with all the additional IMD, and it's just asking for trouble.
A 3-way should be already better at, quite significantly actually.
Don't even have to break the bank for that.
New big 3d printed optimizer port!
