I have some weird pipe found in my garage that's the right size. It's 36'' long and I put it in front of only one of the ports. Although I see a difference in sound it doesn't make things better. I know I need to make both ports longer and everything to check what really happens, but maybe information above helps somehow. How about exceeding the Xmax at around 60Hz when it goes louder? Will that disappear too?
Edit: I can't believe that this has become 5 pages long.... This project didn't seem to be so complicated at first
Edit: I can't believe that this has become 5 pages long.... This project didn't seem to be so complicated at first
While I know this is the common wisdom; there is a thread ( can't find it tho ) about tuning with ports of different lengths; something was said about averaging the lengths to get the effective length.
I am aware of one Australian maker ( Colin Whatmough ) who uses this in his subwoofers
I modeled your design in BassBox Pro. I had to assume you wired the two voice coils in series.
Two 3" diameter vents yield a port speed of just under 14 m/s. That is acceptable for vent speed, however, a 12" vent reaches back to 1" or less to the rear wall. That is your problem! If the rear wall has stuffing (as it should), you have even more unneeded port restriction that you are hearing.
The easiest way to deal with this is to add 90° elbows midway down the pipes and point them into the open space of the cabinet. Keep any stuffing away from the port openings.
To calculate the length of the port use the line that runs through the mid section of the elbow as part of the length. Parts Express sells elbows, but you might want to look at Schedule 40 PVC pipe.
Any other combination of vent changes (rectangular, more vents, larger vent diameters, etc.) require an even longer vent length, which reduces internal volume of the cabinet. Your cabinet volume could be larger if you wanted a little more lower end, so stealing internal volume with a different vent configuration works against your bottom end.
Lastly, you could flare the ends of the ports to further reduce port noise (after you add the elbow). If you flare both ends you need to change the port length to 13" each.
Your current cabinet should be tuned to about 32 Hz with an F3 of 36 Hz. You could get much better bass response with a larger enclosure (up to 8 cubic feet), but that is another story.
Last edited:
Converting the box to a sealed box will raise the point where xmax is exceeded from 26 Hz to about 37 Hz.
F3 raises from 36 Hz to 60 Hz in a sealed design.
The change in Fs seems a bit much but have not calculated. There is no plausible way I know of to predict response of a ported system in a room. Repeatedly have found a sealed box provides clearer more defined bass than the just rumbling sound of a small port system. Port results in lack of notes and sealed low frequencies have notes.
Roll off the the sealed box is -12dB per octave and the port type are -24dB per octave so EQ of a port is not to practical and EQ of a sealed box works really well. This also means at the very bottom of the audio spectrum this woofer in a sealed box will have more output than the ported one.
If all other things are equal - same driver, enclosure volume, power applied, etc., that simply is not true - just run the numbers through any sim software
The sealed enclosure (in blue) has substantialy less output at lower frequencies than the ported enclosure (in red)
The sealed enclosure will also exceed xmax at a higher freq than the ported enclosure
An externally hosted image should be here but it was not working when we last tested it.
Looking at the two plots I pick the vented enclosure.
Yes, you can EQ the sealed box to raise the output level to the same level, but you would quickly exceed xmax of the driver long, long before you would get there, so it is pointless to try.
To do what you suggest would require a different driver with a much longer xmax. Then you could EQ the lower end and make a superior system as you described, but you will need a huge amp. Every 3 dB of level change requires 2X the power Wattage.
An externally hosted image should be here but it was not working when we last tested it.
Problem with these curves...
That response curve may be right calculated at 1mW but at 10 watts it is clearly wrong because the ports in this system will go into power compression long before that power level is reached. So the nice low power graph does not apply at 10 watts input for the ported system. This means the sealed box at 10 watts will provide a lot more output level down low than the ported version. I estimate power compression sets in about 2 watts on this design effectively limiting port output at that level.
Second you make like the port and that is fine however, I prefer notes to rumbling sound produced by small ports that stilll do not work.
Third I see the peak output response are aligned for level with both curves. It is more realistic (in my opinion) to align the asymptomatic response and see where the bass falls rather than aligning the peak in the response unless the woofer is used as a sub only. If it is only a sub then other driver levels may be matched appropriately to the peakiness of the curve.
Forth, the difference in level at 30 Hz is nay but 6dB so EQ to achieve the same output as the port at 1mW is the same 6dB. Not to much EQ in the bass really and especially considering the port will never play that loud because of power compression at that frequency.
These things taken together leave me to choose a sealed box that
is inferior at 1mW than the ported box that is superior at 1mw. At 10 watts in the sealed system will be louder at all frequencies below 50 Hz than the ported system and without the useless rumbling of a port to small.
You may wish to discover the facts in the laboratory like I have and rely a little less on 1mW graphs for realized final results. There are a lot of troubles with many of the models so I suggest caution using them other than as a guide. Empirical results are the fact and the fact is SMALL PORTS STILL DO NOT WORK!
That response curve may be right calculated at 1mW but at 10 watts it is clearly wrong because the ports in this system will go into power compression long before that power level is reached. So the nice low power graph does not apply at 10 watts input for the ported system. This means the sealed box at 10 watts will provide a lot more output level down low than the ported version. I estimate power compression sets in about 2 watts on this design effectively limiting port output at that level.
Second you make like the port and that is fine however, I prefer notes to rumbling sound produced by small ports that stilll do not work.
Third I see the peak output response are aligned for level with both curves. It is more realistic (in my opinion) to align the asymptomatic response and see where the bass falls rather than aligning the peak in the response unless the woofer is used as a sub only. If it is only a sub then other driver levels may be matched appropriately to the peakiness of the curve.
Forth, the difference in level at 30 Hz is nay but 6dB so EQ to achieve the same output as the port at 1mW is the same 6dB. Not to much EQ in the bass really and especially considering the port will never play that loud because of power compression at that frequency.
These things taken together leave me to choose a sealed box that
is inferior at 1mW than the ported box that is superior at 1mw. At 10 watts in the sealed system will be louder at all frequencies below 50 Hz than the ported system and without the useless rumbling of a port to small.
You may wish to discover the facts in the laboratory like I have and rely a little less on 1mW graphs for realized final results. There are a lot of troubles with many of the models so I suggest caution using them other than as a guide. Empirical results are the fact and the fact is SMALL PORTS STILL DO NOT WORK!
I buy into your port non-linearity issue. For home use where volume levels are not at extreme levels you can get away with vent sizes that are closer to a 6" diameter vent for a 12" speaker. The original poster's twin 3" ports are less than ideal for the circumstance (4" would be better), but we have what we have.
I don't agree with the claim that 2 Watts of power is all you get out of the ports at input power levels greater than 2 Watts. Compression may set in, but what happens is the port begins to operate in a non-linear fashion, which is different than simply clipping of the port's power level. Maybe I misunderstood what what you were trying to imply.
However, and I reiterate my previous point, the 12" woofer currently employed is in no way capable to operate in a sealed enclosure (particularly at 2.44 cubic feet). Get another driver.
I don't think fouchagalaga wants to scrap all the work he has done and start over, so we should endeavor to make the best of what we have.
I don't agree with the claim that 2 Watts of power is all you get out of the ports at input power levels greater than 2 Watts. Compression may set in, but what happens is the port begins to operate in a non-linear fashion, which is different than simply clipping of the port's power level. Maybe I misunderstood what what you were trying to imply.
However, and I reiterate my previous point, the 12" woofer currently employed is in no way capable to operate in a sealed enclosure (particularly at 2.44 cubic feet). Get another driver.
I don't think fouchagalaga wants to scrap all the work he has done and start over, so we should endeavor to make the best of what we have.
You obviously have expressed what you believe. (btw the sim was calc'd at 100W with the amp & driver the OP has, not at 1 mW )
The point of this tread was to help the OP fix an issue that he is having but you seem determined to hijack this thread for your own purposes for whatever reason.
If you can't contribute helpful suggestions for the original purpose of this thread and within the limitations of what the OP has to work with, I would respectfully suggest do not contribute extraneous and potentially misleading content.
The point of this tread was to help the OP fix an issue that he is having but you seem determined to hijack this thread for your own purposes for whatever reason.
If you can't contribute helpful suggestions for the original purpose of this thread and within the limitations of what the OP has to work with, I would respectfully suggest do not contribute extraneous and potentially misleading content.
First, I think he has a valid point and he clearly has a strong opinion about it.
Second, I do think that the discussion has veered off-topic because the point is to try to make the situation better for fouchagalaga without ripping up everything.
However, in all fairness, you may be laying your criticism on a bit heavy.
Second, I do think that the discussion has veered off-topic because the point is to try to make the situation better for fouchagalaga without ripping up everything.
However, in all fairness, you may be laying your criticism on a bit heavy.
My best suggestions and thoughts
I originally suggested fouchag... listen to the system with the port blocked and with the port open. Everyone who has tried this within my limited experience, about 150 times, has discovered the bass is much clearer with notes well defined when the port(s) are blocked solidly. When open there is more low bass but the sound is always kind of rumbling and very non-distinct often with no discernible tone or note. The end result of this for many has been to slightly increase the bass level in active systems (port blocked) or in passive systems slightly turn down the mids and highs with a loss pad in the crossover. This gentleman has the choice to turn up his active bass so the plugged port will not be a problem to adjust for a little less bass to balance out the sound with the mids and highs. A lucky break.
I suppose of lot of the trouble is cause by the fact the port signal is one full cycle behind the cone signal at the tuned frequency. In this system it is 35Hz I believe so the time delay for the port signal is 29ms which is getting dangerously close to echo defined as 50ms delay.
The other problem is the power compression thing. I have measured a lot of ported systems, even ones I built, and the small port fairly much functionally clips. 10dB increase input increases output only 1dB at the port tuned frequency. The clipping effect is not to difficult to measure empirically which is where all my useful data comes from. This data led to the end of the small port.
Some decided to defend the small port but they have never put a microphone on a test port system. I would like that to occur. LEAP shows these small ports as a lot of port distortion and nonlinear BL action. So different models give different results. There's a surprise. In the end I use a microphone and on every speaker with a port it gets blocked before install.
And we never did mention all the problems from midrange tones inside the box exiting out the port hole/tube.
My other thought was to modify his amp slightly (schematic link was posted top MCM) and use a bass boost HP filter with a Q of 1-1.5 to EQ the now sealed box woofer. To me this would be the most satisfactory result. Exceeding Xmax and so on is usually not that big a problem until something hits a mechanical stop so boost is fine by my experience. And do not even get me started on woofer distortion please. Most of the woofers measured at 22volts are over 25%THD so a little exceeding the Xmax is not going to affect that much.
I have no opinion but I have a lot of test data beginning from the dark ages of college specializing in room acoustic and loudspeaker design in dept of physics. Have wasted to much of my life on the facts versus a lot of ill conceived models, horrible electronics designs, and bad box implementation. I have a lot of modeling softwares and way more test equipment. Now days things work pretty good overall. One of the models even gives accurate results for some things...go figure.
Really I just want people to make solid measurements and use these modeling softwares as guide only. The measured result in the room is reality and the rest is simply conjecture.
I originally suggested fouchag... listen to the system with the port blocked and with the port open. Everyone who has tried this within my limited experience, about 150 times, has discovered the bass is much clearer with notes well defined when the port(s) are blocked solidly. When open there is more low bass but the sound is always kind of rumbling and very non-distinct often with no discernible tone or note. The end result of this for many has been to slightly increase the bass level in active systems (port blocked) or in passive systems slightly turn down the mids and highs with a loss pad in the crossover. This gentleman has the choice to turn up his active bass so the plugged port will not be a problem to adjust for a little less bass to balance out the sound with the mids and highs. A lucky break.
I suppose of lot of the trouble is cause by the fact the port signal is one full cycle behind the cone signal at the tuned frequency. In this system it is 35Hz I believe so the time delay for the port signal is 29ms which is getting dangerously close to echo defined as 50ms delay.
The other problem is the power compression thing. I have measured a lot of ported systems, even ones I built, and the small port fairly much functionally clips. 10dB increase input increases output only 1dB at the port tuned frequency. The clipping effect is not to difficult to measure empirically which is where all my useful data comes from. This data led to the end of the small port.
Some decided to defend the small port but they have never put a microphone on a test port system. I would like that to occur. LEAP shows these small ports as a lot of port distortion and nonlinear BL action. So different models give different results. There's a surprise. In the end I use a microphone and on every speaker with a port it gets blocked before install.
And we never did mention all the problems from midrange tones inside the box exiting out the port hole/tube.
My other thought was to modify his amp slightly (schematic link was posted top MCM) and use a bass boost HP filter with a Q of 1-1.5 to EQ the now sealed box woofer. To me this would be the most satisfactory result. Exceeding Xmax and so on is usually not that big a problem until something hits a mechanical stop so boost is fine by my experience. And do not even get me started on woofer distortion please. Most of the woofers measured at 22volts are over 25%THD so a little exceeding the Xmax is not going to affect that much.
I have no opinion but I have a lot of test data beginning from the dark ages of college specializing in room acoustic and loudspeaker design in dept of physics. Have wasted to much of my life on the facts versus a lot of ill conceived models, horrible electronics designs, and bad box implementation. I have a lot of modeling softwares and way more test equipment. Now days things work pretty good overall. One of the models even gives accurate results for some things...go figure.
Really I just want people to make solid measurements and use these modeling softwares as guide only. The measured result in the room is reality and the rest is simply conjecture.
Good point
I did do that..
Sorry for not beiong active lately.. I have so much work to do!
I still have my 2x3'' ports with the elbow. I have the 3'' pipe I found in my garage that I will cut and seal with silicone or glue or whatever this week-end. If lowering the tuning frequency works, I will try to integrtae 25'' long ports to my box one way or another. If ever this does not work, I will either need to get an external HPF or play in the sub-amp (I most likely will not do that since I will blow everything up). I agree on the fact that small ports do not work, but for now I do not want to make bigger ports since I do not know if that will solve my problem, I will try extending the current ports. I did try blocking ports with a ball and it seemed as if the speaker reached Xmax as much as when ports not blocked. I even bought 3/8'' weather strip that I will put around handles and driver (even if driver already has a seal) I will put a little more eggcrate foam behind driver and make sure everything is perfect when the port extension will be attached. Once again thank you for the helpful replies!
Edit: and BTW Loren42 voicecoils are in Parallel.
Last edited:
One thing that you might consider. I believe this is intended to be a sub. As it has been beaten into us again and again, larger diameter ports would be helpful.
You could do one of two things. Increase the port diameters to 4". If you want, I could model you port lengths in Bass Box Pro. The more port pipe you add inside the box, the lower your internal volume, which impacts port tuning. It's a vicious circle. Normally, I would not get too excited about the small difference, but your internal volume is small at 2.44 cubic feet and port changes have a greater magnitude of change.
The second option is to go to a shelf type port. These are a little trickier because they are harder to tune, but offer much more port area and lower chuffing, particularly in a sub.
My suggested way of dealing with this is to build a shelf port external to your existing box. That is, make an external tunnel that goes under the existing cabinet, wraps around the back, and cut an opening into the cabinet at the desired port length. This makes the external cabinet larger, but keeps the internal volume constant. You can be a little creative with the design, too.
If you have a serious thought about this approach I can help model it in CAD if you give me your current cabinet dimensions.
You could do one of two things. Increase the port diameters to 4". If you want, I could model you port lengths in Bass Box Pro. The more port pipe you add inside the box, the lower your internal volume, which impacts port tuning. It's a vicious circle. Normally, I would not get too excited about the small difference, but your internal volume is small at 2.44 cubic feet and port changes have a greater magnitude of change.
The second option is to go to a shelf type port. These are a little trickier because they are harder to tune, but offer much more port area and lower chuffing, particularly in a sub.
My suggested way of dealing with this is to build a shelf port external to your existing box. That is, make an external tunnel that goes under the existing cabinet, wraps around the back, and cut an opening into the cabinet at the desired port length. This makes the external cabinet larger, but keeps the internal volume constant. You can be a little creative with the design, too.
If you have a serious thought about this approach I can help model it in CAD if you give me your current cabinet dimensions.
Last edited:
good luck!
Good luck with your studies. If you look back thorough the post from me you can get a free computer based audio oscillator from Marchand here: Function Generator, Software for PC and Hardware models With my sound card this gave 0.03%THD at full output. Just slowly adjust frequency watching the woofer cone move. I put a very small white piece of tape on the cone to make it easy to see the cone motion amplitude. As you cross the port tuning frequency cone motion amplitude will become minimum. The amplitude of motion will increase at either higher or lower frequencies than the port tuned frequency. Simple test with accurate result. Use a good light (not cheap fluorescent) like a halogen to see the tape. Do test at moderate power like around 5-10 watts.
Look forward to seeing your results.
As a note I do not care for foam. I use this item 2 here is not to expensive:
Soundproofing Materials, Soundproofing Products, Discount Soundproof Products
work really well to simply line the cabinet with this and not stuff. stuff has not proved to work to well with ports and such.
Loren42 has I think has good suggestions
Good luck with your studies. If you look back thorough the post from me you can get a free computer based audio oscillator from Marchand here: Function Generator, Software for PC and Hardware models With my sound card this gave 0.03%THD at full output. Just slowly adjust frequency watching the woofer cone move. I put a very small white piece of tape on the cone to make it easy to see the cone motion amplitude. As you cross the port tuning frequency cone motion amplitude will become minimum. The amplitude of motion will increase at either higher or lower frequencies than the port tuned frequency. Simple test with accurate result. Use a good light (not cheap fluorescent) like a halogen to see the tape. Do test at moderate power like around 5-10 watts.
Look forward to seeing your results.
As a note I do not care for foam. I use this item 2 here is not to expensive:
Soundproofing Materials, Soundproofing Products, Discount Soundproof Products
work really well to simply line the cabinet with this and not stuff. stuff has not proved to work to well with ports and such.
Loren42 has I think has good suggestions
Last edited:
Loren42, I really like your ideas. I don't fully understand what they are thought . I know that larger ports would be the best, but I don't want to increase my holes for ports yet. As i said before I have a feeling that lower tuning might not solve my problem completely (it most likely will, but I have a feeling) if lowering tuning works well, I will obviously put larger ports. The problem would be that they would be really long. That's where you can help . Can you explain what the shelf type port are like and what to expect when trying to make (or buy, I don't know what it is) one? And about the tunnel underneath the box, that would be an interesting solution. As I said, I don't fully understand what it involves. Maybe a drawing? I googled it but it isn't really conclusive..
. I know that larger ports would be the best, but I don't want to increase my holes for ports yet. As i said before I have a feeling that lower tuning might not solve my problem completely (it most likely will, but I have a feeling) if lowering tuning works well, I will obviously put larger ports. The problem would be that they would be really long. That's where you can help . Can you explain what the shelf type port are like and what to expect when trying to make (or buy, I don't know what it is) one? And about the tunnel underneath the box, that would be an interesting solution. As I said, I don't fully understand what it involves. Maybe a drawing? I googled it but it isn't really conclusive..This might sound a little retarded but... When doing this test, how do I measure cone displacement? What exactly am I looking for? Also, if I ever do it, what will the results mean? how can I interpret them?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.