After having a good look at those tables/calculations I found no mention of the diameter of the column which I found strange. Surely the resonant frequency would be affected by that, one would think?
The other odd thing was it said that frequencies were the same in a conical or straight tube.
Again I would question this and say that surely the overall volume must be a factor in these calculations.
I have a small understanding of tuning pipes as I once played with windchimes. I tuned them to a minor scale so at times they would play what might sound like a small Bach tune.
I thought, coming from guitars that if I shortened a pipe by the distance of a fret space that the tube would resonate a semitone higher but in fact I discovered that twice a fret distance had to be cut off to make one semitone difference.
This appeared to be due to the way the pipe resonates so it's fundamental is changed around a central node where a string just works from changing one end.
I also found that for a pipe to vibrate freely it had to be held a small distance in from the end because there must be a node at that point on one of the higher harmonics whereas if held right at the end the sound is muffled or it simply won't ring at all.
There was also an ideal place for the striker to hit and different harmonics were produced at different striking places, which obviously were at various nodes.
So this seems to be relevant to a TL design.
Anyway Thanks Dave, I was kind of expecting you to say that so maybe I am 'getting it' a bit now. I will be doing your design first once I access a pair of passive B139 units.
Cheers
Dan
The other odd thing was it said that frequencies were the same in a conical or straight tube.
Again I would question this and say that surely the overall volume must be a factor in these calculations.
I have a small understanding of tuning pipes as I once played with windchimes. I tuned them to a minor scale so at times they would play what might sound like a small Bach tune.
I thought, coming from guitars that if I shortened a pipe by the distance of a fret space that the tube would resonate a semitone higher but in fact I discovered that twice a fret distance had to be cut off to make one semitone difference.
This appeared to be due to the way the pipe resonates so it's fundamental is changed around a central node where a string just works from changing one end.
I also found that for a pipe to vibrate freely it had to be held a small distance in from the end because there must be a node at that point on one of the higher harmonics whereas if held right at the end the sound is muffled or it simply won't ring at all.
There was also an ideal place for the striker to hit and different harmonics were produced at different striking places, which obviously were at various nodes.
So this seems to be relevant to a TL design.
Anyway Thanks Dave, I was kind of expecting you to say that so maybe I am 'getting it' a bit now. I will be doing your design first once I access a pair of passive B139 units.
Cheers
Dan
No, pipe, box volume [Vp, Vb] only sets its acoustic efficiency = peak SPL.
Correct, they are both 1/2 WL resonators.
??? A pipe's fundamental doesn't change unless you change its length and/or taper.
Right, take a dust mop and mark off the handle at 3rds, 5ths and holding it at the end, shake it at these points to see how it impacts the mop's flailing to see how much a pipe's harmonics is acoustically impacting the terminus/vent's output and if strong enough you'll find that the 2/5th impacts ['snaps'] it most, decaying with increasing distance.
Indeed! With rare exception, we ideally want the driver, vent at an odd harmonic.
I wasn't sure where to post for some advice, but it seemed fitting here, and I don't think I need a thread of my own at this stage (at least). 
Anyhow, I've been working slowly on/off with my simultaneous build of Amplifier and Speakers, lots of stuff to learn and mistakes to hopefully avoid (and money, as always).
But I'm at the stage that I need to measure the speakers and do the start setup of the DSP (Amplifier and DSP is 3E, 6-channel so fully active).
The question: How would you suggest to measure them?
I do live with tons of open space on the outside, but it's Always windy and I've got a stationary PC so the speakers would either be pretty much on the doorstep or I would have to arrange to move everything outside.
My largest room is where they'll live when all is said and done, should I perhaps just start there and take it from there, cross the bridge of moving things outside in worst case scenario...
Would you measure them in a specific order, all at once with something like rePhase (just found it, don't know much about it). As I'm not building a cross-over for some arbitrary situation but a known location and listening position I simply wondered if you would do it differently, I guess.
Please feel free to give suggestions, direct or about where I might find good useful information.
Anyhow, I've been working slowly on/off with my simultaneous build of Amplifier and Speakers, lots of stuff to learn and mistakes to hopefully avoid (and money, as always).
But I'm at the stage that I need to measure the speakers and do the start setup of the DSP (Amplifier and DSP is 3E, 6-channel so fully active).
The question: How would you suggest to measure them?
I do live with tons of open space on the outside, but it's Always windy and I've got a stationary PC so the speakers would either be pretty much on the doorstep or I would have to arrange to move everything outside.
My largest room is where they'll live when all is said and done, should I perhaps just start there and take it from there, cross the bridge of moving things outside in worst case scenario...
Would you measure them in a specific order, all at once with something like rePhase (just found it, don't know much about it). As I'm not building a cross-over for some arbitrary situation but a known location and listening position I simply wondered if you would do it differently, I guess.
Please feel free to give suggestions, direct or about where I might find good useful information.
Oh I see, yes I wonder if you would measure every driver by itself like when you design a passive crossover (at their height and in different angles), or if it's possible to measure all at the height of the tweeter and possibly even do adjustments like it's 'one speaker' instead of several drivers. I'm guessing that it's how the program works that writes ready FIR code for the DSP.
I have simulated them in VituixCAD so I have an idea of what to expect for crossover frequency and BSC, but that's still a fair bit from something exact and precise.
I'm really a knob at this, I've somehow through all the reading about designing speakers missed all about this. lol
How much free space is recommended? To figure out where in my yard I would end up, and how I could possibly arrange it.
I have simulated them in VituixCAD so I have an idea of what to expect for crossover frequency and BSC, but that's still a fair bit from something exact and precise.
I'm really a knob at this, I've somehow through all the reading about designing speakers missed all about this. lol
How much free space is recommended? To figure out where in my yard I would end up, and how I could possibly arrange it.
First a disclaimer, I have yet to do anything with DSP's so I'm coming from experience with passive and analogue active crossovers.
I would say it mostly depends on how deep you want to go with attaining desired acoustic slopes (of individual drivers), and how the software you are going to use to accomplish that works.
If you measure everything on one axis (say with the tweeter) then your measurements have relative offsets built in so you do not need to account for those in whatever simulation software you use.
If you measure everything on axis with each driver, you will need to put all of the offset information into whatever sim/design package you are using, but this allows for potentially more flexibility in checking things like off axis response variations.
If measuring inside, provided you have a large room, your main limiting factors for low frequency measurements will be your ceiling height, positioning the speaker roughly half way between the ceiling and floor, putting in roughly in the middle of the room, and having speaker / mic not parallel to any walls will likely give you the best results for semi anechoic measurements. The main advantage of moving outdoors is you can potentially get usable measurements down to a lower frequency. My own experience there (with not a particularly big outdoor space) was that I could get gated down to maybe around 300Hz, but that in general the measurements were good enough to use down to zip if I used 1/8th octave smoothing of the raw measurements.
Tony.
I would say it mostly depends on how deep you want to go with attaining desired acoustic slopes (of individual drivers), and how the software you are going to use to accomplish that works.
If you measure everything on one axis (say with the tweeter) then your measurements have relative offsets built in so you do not need to account for those in whatever simulation software you use.
If you measure everything on axis with each driver, you will need to put all of the offset information into whatever sim/design package you are using, but this allows for potentially more flexibility in checking things like off axis response variations.
If measuring inside, provided you have a large room, your main limiting factors for low frequency measurements will be your ceiling height, positioning the speaker roughly half way between the ceiling and floor, putting in roughly in the middle of the room, and having speaker / mic not parallel to any walls will likely give you the best results for semi anechoic measurements. The main advantage of moving outdoors is you can potentially get usable measurements down to a lower frequency. My own experience there (with not a particularly big outdoor space) was that I could get gated down to maybe around 300Hz, but that in general the measurements were good enough to use down to zip if I used 1/8th octave smoothing of the raw measurements.
Tony.
You will get the expected good results by taking all the same steps that you would for a passive crossover.. for some this involves both techniques, or even others.
Have you designed a crossover before? Some of the techniques you'd read about have specific reasons which are usually not made clear. FIR is not one I'd expect to begin with as a crossover should come together using conventional techniques as a sign you are doing things right.
Have you designed a crossover before? Some of the techniques you'd read about have specific reasons which are usually not made clear. FIR is not one I'd expect to begin with as a crossover should come together using conventional techniques as a sign you are doing things right.
Thank you both for your thoughts!
Maybe it's easier to go through my thought process during this.
Tweeter (SB 26ADC), Midrange (SB 15NBAC) was chosen for two reasons, people seem to like them, and I believe I made a passive crossover work in VituixCAD so if I decide to build a 2-way for surroundsound later I could use the same drivers.
Bass driver (Scan-Speak 26W/8534G00) was one out of 3 in my price range that looked good on paper/tests and should work well in closed enclosure. (I've got 2 channels left on the DSP for subwoofer if needed later).
Simulations in VituixCAD was primarily aimed at baffel design, enclosure volume and trying different crossover frequency based on public measurements of the drivers.
So I guess I could use at least one test to confirm each driver is like 'I was told', or else I would be guessing and possibly trying to correct 'system deficiency' when it's a driver that doesn't behave like I thought... I'm beginning to see the light now! lol
I would also need to adjust output level for each channel, but I'm not sure if that's best to do on-axis for each driver or at tweeter (ear) level...
Time alignment need to be checked, and I'm guessing I need to do that at ear level and one driver at the time so they don't impact each other (or possibly by using tone so far away from crossover frequency that it has no impact...)
At some point I'm not really adjusting crossover frequency, time alignment etc but EQ and BSC. I'm guessing that apart from EQ possible baffel problems (that could be done outside) I'm now adjusting to my room and listening position (which might change as I'm positioned the wrong way in my room for audio now by having the speakers off-set on the long wall), and even to my own preference possibly.
This is where I'm more at home, I started with car audio and adjusting by ear some 25 years ago and have helped others as well, microphone is a new experience but maybe I'll like it enough that I'll use it for cars as well.
*AllenB, I meant IIR, something went wrong in the typing.
Maybe it's easier to go through my thought process during this.
Tweeter (SB 26ADC), Midrange (SB 15NBAC) was chosen for two reasons, people seem to like them, and I believe I made a passive crossover work in VituixCAD so if I decide to build a 2-way for surroundsound later I could use the same drivers.
Bass driver (Scan-Speak 26W/8534G00) was one out of 3 in my price range that looked good on paper/tests and should work well in closed enclosure. (I've got 2 channels left on the DSP for subwoofer if needed later).
Simulations in VituixCAD was primarily aimed at baffel design, enclosure volume and trying different crossover frequency based on public measurements of the drivers.
So I guess I could use at least one test to confirm each driver is like 'I was told', or else I would be guessing and possibly trying to correct 'system deficiency' when it's a driver that doesn't behave like I thought... I'm beginning to see the light now! lol
I would also need to adjust output level for each channel, but I'm not sure if that's best to do on-axis for each driver or at tweeter (ear) level...
Time alignment need to be checked, and I'm guessing I need to do that at ear level and one driver at the time so they don't impact each other (or possibly by using tone so far away from crossover frequency that it has no impact...)
At some point I'm not really adjusting crossover frequency, time alignment etc but EQ and BSC. I'm guessing that apart from EQ possible baffel problems (that could be done outside) I'm now adjusting to my room and listening position (which might change as I'm positioned the wrong way in my room for audio now by having the speakers off-set on the long wall), and even to my own preference possibly.
This is where I'm more at home, I started with car audio and adjusting by ear some 25 years ago and have helped others as well, microphone is a new experience but maybe I'll like it enough that I'll use it for cars as well.
*AllenB, I meant IIR, something went wrong in the typing.
You mention time alignment, which suggests you are planning to take one measurement per driver, and need to know where to work from, is that right? Based on this it makes sense to put the mic where the difference in distance is the same as for the listening position.
[It is also preferrable to measure at the same angle to the driver, but that is only possible to also do at the listening distance which makes reflection contamination likely and pushes you outside. Fortunately it isn't so critical. This technique is based on many assumtions to begin with, such as what is the optimum toe in.]
So focus for now on the distance. It isn't about time aligning, it isn't about finding how much delay there is, it is about measuring what there is.. make sure you use a measurement technique that preserves the distance between measurements.
[It is also preferrable to measure at the same angle to the driver, but that is only possible to also do at the listening distance which makes reflection contamination likely and pushes you outside. Fortunately it isn't so critical. This technique is based on many assumtions to begin with, such as what is the optimum toe in.]
So focus for now on the distance. It isn't about time aligning, it isn't about finding how much delay there is, it is about measuring what there is.. make sure you use a measurement technique that preserves the distance between measurements.
That's really great advice, almost obvious when you say it, but I believe those are the things that's easiest to miss sometimes.
My three options for measuring is in my living room, outside the window of my living room (cables will reach there from PC, power jack, amplifier etc) but it will only be open space in front of the speaker (half space) and the back of the speaker will be closer to a wall compared to the inside, or I could move it out in the yard (but it will be difficult as by the time I've moved Everything outside the wind could have changed significantly as I'm literally living in the middle of a field).
It's been a dilemma in all of this to turn every coin over at least three times before making a decision. lol I'm not equipped for woodworking, my garage is set-up for working with cars (engines and metal work), so every step through this process have been with one eye on what tools I can afford or build. I spent more time on that then building/designing the speakers. An old circular saw that was outdated 20 years ago, a cheap router that I built a number of things for to make it do the work of all the tools that I don't have.
Even the design of the cabinets was based on what router bits I could get in Europe, and make work.
So when I found something I had to check in VituixCAD how those tools 'would make it sound', then do CAD drawings to see how it looked from an engineering point of view as well.
It's been a wild and interesting experience to say the least! lol
Now I just need to start tie it together, but I'm still very much limited by budget. For instance the speaker cables used for my current system will be used for the bass driver (no connections on the cabinet, cables directly from drivers to speak-on at amplifier to save cost), so once I cross that bridge I might be pretty committed to making it work.
Regarding time alignment, both baffle and sides are 5 degrees, so I'm guessing it's 'halfway there' already from the drivers point of view (I'm not sure if there's latency in the DSP depending on how much there is to compute).
And I know from experience that it takes time to listen and making adjustments until you're happy, it's not done in a day usually.
My three options for measuring is in my living room, outside the window of my living room (cables will reach there from PC, power jack, amplifier etc) but it will only be open space in front of the speaker (half space) and the back of the speaker will be closer to a wall compared to the inside, or I could move it out in the yard (but it will be difficult as by the time I've moved Everything outside the wind could have changed significantly as I'm literally living in the middle of a field).
It's been a dilemma in all of this to turn every coin over at least three times before making a decision. lol I'm not equipped for woodworking, my garage is set-up for working with cars (engines and metal work), so every step through this process have been with one eye on what tools I can afford or build. I spent more time on that then building/designing the speakers. An old circular saw that was outdated 20 years ago, a cheap router that I built a number of things for to make it do the work of all the tools that I don't have.
Even the design of the cabinets was based on what router bits I could get in Europe, and make work.
So when I found something I had to check in VituixCAD how those tools 'would make it sound', then do CAD drawings to see how it looked from an engineering point of view as well.
It's been a wild and interesting experience to say the least! lol
Now I just need to start tie it together, but I'm still very much limited by budget. For instance the speaker cables used for my current system will be used for the bass driver (no connections on the cabinet, cables directly from drivers to speak-on at amplifier to save cost), so once I cross that bridge I might be pretty committed to making it work.
Regarding time alignment, both baffle and sides are 5 degrees, so I'm guessing it's 'halfway there' already from the drivers point of view (I'm not sure if there's latency in the DSP depending on how much there is to compute).
And I know from experience that it takes time to listen and making adjustments until you're happy, it's not done in a day usually.
I'm not sure why I bother with time alignment, maybe because so many write about it.
I doubt much of what I've had has been good in that aspect. Maybe it's a different thing, but I never seem to like vented enclosures as much, but maybe that's respons more than the time delay.
The way I have my stereo setup right now is beyond bad, in any aspect, and it works okay for me. I'm 90% concerned about not getting tired of listening for a whole day, that frequently happens. Half the time I'm not even in the same room, and definitely not in a perfect spot in front of the speakers, and the material is often old live recordings that I have a thing for. So the only time I would want a precise and distinct system is when I watch a film.
I guess that's one of the reasons why I'm less concerned about perfect measurements and ruler flat respons, but one does not have to disclude the other I guess. Even my possibly odd taste could most likely be quantified by measurements. lol
I doubt much of what I've had has been good in that aspect. Maybe it's a different thing, but I never seem to like vented enclosures as much, but maybe that's respons more than the time delay.
The way I have my stereo setup right now is beyond bad, in any aspect, and it works okay for me. I'm 90% concerned about not getting tired of listening for a whole day, that frequently happens. Half the time I'm not even in the same room, and definitely not in a perfect spot in front of the speakers, and the material is often old live recordings that I have a thing for. So the only time I would want a precise and distinct system is when I watch a film.
I guess that's one of the reasons why I'm less concerned about perfect measurements and ruler flat respons, but one does not have to disclude the other I guess.
Even my possibly odd taste could most likely be quantified by measurements. lol
Yes, but you need to measure the right things. Be as precise as you like on the wrong things, it won't help.
So when you have the filters dividing things smoothly and phase coming together the way you want it, then time is good too. You'd see it if it were not. (I'll show an example next.)
The future would be making the speaker sound the same in and out of the room. It is considerably beyond these steps of designing a filter.
So when you have the filters dividing things smoothly and phase coming together the way you want it, then time is good too. You'd see it if it were not. (I'll show an example next.)
The future would be making the speaker sound the same in and out of the room. It is considerably beyond these steps of designing a filter.