• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

"To plug or not to plug ... this is here the question" (vented mains integrated with sealed sub(s))

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
Hi, I would need some help / technical advice for the above question as it has been bugging me for a while, but I simply cant get my head around it and the internet searches seem not to give me the clarity I am after.

Question:
Is it better to plug vented mains (e.g. KEF Ref 3) when combining (DSP Room EQ, x-over at e.g. 80Hz) them with sealed sub(s) (e.g. KF92) ?
Yes, I haven't defined "better", but any advantage / disadvantages would be great to hear in your responses. Audibility would be a separate, but of course relevant, discussion.

My current state of thinking is (but please correct / add to it):
For the vented mains I am thinking of e.g. the KEF Ref 3, the port only becomes relevant well below the sub cross-over frequency of 80Hz, so it might not matter whether one plugs the mains ports or not (?). I would guess the ports starts contributing at about 50Hz and lower or so, although I am not 100% sure, as I have not seen exact data, but the R7/R700, R5 ports are about in this ballpark too afik.

Thanks and I appreciate any answers / contributions.

Also I came across this one too (Source: Neumann FAQ, page 10ff):

1651416987523.png


 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,425
Likes
7,941
Location
Brussels, Belgium
What is your current bass management? I mean how flexible is your setup with regards to crossover slope and crossover frequency.
 
OP
H

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
What is your current bass management? I mean how flexible is your setup with regards to crossover slope and crossover frequency.
I use XT32 with the new MultiEQ X Microsoft app from Denon. I can set filters with slopes of 2nd, 3rd and I believe 4th order. But currently have the automated algorithm set everything.
I believe vented rolls off at 24db and sealed at 12dB/octave, but not sure if the automated XT32 takes care of that.
I also have REW, but only use it to measure FR at LP just fyi. And as far as I can tell it takes care nicely of the room induced nulls / peaks.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,425
Likes
7,941
Location
Brussels, Belgium
I use XT32 with the new MultiEQ X Microsoft app from Denon. I can set filters with slopes of 2nd, 3rd and I believe 4th order. But currently have the automated algorithm set everything.
I believe vented rolls off at 24db and sealed at 12dB/octave, but not sure if the automated XT32 takes care of that.
I also have REW, but only use it to measure FR at LP just fyi.
Well I had a look at some data from Erin's and Amir's review and I would cross the speaker at 70Hz when it's unplugged an 100Hz when it's plugged.

2nd order for the speaker, 4th order for the subwoofer.

plugging the speaker and doing 100Hz will increase theoretical output but it will (very slightly noticeably) pollute the soundstage.
 
OP
H

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
Well I had a look at some data from Erin's and Amir's review and I would cross the speaker at 70Hz when it's unplugged an 100Hz when it's plugged.

2nd order for the speaker, 4th order for the subwoofer.

plugging the speaker and doing 100Hz will increase theoretical output but it will (very slightly noticeably) pollute the soundstage.
Thx. Really appreciate it.

When plugged also 4th order for the speaker?

When plugged, would that reduce or increase the group delay as described in the Newmann source? Or would it not matter as the decrease of the delay would be canceled by the increase in delay of the higher order filter?

Would you mind, when you get the chance to explain your thinking please? Really curious to learn. Thx.

Also can I measure the effects with REW and if yes which measures would I have to conduct? FR? Phase? IR?
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,425
Likes
7,941
Location
Brussels, Belgium
When plugged also 4th order for the speaker?

you always want to use 2nd order for speaker and 4th order for subwoofer. plugged or not plugged.

When plugged, would that reduce or increase the group delay as described in the Newmann source? Or would it not matter as the decrease of the delay would be canceled by the increase in delay of the higher order filter?
Good Question. I didn't bother really testing this because I needed the crossover anyway.

However I don't really like bass-reflex (ported) bass. Whether it's the subwoofer or the speaker. it's likely to be group delay, but there is no point overthinking these things.

Would you mind, when you get the chance to explain your thinking please? Really curious to learn. Thx.

I'm trying to get the summation between the subwoofer and the speaker to ACOUSTICALLY sum to flat on-axis.

kh750_kh80_freq_resp_510.gif


This is often achieved with 80Hz Linkwitz riley acoustical crossover at 80Hz (4th order or 24db/oct).

One thing to keep in mind that a linkwitz - riley filter is achieved by stacking two Butterworth crossovers. So i checked the response of the speaker and found out at which frequency it resembles a Butterworth response.

1651430243907.png


Then i let the processor set a second order on top at the same frequency and a fourth order on the sub at the same frequency and voila!
 
OP
H

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
you always want to use 2nd order for speaker and 4th order for subwoofer. plugged or not plugged.


Good Question. I didn't bother really testing this because I needed the crossover anyway.

However I don't really like bass-reflex (ported) bass. Whether it's the subwoofer or the speaker. it's likely to be group delay, but there is no point overthinking these things.



I'm trying to get the summation between the subwoofer and the speaker to ACOUSTICALLY sum to flat on-axis.

kh750_kh80_freq_resp_510.gif


This is often achieved with 80Hz Linkwitz riley acoustical crossover at 80Hz (4th order or 24db/oct).

One thing to keep in mind that a linkwitz - riley filter is achieved by stacking two Butterworth crossovers. So i checked the response of the speaker and found out at which frequency it resembles a Butterworth response.

View attachment 203843

Then i let the processor set a second order on top at the same frequency and a fourth order on the sub at the same frequency and voila!
Mhhh… I wonder if XT32, Dirac, GLM, MA1 etc do this automatically if it is so commonly known among experts from this field or if I would have to “force” the 24dB/oct into it.

Thx. Really appreciate your help.

One more please. How did you determine the 70, 100Hz crossover frequencies you recommended?
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,425
Likes
7,941
Location
Brussels, Belgium
How did you determine the 70, 100Hz crossover frequencies you recommended?
Like i said, i use the REW EQ tool with the speaker driver feature in the target setting to check at which frequency the speaker's frequency response matches best a 2nd order butterworth filter.

when i stack a second butterworth 2nd order filter on top via the processor i get a Linkwitz-riley 4th order crossover.
 
OP
H

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
Like i said, i use the REW EQ tool with the speaker driver feature in the target setting to check at which frequency the speaker's frequency response matches best a 2nd order butterworth filter.

when i stack a second butterworth 2nd order filter on top via the processor i get a Linkwitz-riley 4th order crossover.
Ah got it.
 

FeddyLost

Addicted to Fun and Learning
Joined
May 24, 2020
Messages
752
Likes
542
Personally I see no reason to plug the port if you are already cut off LF from speaker with high order crossover.
As we can see from Erin's measurements, port still contributes at frequencies higher than 80 Hz, and if we plug it, distortion at high SPL will rise.
Port plugs is a kludge for case when signal processing is not available. Even if they are being sold with speakers.

Anyway, you can check sound with port plugs and without them and keep the best one. It will not damage anything.
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
My advice as well. The only valid reason to plug the port is if it emits resonances above the crossover frequency of the sub, which is the case for many speakers, unfortunately.
 
OP
H

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
Personally I see no reason to plug the port if you are already cut off LF from speaker with high order crossover.
As we can see from Erin's measurements, port still contributes at frequencies higher than 80 Hz, and if we plug it, distortion at high SPL will rise.
Port plugs is a kludge for case when signal processing is not available. Even if they are being sold with speakers.

Anyway, you can check sound with port plugs and without them and keep the best one. It will not damage anything.
Thank you very much. To both of you and LtIg.

BTW. Where did you find the KEFs Ref 3 detailed measurements at Erin’s or Pierre site. I only see the R3 but not the Ref 3. Am I looking at the wrong place? Thx.
 
Last edited:
OP
H

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
My advice as well. The only valid reason to plug the port is if it emits resonances above the crossover frequency of the sub, which is the case for many speakers, unfortunately.
Thx. Ports of the Kef Ref 3 are very fine and not making noise.
 
Last edited:

FeddyLost

Addicted to Fun and Learning
Joined
May 24, 2020
Messages
752
Likes
542
Where did you find the KEFs Ref 3 detailed measurements at Erin’s or Pierre site.
It was a mistake.
In case of Reference 3 with 2 exchangable ports my opinion is the same: with proper filtration of input signal blocking the ports is useless.
 
OP
H

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
It was a mistake.
In case of Reference 3 with 2 exchangable ports my opinion is the same: with proper filtration of input signal blocking the ports is useless.
No problem. Thanks for clarifying.
 
Top Bottom