Based on an off-topic discussion in the CHR-70 destruction thread, I have taken 5th Element's advice and played a bit with port tuning on my Castle microTowers.
Based on his calculations and some of my own playing around in WinISD, I decided to extend the ports from the stock 4" long to 9.4" long (24cm) in an attempt to tune the port somewhere in the 40hz range.
I accomplished this with 5.4" lengths of ABS pipe and some duct tape. The change was simple to make since my ported bottom panels are simple to remove.
The result? Well, I asked my wife and her response was "....mmm...deeper."
Indeed, deeper. Even with the left end of the EQ sliders down about 2 or 3 dB (which results is less escursion) the bass from 50-ish hz through to the mid 30s is quite obviously "better", smoother perhaps.
Subjective, I know, but if my wife can hear the difference then it is not subtle. I am pleased. I was happy with my microTowers as they were, but the low end is just that much better now...
Anyone else ever play with port tuning their microTowers? I'd love to hear about your setups and your results.
Based on his calculations and some of my own playing around in WinISD, I decided to extend the ports from the stock 4" long to 9.4" long (24cm) in an attempt to tune the port somewhere in the 40hz range.
I accomplished this with 5.4" lengths of ABS pipe and some duct tape. The change was simple to make since my ported bottom panels are simple to remove.
The result? Well, I asked my wife and her response was "....mmm...deeper."
Indeed, deeper. Even with the left end of the EQ sliders down about 2 or 3 dB (which results is less escursion) the bass from 50-ish hz through to the mid 30s is quite obviously "better", smoother perhaps.
Subjective, I know, but if my wife can hear the difference then it is not subtle. I am pleased. I was happy with my microTowers as they were, but the low end is just that much better now...
Anyone else ever play with port tuning their microTowers? I'd love to hear about your setups and your results.
Last edited:
As you know, I have really no idea what I am doing
I know TL is Transmission Line, but I have no idea what that means. Also, the Fs is 64Hz, according to Mark's pdf.
Is there a danger in doing what I have done? Does it make any sense that it sounds "deeper" and/or "smoother"?
By "a dip up higher" do you mean in the upper bass, or somewhere in the midrange, or...?
I have no idea what group delay is, but I'll try to google that...
I'm listening to Rage Against the Machine right now and I must say that the best way to describe the change that this longer port offers is "more balls."
I'm surprised that the micro tower actually shows any significant TL loading considering it looks in essence like a large ported cabinet. Sure the overall height of the thing will produce Quarter wave loading at something like 100hz, but then every floor standing ported speaker surely would too. Maybe they do and it's neglected in some way by design.
I have to admit though I've never looked into MLTLs before as I've never wanted to build one. A quick browse through a couple of papers seems to indicate that for a system to be a MLTL, the port has to be placed at the end of what would otherwise be the opening in a traditional transmission line. Then as the port changes the loading presented by the transmission line it pushes the tuning down in frequency.
Now from what I understand a TL doesn't really have a 'tuning' it more controls the air load presented to the driver and as a result controls the drivers behaviour. This occurs because the air in the line sets up standing waves at various frequencies that alter the pressure within the line and loads the driver so as to control it. Once you go below the lowest frequency the line can offer any control over though the driver excursion should shoot way up, much like a ported loudspeaker unloading.
Now the thing I am wondering is that the transmission line behaviour occurs because of the length of the pipe relative to the wavelength of sound going through it. So if the microtowers 1/4 wave is @ ~100hz how does the port go about altering this down to something ~40hz? Does the change in loading presented at the terminus of the line, by the port, set up a resonance that now occurs much lower in frequency where the port and the tower height actually end up working in sympathy? In a ported cabinet this ends up happening, but when it occurs, quarter waves aren't setup like they are in transmission lines. So does the 'mass loaded' presence of the port at the end of the line, actually allow quarter wave resonances @ ~40hz, to start occurring down the length of the line?
The tuning of the port in an MLTL must be quite critical I would imagine. If you keep altering the tuning is there a point where the cabinet actually stops behaving like an MLTL and transitions more towards behaving like a traditional ported cabinet? And does the Helmholtz resonance still occur in an MLTL or does the quarter wave loading act to suppress it?
Group delay is fairly simple in definition but probably a lot harder to actually know where to draw the line at. Group delay simply means how long you have to wait, after applying a stimulus to a system, for it to respond and produce something on the output. If this delay were identical for the entire system at all given frequencies then there wouldn't be a problem. The trouble is that in some systems the group delay differs depending on the frequency involved and this will/should degrade sound quality if large enough.
I have to admit though I've never looked into MLTLs before as I've never wanted to build one. A quick browse through a couple of papers seems to indicate that for a system to be a MLTL, the port has to be placed at the end of what would otherwise be the opening in a traditional transmission line. Then as the port changes the loading presented by the transmission line it pushes the tuning down in frequency.
Now from what I understand a TL doesn't really have a 'tuning' it more controls the air load presented to the driver and as a result controls the drivers behaviour. This occurs because the air in the line sets up standing waves at various frequencies that alter the pressure within the line and loads the driver so as to control it. Once you go below the lowest frequency the line can offer any control over though the driver excursion should shoot way up, much like a ported loudspeaker unloading.
Now the thing I am wondering is that the transmission line behaviour occurs because of the length of the pipe relative to the wavelength of sound going through it. So if the microtowers 1/4 wave is @ ~100hz how does the port go about altering this down to something ~40hz? Does the change in loading presented at the terminus of the line, by the port, set up a resonance that now occurs much lower in frequency where the port and the tower height actually end up working in sympathy? In a ported cabinet this ends up happening, but when it occurs, quarter waves aren't setup like they are in transmission lines. So does the 'mass loaded' presence of the port at the end of the line, actually allow quarter wave resonances @ ~40hz, to start occurring down the length of the line?
The tuning of the port in an MLTL must be quite critical I would imagine. If you keep altering the tuning is there a point where the cabinet actually stops behaving like an MLTL and transitions more towards behaving like a traditional ported cabinet? And does the Helmholtz resonance still occur in an MLTL or does the quarter wave loading act to suppress it?
Group delay is fairly simple in definition but probably a lot harder to actually know where to draw the line at. Group delay simply means how long you have to wait, after applying a stimulus to a system, for it to respond and produce something on the output. If this delay were identical for the entire system at all given frequencies then there wouldn't be a problem. The trouble is that in some systems the group delay differs depending on the frequency involved and this will/should degrade sound quality if large enough.
More response down low will translate loosely to "more balls", the ear perceives a 10 dB change around 1000 Hz as twice or half as loud, while down at 20 Hz it only takes a 5 dB change to sound twice a loud.OK, I understand what "group delay" is, but I have no idea what it might sound like. Perhaps "more balls"?
Generally, tuning below Fs is "less bang for the buck", the output goes deeper, but is not as loud. The lower the tuning below Fs, the more the speaker sounds/looks like a sealed cabinet.
You can easily find the cabinet tuning (Fb) by sweeping down with a sine wave tone generator, the cone movement (excursion) will be at minimum at Fb, it will go up rapidly below Fb, and will also rise above Fb.
If you don't mind putting a dot on the cone, it makes it easier to read excursion. If you are using very low test level, where cone movement is hard to see, you can use a piece of paper (or a finger) next to the cone, at Fb the paper won't be hit, but above and below Fb you will feel movement of the paper, or vibration of your finger.
Using this technique it will be easy to adjust the length of the port to get the Fb close to the speakers Fs.
If you don't have a sine wave generator, there are many test CDs and online free down loads of sine wave tones.
This will also get you familiar with the safe level of excursion vs. frequency your speakers can operate in, simulations often don't always relate all that well to the real world.
Have fun,
Art
A TL definitly has a tuning. Sometimes damped until close to aperiodic, this tuning is "smooshed" to use a highly technical term.
A vented box transitions from a BR towards am ML-TL as the dimensions shift such that one dimensions gets larger & larger compared to the others.
For good technical coverage one needs to read both Martin King works & Augspurger's AES paper (also published in aXp)
dave
Hmmm. Well, I am definitely getting more ballsy low end, and if there is less fidelity due to group delay, then either my choice of music does not reveal it or I simply cannot hear it. I suspect the latter, because my choice of music is very eclectic.
I think I might have lost a bit of oomph in the upper bass, but this was a region that was a bit pronounced before so I was compensating with EQ. With the longer port I have considerably more in the very low end and I no longer need to pull the upper bass down with EQ.
I am going to experiment with other port lengths, seeing as it is so easy for me to do so.
Did you read (and understand) my post #10 ?
I'll be curious as to what your actual Fb of choice turns out to be.
I assume I should disable my EQ for this type of analysis?
Since one driver is mounted in the top of the cabinet, it is quite easy to view the excursion...
Forgive my ignorance, but I still don't understand the benefit of matching Fb to Fs (if my goal is extending the bass response lower).
Yes, I can generate tones from the Linux command-line and I have some recorded sweeps I can use as well.
I appreciate all your input and I will do some testing as you suggested. I just hope to understand some of this theory as well.
Bypassing the EQ would eliminate a variable and simplify testing, an EQ that sounds good or flattens one tuning will not be appropriate for other tunings.
Extended frequency response also includes in room response, small rooms can give you "cabin gain" which can make up for a drooping bass response. Generally a good balance is Fb = Fs or a bit below. If you tune too low you may find that the box may as well be sealed. Finding the balance between a "flat" response, which will be louder, but will have a steeper rolloff below Fb, and a lower tuning, which will have less output above Fb, but a more gradual rolloff below is a subjective choice and compromise.
I'd suggest generating the tones in 5 Hz intervals from around 80 Hz to 50 Hz, then 1 Hz increments below, recording them for easy playback so you can increment /decrement with some speed, if you have to do keystrokes between tests it is hard to remember sound, feel, or sight of excursion from tone to tone.
I have been using test tones and visual cone confirmation of Fb since discovering a couple of my long duct tunings were off by as much as 1/2 octave from what was predicted
.Also, from a sonic standpoint, while testing it is a good idea to short out other speakers in the room, their Fb can cause dips in response like a bass trap.
Good vibes to you.
There is always the option of using LIMP as a part of the ARTA package to measure the impedance response of your loudspeakers. All you really need to make this work is the required connectors and a resistor. All I use is a pair of phono sockets for the sound card inputs + resistor and that's it.
Hi,
Detuning a vented box gives highly predictable results :
more low end
less upper bass
better transient response
In nearly all cases the best result depends on the room, not sims.
So experiment and choose what is best for the room and speaker.
I do it by foam lining the ports rather than extending them, to reduce
CSA for lower tuning and creating a velocity profile than can handle
higher total port velocities without chuffing, it works well.
rgds, sreten.
Detuning a vented box gives highly predictable results :
more low end
less upper bass
better transient response
In nearly all cases the best result depends on the room, not sims.
So experiment and choose what is best for the room and speaker.
I do it by foam lining the ports rather than extending them, to reduce
CSA for lower tuning and creating a velocity profile than can handle
higher total port velocities without chuffing, it works well.
rgds, sreten.
Well, rather than try to work with the suggested tools/methods to tweak something that I barely understand, I decided to simply listen to the system. A lot.
I have come to realize that the longer port is only beneficial for music that relies a lot on very low bass. Initially, I was specifically listening to various types of music, but I was selecting tracks which would "display" the low end bass response because that's what I was looking for.
However, I have done much more listening at this point, all of it random, and I have to say that in general the sound just wasn't as pleasant to listen to with the longer port. I cannot even put my finger on exactly what wasn't right about it, but perhaps I am now closer to understanding what the elusive "fidelity" is. It came to a head last night while listening to The Doors "Morrison Hotel". I have heard this album hundreds of times, and this time something just wasn't right. Within minutes I had the port extensions removed. Whatever was missing had returned. I continued to listen to random music and realized that there was just more "life" to everything...
I am glad I tried this experiment. Someday I hope to really understand what is going on here. I suspect there might be a port length somewhere between 4" and 9.5" that works best for me and my room, but for now I will enjoy the 4" tuning. I still have plenty of ABS left, so...
I have come to realize that the longer port is only beneficial for music that relies a lot on very low bass. Initially, I was specifically listening to various types of music, but I was selecting tracks which would "display" the low end bass response because that's what I was looking for.
However, I have done much more listening at this point, all of it random, and I have to say that in general the sound just wasn't as pleasant to listen to with the longer port. I cannot even put my finger on exactly what wasn't right about it, but perhaps I am now closer to understanding what the elusive "fidelity" is. It came to a head last night while listening to The Doors "Morrison Hotel". I have heard this album hundreds of times, and this time something just wasn't right. Within minutes I had the port extensions removed. Whatever was missing had returned. I continued to listen to random music and realized that there was just more "life" to everything...
I am glad I tried this experiment. Someday I hope to really understand what is going on here. I suspect there might be a port length somewhere between 4" and 9.5" that works best for me and my room, but for now I will enjoy the 4" tuning. I still have plenty of ABS left, so...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.