Can you specify "doing" because that is quite vague?
As for output, just measure outside the port.
Measuring inside a port opens up a whole bunch of other issues.
A woofer is always "contaminating" the measurements, since that is our source?
Obviously?
What's you point?
I think the point is getting results you can draw at least some conclusions from.
Sticking mics into ports can create certain issues that can make your findings totally invalid.
thanks @Tenson for your measurements!
any experience is very welcome here, even more useful if the measurement parameters are known and provided.
I also agree with @b_force that measuring inside the port the results have different meaning (but I would not call those measurements invalid) compared to measuring outside the port, which is how I did it.
I hope we all can profit from sharing and discussing experiences.
any experience is very welcome here, even more useful if the measurement parameters are known and provided.
I also agree with @b_force that measuring inside the port the results have different meaning (but I would not call those measurements invalid) compared to measuring outside the port, which is how I did it.
I hope we all can profit from sharing and discussing experiences.
Thanks for your kind words STV.
My measurements should only be compared with mine. Within that set you can see different port construction materials and the layer damped version has the highest damping of unwanted resonances. That being so, I think it could be an approach worth trying for anyone interested.
I expect in your first tests where the port tube is sticking a long way outside the speaker you will find little difference with the mic near the mouth or slightly inside. Perhaps you can try it and report back? In my case the woofer was right next to the port so the only way to get enough 'signal to noise' to see the port resonances clearly is to go slightly inside the mouth.
I also ran the mic along the whole port length to see where the hot spots for the resonances were. Unfortunately I can't find that data now but I do remember the lower freq. resonance was highest in the middle and the upper freq. resonance was at each end.
I think I misspoke earlier saying the port should be as stiff and light as possible. I think there is an optimal stiffness where the port doesn't take on new resonance modes because it is too weak, but also has enough flexibility to damp the higher freq. pipe resonances. Once you find that stiffness sweet spot, keeping it light will help the damping layer be most effective.
Here is the mic I used.
My measurements should only be compared with mine. Within that set you can see different port construction materials and the layer damped version has the highest damping of unwanted resonances. That being so, I think it could be an approach worth trying for anyone interested.
I expect in your first tests where the port tube is sticking a long way outside the speaker you will find little difference with the mic near the mouth or slightly inside. Perhaps you can try it and report back? In my case the woofer was right next to the port so the only way to get enough 'signal to noise' to see the port resonances clearly is to go slightly inside the mouth.
I also ran the mic along the whole port length to see where the hot spots for the resonances were. Unfortunately I can't find that data now but I do remember the lower freq. resonance was highest in the middle and the upper freq. resonance was at each end.
I think I misspoke earlier saying the port should be as stiff and light as possible. I think there is an optimal stiffness where the port doesn't take on new resonance modes because it is too weak, but also has enough flexibility to damp the higher freq. pipe resonances. Once you find that stiffness sweet spot, keeping it light will help the damping layer be most effective.
Here is the mic I used.
Last edited:
That doesn't make sense.
In an open tube/pipe, the maximum is at the end.
Otherwise we also wouldn't be able to hear that resonance.
In an open pipe this is the same for all harmonics.
The first harmonic has a minimum in the middle of the pipe. The 2nd order a max, 3rd a max again etc etc.
This is also one of the reasons why measuring inside a pipe isn't very helpful.
That doesn't have anything to do with just your method.
It has to do with the fact that you have to change the position of the microphone with every new pipe.
Depending on its acoustic length.
Misplacing the mic just by a bit can already give very different results.
Also, depending on the surface area of the tube, you're directly influencing the air flow and therefor the results.
Again, I am not talking about comparing to other people, but I am talking purely about your own results.
Which has the potential to make the results on themselves invalid all together.
Most importantly it's looking for unnecessary problems.
If that is the case then it seems reasonable using 25 mm MDF for a slot port or gluing one or multiple reinforcing ribs along its length should raise the mechanical resonance above the woofer's pass band. I have trouble visualising how a 3-side fixed piece of MDF small enough to serve as the slot partition of a 2.5 L cabinet could sustain a significant 400 Hz mechanical excitation.
When I talk about the lower resonance I do not mean the fundamental, but the first unwanted resonance around 400Hz.
Moving the mic a few cm does not affect the results much at all, I tested it.
You criticise a lot based on theory alone without practical experimental evidence. You said you purchased some parts for testing so I would love to see what you find.
Moving the mic a few cm does not affect the results much at all, I tested it.
You criticise a lot based on theory alone without practical experimental evidence. You said you purchased some parts for testing so I would love to see what you find.
I also agree the 400Hz resonance is not a 'pipe resonance'. There are clearly other factors at play in my tests. This is a slot port so I beleive it can be a membrane / limp mass panel resonance. Also I think it can be cabinet panel resonances leaking out the port. For this reason I think it makes a lot of sense to build a highly braced box for any port testing if you want to eliminate variables.
Regarding making the port stiffer, this will most likely reduce the panel type resonance but it enhances the pipe resonance. I think slot ports are good for getting more port area and length inside a small cabinet, but probably tube ports are better structurally in a suitable cabinet.
I'm actually still in the pipe fundamental resonance camp, which would explain why my measurements showed strong resonance suppression effects from perforating and damping the port. That said a tuned flexible wall sounds like it could achieve the same goal without introducing potential flow disturbances inside the port.
If you understood a bit of what I just wrote before, you maybe also understand that;
- What I was writing was a general statement
- I also mentioned the length or in other words, the ratio that is important.
Which means that for smaller pipes this is gonna be a lot more critical than longer pipes.
If you want to measure it the actual proper way, I would suggest reading on Kundt tube measurements;
https://en.wikipedia.org/wiki/Kundt's_tube
Also known as plane wave tube measurements.
You will see that microphones are always placed on the side part of the tube.
Which works well, since we are doing sound pressure level measurements.
In this case we also don't obstruct our airflow.
What also this information tells us, is that these in-pipe measurements are just needed to get information harmonics.
It can provide useful information when using the data below the first mode.
For example T/S can derived from this, but there are other applications as well.
You are very obviously missing the point here.
Nothing here is about how important people are.
But it does say something about people experience, skills and knowledge.
Again, no reason to get personal, pretty lame thing to do.
Come up with actual arguments.
Btw, people can contribute to a discussion just fine with always making their hands dirty.
In fact ,it shows that there is a bit of lack of fundamental basics.
And yes, eventually I will do some additional measurements myself, but once again, that is not the point here at all!
Perforating the port provides a pressure release which will both minimise pipe resonance as well as membrane panel resonance of the structure.
I encourage you to play a tone at the port fundamental and check what level you can achieve before unpleasant noises onset.
Below is the same port with a 60Hz tone. Red is a smooth wall / Green is with some 1mm holes in the middle covered on the back side with a bit of felt. P.S. This was measured in front of the speaker and port to test the system as a whole.
Last edited:
Arguments like:
1) I tried moving the mic, it made minimal difference to my data.
2) Measuring outside the mouth of the port contaminates the data with direct sound from the woofer.
3) The mic is 5mm it has minimal effect on airflow and is a constant in the data.
4) I only posted my data when requested and explained it was only intended for my personal use. I did not share it as a research paper.
In my view, you are the person being negative. Your attitude makes me regret sharing anything.
Than we share the same feeling. I find dismissing people extremely negative.
Makes me doubt even sharing all those years of experience and knowledge at all.
To me you come across of not giving a damn thing about people with actual experience an knowledge in the field.
Like if tomorrow Linkwitz of John Eargle would show up and you just totally dismiss them.
You probably don't mean it that way, but that is how it very clearly comes across at least to me.
In that case, let's just bury to war axes?
All the things you just pointed out could be true.
But for us, it's impossible to check if that is actually the case.
That could be totally unintentionally.
It's not about sharing as a research paper, it's about figuring out that even your own data can be meaningful.
All I am trying to explain, is that you're making it very difficult for yourself atm.
Leading to potential problems. The emphasis here is on the word potential.
Mostly only all because of point 2.
Also leaving some things out is also not clear, like the maximums and minimums we were talking about.
There ARE just things we expect from theory and they simply must be there in practice (within a certain margin of error).
So clarifying things afterwards is fine.
Without photos and (block)diagrams but only verbal explanation, it's hard to follow what's going on sometimes.
I tried this, trying to see of Kef's idea worked well. I did a version with TPU filament which was sort of stiff, and another using foam pipe insulation, which was reasonably flexible and had better results. The pic below is the TPU version so you get the idea. I also compared the foam version against a control. It sorta did something, but I wasn't too impressed. I thought maybe because the port is short there wasn't enough to flex but Kef's port is pretty short in the LS50 too, so? A more ideal flexible material might work better if anyone has ideas. Maybe a much lower hardness TPU, I think mine was around 96, and they make 90 or 92 also. Tougher to print that stuff though.
as far as I have measured and as I understand it when speaking about 2 side open tube resonance there will e a fundamental resonance with wavelenght roughly 2 * tube lenght. there will be a speed maximum at the ends and a pressure maximum at the center of the port.
an absorber for the fundamental resonance (not the helmholtz frequency, obviously) is best located at the center / pressure node of the port.
there is a second harmonic of the fundamental with wavelength = port lenght (double frequency). this resonance can be tamed with absorbers at 1/4 and/or 3/4 of the port length.
have a look at my first postings in this thread. I will also provide measurements shortly for practical ports, where my initial assumption of independent resonators has proved to be wrong.
My measurements were level calibrated and don't show that green behaviour:
https://www.diyaudio.com/community/threads/port-measurement.380629/post-6879909
Edit: check the impulse responses as well.
The perforations were simple ad hoc experiments using Home Depot drain liners, cut with a taper bit and filed clean internally, but the holes still had square edges so the limitations are recognized. The next step would be a printed port with tapered internal edges to minimize turbulence but I moved to sealed for now. That said I (casually) found the onset of turbulence using ~1 mm side wall ports to be quite high, approaching the driver's limits where it can be an arguable benefit to limit port flow and drive the enclosure towards sealed.