Good question.
I am not, however, promoting a complete loudspeaker design here. The baffle used in the testing may or may not be the baffle for any given application.
The impulse response is useful to determine what might possibly be baffle and what is transducer response. In this case, baffle dimensions are not infinite and might even be considered small.
Before I just assume the need for baffle step correction, I would like to see the near field response of the enclosure and low frequency transducer and then compare it to a far field response. Would you not agree that enclosure design will have some importance to determining baffle size?
Baffle step correction destroys sensitivity. I would prefer a system that does not require it.
I am open to see what may come with the W5-704, used either with or without my modifications and two-way design.
Mark
I am not, however, promoting a complete loudspeaker design here. The baffle used in the testing may or may not be the baffle for any given application.
The impulse response is useful to determine what might possibly be baffle and what is transducer response. In this case, baffle dimensions are not infinite and might even be considered small.
Before I just assume the need for baffle step correction, I would like to see the near field response of the enclosure and low frequency transducer and then compare it to a far field response. Would you not agree that enclosure design will have some importance to determining baffle size?
Baffle step correction destroys sensitivity. I would prefer a system that does not require it.
I am open to see what may come with the W5-704, used either with or without my modifications and two-way design.
Mark
chuck55 said:
Each driver is different, modifications to the cone (and other parts) for one driver may show little or no benefit on another.
Mark:
I'm confused, how can they be time aligned without having the voice coils aligned vertically and without any sort of electronic delay?
How can the phase be so flat? If you look at the phase of the raw driver, don't the driver's raw response phase shift significantly throughout the frequency range?
For a long time I've been wanting a speaker with "just a cap on the tweeter" but I've always thought that I'll have to use some method to move the tweeter back so that the voice coils will be vertically aligned, such as using a wave guide, or at least some sort of offset baffle shape?
Every time I start prototyping speakers, I always first just throw a cap on the tweeter and try various diff. values.... Without your modifications the top end of the W5 wouldn't be nice enough for that approach to work very well, of course.
Would you discuss the narrow "beaming" directivity that we would expect to see using a 5" driver this high?
Oh! I see the SCALE that you use for the phase is far different from what we're used to seeing... THAT's why it looks so "flat". Sorry I didn't expect that and didn't look at the scale before I made my last post...
Mark,
Would it be any trouble to add a picture of the individual + summed responses of the drivers? Description of the baffle that you used for your test, and, possibly, .FRD and .ZMA files?
I don't intend to manufacture and market speakers on any large scale myself (except for a very unique design which remains an unlikely fantasy, but maybe someday, that would be an quasi-omnidirectional speaker that Ohm Acoustics would produce and sell) but, if some of us produce and sell / trade one or two sets of speakers using your stepped plug modification would you want some sort of license fee?
Mark,
Would it be any trouble to add a picture of the individual + summed responses of the drivers? Description of the baffle that you used for your test, and, possibly, .FRD and .ZMA files?
I don't intend to manufacture and market speakers on any large scale myself (except for a very unique design which remains an unlikely fantasy, but maybe someday, that would be an quasi-omnidirectional speaker that Ohm Acoustics would produce and sell) but, if some of us produce and sell / trade one or two sets of speakers using your stepped plug modification would you want some sort of license fee?
First, what application uses .FRD and .ZMA files? Is it speakerworkshop? If so, wouldn't .IMP files be better?
Second, my purpose with the phase response was to show the tweeter onset lead. This is just a little over 200 degrees. I also wanted to get this on the graph without shifting center point for the frequency response graph.
Next, voice coil location has little to do with apparent acoustic center. Somebody long ago just made this statement and it caught on. I hope that any designers out there will have stopped thinking about time alignment by location of voice coils.
Next, I missed Chuck's post when I made my last reply. The breakup or vibration mode pattern of each model transducer is unique. With a paper diaphragm, glue applications (including puzzle coat) are ineffective in controlling vibration modes. Also, for the transducer model specified, there is a clear mode at around 12 kHz. That mode may be caused by the way the plug is air mass loading the diaphragm. If this is the case (and I do not know if it is), then attempts to correct it by changing the cone may be ineffective.
Last, there should be no selling of loudspeakers made using my modification designs (including my new plug concepts). This is a diy forum and the information is supplied for the purpose of making loudspeakers for the use of the person making the loudspeaker.
Mark
Second, my purpose with the phase response was to show the tweeter onset lead. This is just a little over 200 degrees. I also wanted to get this on the graph without shifting center point for the frequency response graph.
Next, voice coil location has little to do with apparent acoustic center. Somebody long ago just made this statement and it caught on. I hope that any designers out there will have stopped thinking about time alignment by location of voice coils.
Next, I missed Chuck's post when I made my last reply. The breakup or vibration mode pattern of each model transducer is unique. With a paper diaphragm, glue applications (including puzzle coat) are ineffective in controlling vibration modes. Also, for the transducer model specified, there is a clear mode at around 12 kHz. That mode may be caused by the way the plug is air mass loading the diaphragm. If this is the case (and I do not know if it is), then attempts to correct it by changing the cone may be ineffective.
Last, there should be no selling of loudspeakers made using my modification designs (including my new plug concepts). This is a diy forum and the information is supplied for the purpose of making loudspeakers for the use of the person making the loudspeaker.
Mark
MarkMcK said:
Hi,
It is also reality and ignoring it destroys transient and phase response,
not to be mention it also sounds very bad, mid and treble forward.
If you have the juice a 0.5 way or bipole with two drivers is one
way around it, restoring the 6dB of sensitivity and doubling
the power handling / maximum levels.
FRD and ZMA files are the bread and butter of the FRD consortium.
http://www.pvconsultants.com/audio/frdgroup.htm
http://www.rjbaudio.com/Audiofiles/FRDtools.html
http://www.geocities.com/woove99/Spkrbldg/
/sreten.
MarkMcK said:
Sure, whatever is most convenient for you, I'm happy to use Speaker Workshop, I was just wanting to take the files and plug them into some speaker simulation software while I'm here at work - I have bits of free time throughout the day between calls and I am trying to learn how to use these various programs currently. I would also like to see if I can use my little Audax tweeters, like the one that I sent you some time ago (
Audax 3/4" Textile Dome Tweeter ), since I have at least a dozen of those on hand already. The other thing I want to do with the files is: figure out which woofer I want to use with the Tang Band serving as a sealed mid. The reason I had asked for .FRD and .ZMA was to use with the PCD 6.20 excel spreadsheet, but I've also installed Speaker Workshop here too, so that's fine.
I had just failed to notice your scaling and, foolishly, thought that you had pulled of something no less than magic, in creating a simple design with very little phase shift...
Well ~ sound does travel faster through an aluminum voice coil than through air, but, despite knowing that, I guess I thought the A.C. was further back than it is? Hopefully, before I would publish my own DIY design, I would do measurements to ascertain that the acoustic centers were aligned, either through physical placement or phase shift in the crossover.
Well... For instance, almost everyone making "DIY" speakers wants to keep exploring and trying new designs and new speakers, so the old ones are typically sold or given to friends/relatives, etc. The prices in these "sales" are usually not at all in line with retail prices, often without charging for labor, for instance, the point not being to make a profit so much as being able to afford to buy the next set of parts (without upsetting the wife)? By no means did I intend to start mass producing something with your glue and plug mods. If I was going to get into the business of selling speakers I would want to produce my own drivers, or, at least my own design outsourced.
Excuse me for my denseness, but how do you convert a .PS2 file to an .FRD or .ZMA?
I don't know what Christopher's point of reference is, but the phase response of my simple two-way design is excellent. More important than the phase response is the impulse response. That is the test that is important.
There are also negative consequences to equalizing to correct for change of space loading when the baffle ends. Transient response and "phase" response will become worse, particularly for onset performance. If the enclosure or eventual system design would benefit from equalization, then do so. It is not clear that BSC is the best technique or assumption for determining the correct amount, order, or shape of equalization.
Mark
I don't know what Christopher's point of reference is, but the phase response of my simple two-way design is excellent. More important than the phase response is the impulse response. That is the test that is important.
There are also negative consequences to equalizing to correct for change of space loading when the baffle ends. Transient response and "phase" response will become worse, particularly for onset performance. If the enclosure or eventual system design would benefit from equalization, then do so. It is not clear that BSC is the best technique or assumption for determining the correct amount, order, or shape of equalization.
Mark
Which program uses .PS2 file? Which program do you use to generate those charts? .SWD file would be fine also...MarkMcK said:
I didn't say its [phase response] is not excellent...
I just meant that I was confused for a bit while I thought that the phase only shifted a few degrees throughout the entire range because I failed to notice that your scale was much larger than typically used to display phase where you see the phase chart goes off the bottom then comes back down from the top again as it shifts past 180 degrees... (as in dlneubec's post above showing FR + Phase) As long as the phase of the woofer and tweeter match up at the crossover point, and don't veer away from each other two rapidly above and below, I'm happy with that. The fact that you use only one line (and not separate lines each) for the woofer and tweeter is interesting.
"my point of reference"
Hehe, my point of reference is: I don't like speakers anymore because I find them all unpleasant to listen to. I'm hoping to change that opinion soon. I used to be a real music lover.Ideally, I want a speaker that can fool us into thinking a live instrument is playing in front of us, if we close our eyes. Though, I don't expect to find that; I would be happy to have a speaker which simply doesn't give me a headache when I listen to it (without spending $20,000). I miss the el-cheapo 2-way paper woofer paper tweeter bookshelf speakers I had from 1977...
A reader has to be careful when interpreting supposed test results from parties you do not know. I have noticed an uncomfortable tendency for certain people to misuse features of given test suites to produce graphs more supportive of the story they are telling (usually about how good they are) than revealing anything about actual performance.
To a point not in this thread, I have seen supporters of horn loading present even impulse responses separately for tweeter and woofer and just overlay them. When you measure separately and then overlay or even mathematically combine just FR, you lose all the information that would tell you how the transducers are acoustically summing. That is a lot of information to lose. Usually when you acoustically sum a loudspeaker designed in this way, as you do when you listen to it, you have something that looks (and sounds) terrible.
When you see a frequency response graph that also includes separate FRs for woofer and tweeter, you should not just assume that the combined response is the acoustic response. It could be a math summing of the separate transducer response. If the tester would include the impulse response, you could tell if the FR and impulse match and if the impulse indicates a quality design.
As for your point of reference in listening . . . 1977 was pre digital. Digital recordings tend to be more fatiguing than good analog. That is one problem. With transiently coherent loudspeakers, like the ones I have been sharing here, I can listen at volume for hours to good digital recordings without fatigue.
As for the development (review) of any particular design using the 704, we still need to do more work on the enclosure and get some valid engineering measurements. Questions have been raised, suggestions have been made, but there is a lack of development data.
Mark
P.S. the file mentioned is the default file format for Praxis. I can provide a .pdx, the text based Praxis format. Can you convert that into something for use with one of the free amateur spreadsheet applications?
To a point not in this thread, I have seen supporters of horn loading present even impulse responses separately for tweeter and woofer and just overlay them. When you measure separately and then overlay or even mathematically combine just FR, you lose all the information that would tell you how the transducers are acoustically summing. That is a lot of information to lose. Usually when you acoustically sum a loudspeaker designed in this way, as you do when you listen to it, you have something that looks (and sounds) terrible.
When you see a frequency response graph that also includes separate FRs for woofer and tweeter, you should not just assume that the combined response is the acoustic response. It could be a math summing of the separate transducer response. If the tester would include the impulse response, you could tell if the FR and impulse match and if the impulse indicates a quality design.
As for your point of reference in listening . . . 1977 was pre digital. Digital recordings tend to be more fatiguing than good analog. That is one problem. With transiently coherent loudspeakers, like the ones I have been sharing here, I can listen at volume for hours to good digital recordings without fatigue.
As for the development (review) of any particular design using the 704, we still need to do more work on the enclosure and get some valid engineering measurements. Questions have been raised, suggestions have been made, but there is a lack of development data.
Mark
P.S. the file mentioned is the default file format for Praxis. I can provide a .pdx, the text based Praxis format. Can you convert that into something for use with one of the free amateur spreadsheet applications?
I'm gluing up a pair of vented ~ 20 L cabinets will tune to ~ 42 Hz. I'm not sure if I'm going to glue the fronts on or not... I don't have any of those Dayton tweeters yet, I was wanting to see how well my Audax ones would work, I suppose I should have just bought some anyway, they're so cheap. So I might want to have different front baffles w/different tweeters...
I think I will also build a pair in sealed cabinets.
Mark,
I've never gotten my measurement setup properly/calibrated, etc ~ if you like the idea, I could bring some of them to your place for testing?
I noticed that you didn't discuss the issue of the W5's dispersion at the higher end of it's passband in your design? Do you think you have the tweeter coming in high enough that it's at the edge of most people's hearing so the change in dispersion as you go from the woofer to the tweeter won't be a problem?
I think I will also build a pair in sealed cabinets.
Mark,
I've never gotten my measurement setup properly/calibrated, etc ~ if you like the idea, I could bring some of them to your place for testing?
I noticed that you didn't discuss the issue of the W5's dispersion at the higher end of it's passband in your design? Do you think you have the tweeter coming in high enough that it's at the edge of most people's hearing so the change in dispersion as you go from the woofer to the tweeter won't be a problem?
The discussion about the importance of controlling dispersion is an important discussion. It is also going to be a very involved discussion.
It is not much different than the discussion about baffle step correction. There is too much belief about either that is more myth than good theory.
I am going to take another look at the shared simulation file and then present a text file that will be as easy to convert as I can easily make it. That should be ready fairly soon.
As for using the small Audax tweeter . . . It you hold to the stock baffle diameter, then the woofer to tweeter center spacing will have to be larger than specified for the PE tweeter. The Audax tweeter is very sensitive to baffle size for performance. I have not tried it, but it might produce unexpected performance characteristics for a given application. Just something to look out for when using it.
Mark
It is not much different than the discussion about baffle step correction. There is too much belief about either that is more myth than good theory.
I am going to take another look at the shared simulation file and then present a text file that will be as easy to convert as I can easily make it. That should be ready fairly soon.
As for using the small Audax tweeter . . . It you hold to the stock baffle diameter, then the woofer to tweeter center spacing will have to be larger than specified for the PE tweeter. The Audax tweeter is very sensitive to baffle size for performance. I have not tried it, but it might produce unexpected performance characteristics for a given application. Just something to look out for when using it.
Mark
FWIW, here are the ML-TQWT graphs for my Duo hybrid omni design I posted earlier for the Tangband W5-704D version. The port air velocity is about 1.4% of the speed of sound. Stuffing density is 1 lb/cu.ft. The stuffing is placed in all of the back portion of the line, in the top below the driver and down the front portion of the line, ending about 18" above the bottom of the cabinet.
The port is 2.5 in2 and 2.5" long.
The first graph is the TL line profile:
The next graph is the predicted response:
the next two graphs are driver excursion at 1w and 10w input, though with the driver at about 6ohm nominal impedance, the input powers are more like 1.3w ad 13.3w:
Here is the measured fR for both drivers and the sum, as well as the measured reverse null and phase match.
The port is 2.5 in2 and 2.5" long.
The first graph is the TL line profile:
The next graph is the predicted response:
the next two graphs are driver excursion at 1w and 10w input, though with the driver at about 6ohm nominal impedance, the input powers are more like 1.3w ad 13.3w:
Here is the measured fR for both drivers and the sum, as well as the measured reverse null and phase match.
MarkMcK said:
Some people have trained themselves to listen for the "sweet spot" and pay a lot of attention to the changing levels of HF that the can hear when they move their head around, or stand up. One guy complained that a design that I wanted to do with the tweeter firing up might sound bad to someone at a party who was standing next to the speaker (duh...).
Reminds me of WMD (Weapons of Mass Deception), and, how false rumors change the public's opinion of someone EVEN IF THEY SEE THEM PROVED TO BE FALSE. You hear it enough times, even if someone shows you it's not true, you still believe the lie in some twisted way...
Surely Praxis has some way to export measurements to other Speaker CAD softwares
? I just downloaded the free version and the "Export" portion does seem to be lacking such basic functionality. Tried "Save As" ~ nothing there either... Yes, since before participating in this thread, I had intended to either A) when routing to flush mount the W5, cut as far into the Audax's flange as I can w/o damage; or, B) remove the front tweeter baffle/waveguide and cut my own somewhat deeper waveguide in the baffle (the front wooden board of the speaker enclosure) then route the countersink to flush mount the W5 right up to the edge of that waveguide.
I had also propsed using the ND20FB tweeter before you presented your design, when I was thinking about that, I was considering cutting into the edge of the W5(s) - I was actually thinking about doing an MTM (but psuedo "2.5" crossover) and truncating the flanges to place them closer with the hole drilled for the ND20 offset some, but also cutting into the W5's flanges...
I would have listened to "just a cap on the tweeter", but I was planing on going w/ a 1st order series crossover. That was before you shared your W5 mods - there are very few woofers that roll off nicely enough on the top end to facilitate such a simple crossover (just one cap).
Well - I'll have to learn the hard way (building prototypes) because I'm not familiar enough with software used to simulate that.
------------------------------------------------------------------------------------
[B][COLOR=green]Zaph[/COLOR] [/B] said:
My post was aimed at sreten, not you, Zaph ~ I'm sorry you took offense to it anyway.
Now that Mark has published his modifications, seems like you have a lot more reason to keep those woofers and do something with them, not to put words in his mouth, but, seems like Mark's claiming that with his mods, this woofer will be better than a $200 ScanSpeak.
MarkMcK said:
Hi,
Whilst I agree with your second point that applying a blind 6dB of BSC is not the most informed way of doing things, I would also say
your first point does not seemed the most informed. In a small box
the "onset" will rise ~ 6dB between 150 Hz and 1.5khz far field :
There is nothing in the hemispherically loaded response that
suggests BSC will not be needed. Which goes back to the
original question ? the responses shown are flat farfield
which leads to the assumption of a large test baffle.
I can see why people prefer to ignore BSC, "throwing" away 6dB
seems like sacrilege, but it is sticking your head in the sand IMO.
/sreten.Just a little in reference to the idea of dispersion and beaming . . . There are all sorts of beliefs held about and judgments made about what is required for quality sound reproduction. There are also historical examples of loudspeaker designs that embodied and exemplified any given problem.
I attend art events with DJs and sound reinforcement loudspeakers. The sound at these events just makes me want to leave. The sound changes as you move around, yet there is no place to escape the fatigue inducing sound. But I do not go to these events to listen to recorded music.
Having a loudspeaker that sounds different at different listening locations can be but is not necessarily a problem. Indeed, if you use as your limit what the ear can hear, then all loudspeakers will sound different as the listener moves. This, however, is something different than the classical example of beaming. Beaming should indicate a narrow listening region where a particular aspect of the sound becomes much more prominent. This was often the case of early dome tweeters. For a while, typical dome tweeters would have a diaphragm material vibration mode centered around 12 kHz, with very narrow dispersion. As with most vibration modes, the level of this 12 kHz beam was considerable greater than the average level. This tweeter beaming would be part of the response even when crossed at a low frequency or coupled with an Ohm style designed woofer in what should be a wide dispersion design.
Beaming can exist in a wide dispersion design and it can not exist in a narrow dispersion design. The first step is to eliminate beaming.
Mark
I attend art events with DJs and sound reinforcement loudspeakers. The sound at these events just makes me want to leave. The sound changes as you move around, yet there is no place to escape the fatigue inducing sound. But I do not go to these events to listen to recorded music.
Having a loudspeaker that sounds different at different listening locations can be but is not necessarily a problem. Indeed, if you use as your limit what the ear can hear, then all loudspeakers will sound different as the listener moves. This, however, is something different than the classical example of beaming. Beaming should indicate a narrow listening region where a particular aspect of the sound becomes much more prominent. This was often the case of early dome tweeters. For a while, typical dome tweeters would have a diaphragm material vibration mode centered around 12 kHz, with very narrow dispersion. As with most vibration modes, the level of this 12 kHz beam was considerable greater than the average level. This tweeter beaming would be part of the response even when crossed at a low frequency or coupled with an Ohm style designed woofer in what should be a wide dispersion design.
Beaming can exist in a wide dispersion design and it can not exist in a narrow dispersion design. The first step is to eliminate beaming.
Mark
sreten said:
"It is not clear that BSC is the best technique or assumption for determining the correct amount, order, or shape of equalization."
I think that's Mark's key point in regards to the BSC discussion; saying Mark does not seem "informed" is amusing, but probably not accurate.
I also think that using resistors is not a preferable way to address the baffle step issue. My choices would be: enclosure design and driver(s) selection, and, lastly, in the crossover design (but not with resistors).
Once in a while I consider taking the time to build some enclosure molds for some egg shaped enclosures...
critofur said:
Hi,
I agreed with the second point. I disagreed with the above.
Its simply wrong, for the reasons stated. I do not really care
whether people "dislike" BSC in principle or not, but stating
myths exist about BSC and then effectively stating a myth
about BSC does seem to me an "uniformed" opinion.
TBH I think Marks bag is improving drivers, how they are used
is a different issue. Personally I think the best approach to
BSC issues is at line level in the preamplifier, or an
equaliser between pre-amp and power amplifier.
/sreten.Please people, there are rules to valuable argumentation. The idea is to reach a more informed perspective. When you allow participants to start arguing against the person instead of against their arguments you have failed.
Baffle step exists. Baffle step correction does not exist. What is termed baffle step correction is simply equalization. It does not matter whether you equalize speaker level or line level.
The use of the term "field" is problematic. It is used, for example, to denote types of measurement. You can measure either near or far (or anything in between). It also seems to have generalized to mean other things.
We can talk about the direct and reverberant or ambient sound. Direct sound is not effected by baffle step. Ambient sound is. Ambient sound is also decay sound and not onset. Equalization effects both. If you equalize for the ambient sound and the direct sound does not need it, then you make inaccurate the direct sound.
Also, if you have chosen a ported design (this includes the new age tuned TL designs) and you provide low frequency equalization, you also upset the tuning. I would advise including expected equalization in the tuning for the vented loudspeaker. It seems to make sense to me, but I have never seen any online contributions where this is done.
I will not label people as "idiots" and other derogatory terms, but I also know that it is not productive to argue with people who are dogmatic in their beliefs.
I would like to try something new. Can we talk about the benefits and disadvantages of designing a loudspeaker that will minimize baffle step? What about in-wall mounting or shallow enclosures mounted on the wall?
We can also add a bass module. The two-way design I presented using the modified woofer is of sufficient quality to justify the added expense of bass augmentation despite the inexpensive cost of the two-way design. This will also decrease low frequency distortion.
My reason for returning for one last time to Tang Band transducers was because the W5-704 comes close to being wide band and not needing bass augmentation. It can certainly benefit from it, but compared to four-inch full range applications, it is listenable without.
Mark
Baffle step exists. Baffle step correction does not exist. What is termed baffle step correction is simply equalization. It does not matter whether you equalize speaker level or line level.
The use of the term "field" is problematic. It is used, for example, to denote types of measurement. You can measure either near or far (or anything in between). It also seems to have generalized to mean other things.
We can talk about the direct and reverberant or ambient sound. Direct sound is not effected by baffle step. Ambient sound is. Ambient sound is also decay sound and not onset. Equalization effects both. If you equalize for the ambient sound and the direct sound does not need it, then you make inaccurate the direct sound.
Also, if you have chosen a ported design (this includes the new age tuned TL designs) and you provide low frequency equalization, you also upset the tuning. I would advise including expected equalization in the tuning for the vented loudspeaker. It seems to make sense to me, but I have never seen any online contributions where this is done.
I will not label people as "idiots" and other derogatory terms, but I also know that it is not productive to argue with people who are dogmatic in their beliefs.
I would like to try something new. Can we talk about the benefits and disadvantages of designing a loudspeaker that will minimize baffle step? What about in-wall mounting or shallow enclosures mounted on the wall?
We can also add a bass module. The two-way design I presented using the modified woofer is of sufficient quality to justify the added expense of bass augmentation despite the inexpensive cost of the two-way design. This will also decrease low frequency distortion.
My reason for returning for one last time to Tang Band transducers was because the W5-704 comes close to being wide band and not needing bass augmentation. It can certainly benefit from it, but compared to four-inch full range applications, it is listenable without.
Mark
Hi,
My only point was you should not simply ignore BSC if you are
after some sense of ultimate accuracy. It should be factored
into the design like the other variables and can lead to design
choices (e.g. a wide baffle) that might otherwise not be taken.
FWIW I have no interest in simply disagreeing with someone on a
personal level, I do not have the opinion everyone has a right to
hold an opinion even if it is wrong, technically speaking with the
correct terminology there is usually one way of looking at it correctly.
Your other assertions do not necessarily follow :
You say direct sound is not affected by baffle step, this is only true
nearfield and does not apply to typical farfield listening positions,
consequently applying BSC is not fundamentally flawed.
This seems to be a fundamental difference in the way things are
described. What I disagreed with previously also implies that BS
does not affect the direct sound. As far as I'm concerned if the
direct sound in the first wavefront (and everything else for
~10ms in room) it is undisputed that it is affected by BS,
as anechoic or freefield measurements show.
(See Seas site for drivers measured in boxes and the BS)
You say BSC does not exist but BS does. BSC is the part of the EQ
due to BS and I cannot see any point describing it any other way.
(C= compensation not correction, BS cannot be "reversed")
I do not understand your assertion EQ upsets vented and TL tuning.
Typical narrow baffle BSC affects ~ 150Hz to ~ 1.5KHz and has no
effect below say ~ 100Hz. Bass alignments are unaffected.
I've got no axe to grind. If one does not want BSC in a speaker,
the direct drive and simple crossover appeals, then fine but there
are inevitable consequences to that choice to be considered IMO.
As an example the "Ariel" design chooses to ignore BSC fairly early
on as the ~ 92dB sensitivity is part of its fundamental specifications.
Many builders seem to be very happy with it, though nowhere is
it pointed out to the prospective builder BS has been ignored.
My opinion is that it should pointed out and left to the prospective
builder as to whether they accept that design choice or not.
At the least point them in the direction of :
http://sound.westhost.com/bafflestep.htm
/sreten.
My only point was you should not simply ignore BSC if you are
after some sense of ultimate accuracy. It should be factored
into the design like the other variables and can lead to design
choices (e.g. a wide baffle) that might otherwise not be taken.
FWIW I have no interest in simply disagreeing with someone on a
personal level, I do not have the opinion everyone has a right to
hold an opinion even if it is wrong, technically speaking with the
correct terminology there is usually one way of looking at it correctly.
Your other assertions do not necessarily follow :
You say direct sound is not affected by baffle step, this is only true
nearfield and does not apply to typical farfield listening positions,
consequently applying BSC is not fundamentally flawed.
This seems to be a fundamental difference in the way things are
described. What I disagreed with previously also implies that BS
does not affect the direct sound. As far as I'm concerned if the
direct sound in the first wavefront (and everything else for
~10ms in room) it is undisputed that it is affected by BS,
as anechoic or freefield measurements show.
(See Seas site for drivers measured in boxes and the BS)
You say BSC does not exist but BS does. BSC is the part of the EQ
due to BS and I cannot see any point describing it any other way.
(C= compensation not correction, BS cannot be "reversed")
I do not understand your assertion EQ upsets vented and TL tuning.
Typical narrow baffle BSC affects ~ 150Hz to ~ 1.5KHz and has no
effect below say ~ 100Hz. Bass alignments are unaffected.
I've got no axe to grind. If one does not want BSC in a speaker,
the direct drive and simple crossover appeals, then fine but there
are inevitable consequences to that choice to be considered IMO.
As an example the "Ariel" design chooses to ignore BSC fairly early
on as the ~ 92dB sensitivity is part of its fundamental specifications.
Many builders seem to be very happy with it, though nowhere is
it pointed out to the prospective builder BS has been ignored.
My opinion is that it should pointed out and left to the prospective
builder as to whether they accept that design choice or not.
At the least point them in the direction of :
http://sound.westhost.com/bafflestep.htm
/sreten.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.