New sub design? Constricted Transflex, simple build (series tuned 6th order)

[just a guy]

... well lets call it choking just to differentiate.

Dave

It's usually called port compression but I've heard it referred to as choking, although very infrequently.

 

[Matthew Morgan J]

I just read the last 5 or 10 pages here and see so much talk of chuffing and it's effects.
The biggest problem I see with the whole issue is "audible" chuffing starts after there is already a loss of port output.
This loss of output is very similar to thermal compression in how it reduces the subs output, but it only occurs during the big power hits (high velocities).
I do not know how much the actual velocity is reduced in the middle of a 1/4 wave line but I can say the terminus of that line better have significant cross section to prevent, well lets call it choking just to differentiate.
I just want to try and point out that there are more problems with reduced cross sections than a whistling sound.

Dave

Right , i guess the debate is about at what velocity rate the problems really begin to effect with these 6th order cabinets... There has been numerous examples of successful 6th order designs with a pinch somewhere in the path so the method does work as long as it is not taken to an extreme..

 

[xrk971]

In fluid mechanics, choked flow occurs when the pressure differential across a constriction is large enough to cause the flow to reach the speed of sound at the constriction (an orifice or throat). At this point, increasing the differential pressure does not increase the flow because it is choked with the flow velocity equal to Mach 1. The velocity at the throat cannot exceed the speed of sound. At ambient temperatures the speed of sound is about 342 m/s. Even at 34 m/s flow velocity in the port, it is still only Mach 0.1, far from choked. There may be what people call port compression meaning the pressure losses are high and flow is semi compressible. Even at 34 m/s the flow can be solved quite accurately using the incompressible flow assumption.

 

[freddi]

I think sometimes, the term "port rectification" was used when things choked in a bass reflex - the port and driver become non-linear as result with offsets from "center" - is that an apt term? oil-canning was another term

 

[Matthew Morgan J]

"You can't let this issue go, I guess. I'll play along a bit longer."

Well Jag , when you make ridiculous comparisons like an air compressor line vs a subwoofer cabinet you may get called out on it...

"You can continue to rely on theory "

Nothing wrong with making references to theory in a forum that revolves around science and engineering ... X has referred to some deep theory regarding fluid mechanics, which you don't seem interested in hearing either ...

"Ha ha, clever. XRK already made that joke (about a sealed sub) a few posts back. YOU, on the other hand seem to have chuffing problems in 100 percent of your tested designs, so maybe you might consider IB or OB designs."

Who is joking here? He suggested a sealed box to you , and another gentleman suggested a Bjorno styled tapped pipe, and I suggested OB or IB ..... I cannot speak for the others but i would like to say that my heart goes out to anyone who suffers from a debilitating anxiety regarding chuffing and port compression in an experimental design when there may not be any such issues in a given design...

Diligent & responsible concern are admirable and very practical things which should be included as part of any design process, however that is a far cry from counterproductive paranoia ...

You base my track record on a single short post that mentioned "some" turbulence , you cling onto that single post for the sake of debate, but then you so conveniently ignore the 4 positive (more detailed) posts that followed where the same builder was RAVING about that particular design's performance .............

So think what you will about my track record but lets just take a look at other successful designs that ALSO used a significant mid-path pinch , i have tried to point out a few of those real-world examples to you more than once but you have very little to say about them, once again how convenient ...

I wouldn't be this insistent if you weren't advising people to build your designs without warning them that the designs were never simulated properly, but since you are and no one else is saying anything...

I think it is pretty clear to anyone who frequents DIYaudio that much of what goes on here is experimental, exploratory, developmental etc and that of course is accompanied with some amount of risk if you try building any of these proposed designs..... It is after all Do-It-Yourself as the name would imply ...

If it makes you feel any better about all of this i can provide a canned disclaimer every time i post a new idea here .... Would that satisfy you?

Once again, you made your point last summer, and you made it very clear, good job, but now your just trolling, demoralizing and shooting down people who are trying to explore...

 

[just a guy]

Well Jag , when you make ridiculous comparisons like an air compressor line vs a subwoofer cabinet you may get called out on it...

A restriction in a high velocity duct applies to an air hose and a speaker enclosure. I was trying to give a simplified example so you could understand. Calling me out on a perfectly good example doesn't help your case.

Nothing wrong with making references to theory in a forum that revolves around science and engineering ... X has referred to some deep theory regarding fluid mechanics, which you don't seem interested in hearing either ...

There's nothing wrong with quoting theory but in this case your theory didn't help you to avoid chuffing problems in your design.

XRK posted a formula to describe air flow IN a duct. That formula is useless (or at least not complete) to describe chuffing. I think XRK and I are on the same page with that.

Who is joking here? He suggested a sealed box to you , and another gentleman suggested a Bjorno styled tapped pipe, and I suggested OB or IB ..... I cannot speak for the others but i would like to say that my heart goes out to anyone who suffers from a debilitating anxiety regarding chuffing and port compression in an experimental design when there may not be any such issues in a given design...

Diligent & responsible concern are admirable and very practical things which should be included as part of any design process, however that is a far cry from counterproductive paranoia ...

I have no anxiety about chuffing. I have a HUGE problem with you wasting other people's money on overly complex designs that have not been properly simulated.

You base my track record on a single short post that mentioned "some" turbulence , you cling onto that single post for the sake of debate, but then you so conveniently ignore the 4 positive (more detailed) posts that followed where the same builder was RAVING about that particular design's performance .............

THE DESIGN WAS TESTED AT MODERATE POWER AND HAD AUDIBLE CHUFFING. That negates all the positive comments IMO. If you are happy with designs that exhibit audible chuffing that's fine, but it's a major design flaw when you can get the same (or better) performance from a simple tapped horn without the chuffing.

So think what you will about my track record but lets just take a look at other successful designs that ALSO used a significant mid-path pinch , i have tried to point out a few of those real-world examples to you more than once but you have very little to say about them, once again how convenient ...

I commented briefly about the designs you mentioned. I told you I don't know much about them, haven't read up on them, and I'm not going to study them in depth and simulate them just to find out if they are good designs or not. They might be good designs, they might not. The design I DID look at was yours, and I pointed out a significant and obvious design flaw.

I think it is pretty clear to anyone who frequents DIYaudio that much of what goes on here is experimental, exploratory, developmental etc and that of course is accompanied with some amount of risk if you try building any of these proposed designs..... It is after all Do-It-Yourself as the name would imply ...

If it makes you feel any better about all of this i can provide a canned disclaimer every time i post a new idea here .... Would that satisfy you?

Once again, you made your point last summer, and you made it very clear, good job, but now your just trolling, demoralizing and shooting down people who are trying to explore...

Outside of Danley's recent inventions there isn't a whole lot of new ideas. These designs you are doing are not really unique or what I would call experimentation. This has all been done before and it's easily understood and simulated.

You don't want to admit there's a significant design flaw in your design process, namely unsimulated restrictions in the enclosure. The one design that was tested had audible problems at moderate volume. For pointing this out you have repeatedly called me a troll, implied that chuffing is not a problem, and spent quite a bit of time arguing when you could have been figuring out how to simulate your designs properly.

Again, this wouldn't be a problem if you were "experimenting" by yourself, but you haven't built anything yourself and encourage others to spend their money on your unsimulated designs.

 

[Matthew Morgan J]

Please be sated =P

A restriction in a high velocity duct applies to an air hose and a speaker enclosure.

The differences in CSA scale are gargantuan and the applications are very different so not a very good ANALogy in my opinion thats all ..

I have no anxiety about chuffing. I have a HUGE problem with you wasting other people's money on overly complex designs that have not been properly simulated.

No anxiety ? How about OCD? Call it what you will but it comes across as some sort of unhealthy obsession ... Anyway , trying out new and experimental designs of various types is often what goes on here on DiyAudio, but like i said i can offer a disclaimer with every new submission if it makes you feel better, and i am serious about that .....
By the way , if you really think these designs are overly complicated i would urge you to take a look at many of the other designs floating around out there, some of them make my stuff look incredibly simplistic ...

I commented briefly about the designs you mentioned. I told you I don't know much about them, haven't read up on them, and I'm not going to study them in depth and simulate them just to find out if they are good designs or not. They might be good designs, they might not. The design I DID look at was yours, and I pointed out a significant and obvious design flaw.

The ones i mentioned are 6th order series tuned with a significant squeeze in the mid-path ..... They work, and they work without any reporting chuffing issues , people tell you that they work, and you don't seem to really care, while at the same time you have laserbeam focus on my experimental designs posted here to shoot them down with so much zeal and fervor!!! Your bias is obvious and extreme and it is all based upon a small post made a long time ago about a design that i have long since left behind, while you ignore all of the awesome reviews he left for that same subwoofer later on ...

You say that these types of alignments are nothing new and i can fully agree with that BUT with how that fact pertains to this debate you really cannot have your cake and eat it too .... There is a reason that people keep rehashing and revisiting these 6th order series tuned designs over and over throughout the years, and that is because they work ... I wasn't the first to include a constriction by any means and i do believe i have given proper credit and acknowledgment to the predecessors plenty of times, yet somehow you say my (old) design is horribly flawed and you have little to say about any of those others? ....
So it is now several months later and you feel the need to rake me over the coals about it? again? really? ... Congrats, you are petty,and an incessant cherry-picker and you should probably consider a different mission in life because what you are doing here has become unproductive ..

"spent quite a bit of time arguing when you could have been figuring out how to simulate your designs properly."

You shouldn't assume so much, i have actually put quite a bit of thought and simulation time into developing these recent layouts and i do believe that all of the practice and refinement has been a good thing ... My understanding of how these alignments operate has progressed and i am happy with that ...

Again, this wouldn't be a problem if you were "experimenting" by yourself, but you haven't built anything yourself and encourage others to spend their money on your unsimulated designs.

Nobody is forcing you to build any of these boxes are they? Are you being held captive and against your will by some unsavory brutes who really want you to build these subwoofers??? Are they threatening to do unspeakable things to you unless you build them a Karflex?!?!?!?! Are you in distress?? Is it Snidely Whiplash ?? Someone needs to contact the Mounties right away so they can rescue you!!!

No?

Golly, i certainly hope that nobody else on earth is being forced to build some design i posted ....

JAG, The truth is that people who follow this discussion can choose to build a box or not ... If not that's fine, and if they do then that is fine too ...... Nobody is under duress that i am aware of ... *shrug*

If you feel that i am not being forthright enough about the risks of building a DIY box based on a new layout which is found on a DIY website then like i said earlier i will gladly include a disclaimer on every new design proposal i make here on this website ...... I am serious about that ....

Will you take me up on my offer and will that satisfy you?

........................


My new tablesaw is sitting on the floor of my living room right now, waiting to be mounted out back, and drivers are on the way from PE as we speak... I will build a few Karlflex cabinets for my own use and if they are completely dreadful i will proclaim: "CURSES! FOILED AGAIN!"

 

[Matthew Morgan J]

Delightful !

XKi dual 5in sub RTA measurement and sound clips.
http://www.diyaudio.com/forums/full-range/268524-xki-xs-ab-initio-karlson-6th-order-bandpass-27.html

Xrk ,
I just listened to those recordings of your new new dual 6th order mini-subs , the three tracks you posted a few days go .... I really like the way they sound ....
I don't want to say that they have midbass "coloration" because that word has negative connotations so instead i will say that i really appreciate the lovely midbass presence that i am hearing in these recordings, do you suppose it has anything to do with your well executed front chamber? ... The impression i get is smooth and warm with no offensive midbass honks or holes, with all notes having reasonably even representation ... Good job!

May we hear the sub box all by itself?

 

[just a guy]

If you feel that i am not being forthright enough about the risks of building a DIY box based on a new layout which is found on a DIY website then like i said earlier i will gladly include a disclaimer on every new design proposal i make here on this website ...... I am serious about that ....

Will you take me up on my offer and will that satisfy you?

Just evaluate your designs properly with Akabak. It will take 10 minutes to learn how and after that it will take about 2 minutes per design.

There are a LOT of bad designs out there, and some (maybe even most) of yours might not be bad at all. But until properly simulated there's no way to know. I have NO IDEA why you think I'm picking on you for asking you to properly simulate your designs. You have presented dozens of designs here with the intent that people should build them. That isn't experimentation, it's presenting complete designs to be built. This is unprecedented, I've never seen anything like this. All those other designs you refer to are relatively few and far between and did not have designers that promote so many designs as being complete and ready to build. They weren't warned of the potential issues, they didn't ignore the warnings and continue to post other designs with exactly the same potential issue. This thread has been pinned to the top of the forum for almost a year now, multiple thousands of posts, multiple thousands of views, dozens of promoted designs. Take some responsibility and take 10 MINUTES to learn to properly simulate them. That's all I'm asking. Is that really too much?

This is NOT cherry picking. There's only one design reviewed. It had audible problems. I TOLD YOU it was going to have audible problems before it was built. You refuse to take any responsibility for this at all. Continue to call me names if you like, but be aware that this could all be avoided if you would take 10 MINUTES out of your day to learn to properly simulate your designs. If that's too much work for you then maybe you should post a disclaimer with every new design.

 

[xrk971]

Xrk ,
I just listened to those recordings of your new new dual 6th order mini-subs , the three tracks you posted a few days go .... I really like the way they sound ....
I don't want to say that they have midbass "coloration" because that word has negative connotations so instead i will say that i really appreciate the lovely midbass presence that i am hearing in these recordings, do you suppose it has anything to do with your well executed front chamber? ... The impression i get is smooth and warm with no offensive midbass honks or holes, with all notes having reasonably even representation ... Good job!

May we hear the sub box all by itself?

Hey thanks! There shouldn't be any mid bass coloration - it measures pretty flat throughout the subs bandwidth except for a floor bounce dip. I am integrating it with the PRV 5MR450NDY which has the sensitivity to keep up with the sub. So far it sounds better and with lower distortion because the Vifa was struggling a bit in the 200Hz to 400Hz range whereas it is easy on the PRV with a 5in cone.

Sure I can get you a clip of bass only but it will sound not like music. I will toggle the top on and off in same clip so you get a sense of what sub is adding.

Btw, I have actually done a detailed akabak sim of one of your earlier designs for a dual W5-876SE implementation. Sim included corners, hairpins, finite location of driver near mouth region, etc. it looked pretty good but I was never concerned with port velocity as I would be running it at pretty low drive levels. When I get my computer back from the IT shop I will look for that.

You should get Akabak installed and try the export from HR to Akabak as a start. That will let you check velocities anywhere.

 

[Matthew Morgan J]

Hey thanks! There shouldn't be any mid bass coloration - it measures pretty flat throughout the subs bandwidth except for a floor bounce dip. I am integrating it with the PRV 5MR450NDY which has the sensitivity to keep up with the sub. So far it sounds better and with lower distortion because the Vifa was struggling a bit in the 200Hz to 400Hz range whereas it is easy on the PRV with a 5in cone.

Instead of coloration i think the phrase "excellent midbass presence" is an appropriate way for me to describe my impression, i really liked the rich and warm sound of the recording

Sure I can get you a clip of bass only but it will sound not like music. I will toggle the top on and off in same clip so you get a sense of what sub is adding.

Btw, I have actually done a detailed akabak sim of one of your earlier designs for a dual W5-876SE implementation. Sim included corners, hairpins, finite location of driver near mouth region, etc. it looked pretty good but I was never concerned with port velocity as I would be running it at pretty low drive levels. When I get my computer back from the IT shop I will look for that.

Both of these things would be fantastic

You should get Akabak installed and try the export from HR to Akabak as a start. That will let you check velocities anywhere.

Yessir, Akabak is in my near future, it is a long overdue thing for me ..... What were the restrictions with the install? Something about how it doesn't work with 64 bit operating systems? On the two computers (that i use for designing speakers) the lappy has 32bit win7 and the tower has XP Pro SP3 (also 32 bit)...

 

[xrk971]

Akabak needs 32bit Windows so you are fine on either machine. If using 64bit install a virtual 32bit machine and run in that.

 

[Matthew Morgan J]

I am not opposed to this idea

quote snippets

Just evaluate your designs properly with Akabak. It will take 10 minutes to learn how and after that it will take about 2 minutes per design.

"There are a LOT of bad designs out there, and some (maybe even most) of yours might not be bad at all. " .........

" But until properly simulated there's no way to know."


" This is unprecedented, I've never seen anything like this.".

....."properly simulate them" ....

"Take some responsibility and take 10 MINUTES to learn to properly simulate them. That's all I'm asking. Is that really too much?"

JAG ,
Ok, you claim to be purely well intentioned, and i would like to believe that, however you also keep insisting that my designs are not simulated (?), or not properly simulated (but im pretty sure that what you mean is that they aren't simulated the way that you want them to be simulated) ...

Clearly my designs are simulated as they are all based upon software models and i know you are aware of this because you have followed this discussion long enough to have seen me post enough HR input screenshots, graph screenshots, HR export files, and i have also referred to Hornresponse at least a thousand times in my posts ....
So the fact that these designs are "simulated" is pretty well established, but what we are REALLY talking about here is how i am not simulating them using methods that you would prefer .... Ok , i get it now ...

JAG, Sir , I will level with you as best i can, First of all please understand that i am not just recklessly throwing these layouts together, and i am NOT just posting this stuff on here in a thoughtless or haphazard manner, i come up with a rough concept and revise from there ... I hope i can help to put your mind at ease by letting you know that since we have had your input (regarding the importance of keeping velocity within bounds) last summer i have been much more aware of velocity figures in my sims .... You brought it to my attention and your influence upon my design methods have not faded since ....

Just to let you in on my technique for finding an approximate velocity for a given design i will adjust the CSA of HR's "s5" ( HR's Horn Mouth in TH mode) temporarily to make it match the CSA of the internal vent (constriction) then adjust the "vent" length to again hit my target FB then make a note of that mouth velocity figure and from that decide on whether or not i need to make the internal vent larger ...... To be fair is should also take into account the fact that since the real vent in the real layout will be mid-path instead of end-of-path this suggests a discount in that noted velocity figure ............... The question in this case is of course how much of a reduction can we expect?

I know this may seem like a lot of extra work to figure this out in HR but i have long been determined to come up with all sorts of creative ways to operate this wonderful HR software to see just how versatile it can truly be, this is fun for me ........
I may in fact (as of the last few days) have found a way to derive my "reduction" answer using a basic ML-Transflex model utilizing a very large throat chamber and noting "tapped throat" velocity figures as i move the end of that throat chamber (along with it's port) up and down the path , from the very end to well inside the path.... It is not fast or necessarily convenient, but the experiment seems to work so far...
Using one of the old ML-Transflex (ML-TP) designs with an Alpine driver and a deliberately undersized vent (in this case the throat chamber's port Ap1/Lpt) i was able to simulate the velocity at end-of-path , then 20cm inside of path , then 45cm inside , then 60cm , and so on readjusting vent length and chamber size at every step to maintain size and FB .... . The reductions are apparent albeit seemingly conservative, for example -10% at 40cm , -17% at 60cm , and -25% at 80cm (all rounded numbers) ..... I will run this experiment again with a larger port to see if the numbers scale the same way, and if they do then i have new guideline to use for designing (after verifying with Akabak of course) ..
However with all of the above said , there are some factors that HR does not take into account when calculating velocity such as box losses from panel flex and such (which will reduce the peak in velocity near FB) , and the use of stuffing which will also reduce the velocity peak .... It is good to keep in mind the HR simulates as a concrete box , or a lossless box made of stone and therefore the peaks tend to be exaggerated in sim ....... This phenomenon has been demonstrated repeatedly by builders who compare their real measurements to simulations ...

In these newer offset designs like the Karlflex there is space to add filling/stuffing in the S1 chamber , which will not only knock down particle velocity slightly , but can also reduce an amplitude response peak at FB (if needed) , as well as reduce the peak in group delay and tighten up impulse response ....

Truth is that our software won't be capable of telling us what the velocity figure is with precision, it can only give us an approximation based upon the information it has available and it's algorithms ....

Admittedly my methods have improved tremendously since this discussion was first started several months ago but that was the idea, i was looking for feedback, comments, and guidance was welcomed as well, this website is an excellent peer-review resource when it comes to things such as this .......

As i continue to refine my ideas and products i consequently continue to refine myself and my methods, it is a hermetic sort of process, and i am greatly appreciative of everyone who became involved , even you JAG, (as you have shared your experiences and knowledge and put quite a lot of effort into driving your point home) ...

JAG, I am going to be a decent person and apologize for calling you petty, but honestly how do you expect people to react when you assume and publicly accuse them of being irresponsible?

The truth is that i respect you and all of the other big brains here on this website ...

I am definitely no expert on fluid mechanics, and it seems that both you and X are knowledgeable on the subject, however your (JAG's) personal preference for velocity limits seem impractically low for these particular designs, yet on the other extreme end of the spectrum XRK's calculations suggested that even at velocity figures of 40+ m/s we aren't even remotely close to approaching mach1 and the related non-linearities.... After some research on other forums and websites i am finding that most people consider their maximum allowable figure to be somewhere in the middle of this spectrum and so i am generally shooting for that as well ....

I am also pondering that a vent may behave differently with different alignments, after all can we really expect a vent to work on a 4th order cabinet the exact same way it would on an 8th order cabinet? Maybe , maybe not .... HOPEFULLY our software would give us a general idea, but that's all we can really expect , an approximation just like any of these virtual models ... ...........................................................
I personally feel pretty good about the fact that other closely related 6th order designs do not seem to suffer from any apparent chuffing or port compression, and those designs have been built by highly competent and capable people who would likely notice these sorts of problems (if there were any such problems to notice) ....

Anyway , *end of rant* Yes, Akabak, it is in my future, it is imperative and it is inevitable ...



<3

 

[just a guy]

JAG ,
Ok, you claim to be purely well intentioned, and i would like to believe that, however you also keep insisting that my designs are not simulated (?), or not properly simulated (but im pretty sure that what you mean is that they aren't simulated the way that you want them to be simulated) ...

The velocity thing is only one issue. IIRC I've seen one of your designs with dual mouths at different points in the design. That can't be simulated with any degree of accuracy with Hornresp, but it can be with Akabak and TL.app. Akabak has a steep learning curve if you want to write scripts but TL.app is almost braindead simple to use even without instructions. TL.app can't simulate tapped horns so you can't use that feature as it doesn't provide accurate results. But Akabak can do it all.

Just to let you in on my technique for finding an approximate velocity for a given design i will adjust the CSA of HR's "s5" ( HR's Horn Mouth in TH mode) temporarily to make it match the CSA of the internal vent (constriction) then adjust the "vent" length to again hit my target FB then make a note of that mouth velocity figure and from that decide on whether or not i need to make the internal vent larger ...... To be fair is should also take into account the fact that since the real vent in the real layout will be mid-path instead of end-of-path this suggests a discount in that noted velocity figure ............... The question in this case is of course how much of a reduction can we expect?

I know this may seem like a lot of extra work to figure this out in HR but i have long been determined to come up with all sorts of creative ways to operate this wonderful HR software to see just how versatile it can truly be, this is fun for me ........
I may in fact (as of the last few days) have found a way to derive my "reduction" answer using a basic ML-Transflex model utilizing a very large throat chamber and noting "tapped throat" velocity figures as i move the end of that throat chamber (along with it's port) up and down the path , from the very end to well inside the path.... It is not fast or necessarily convenient, but the experiment seems to work so far...
Using one of the old ML-Transflex (ML-TP) designs with an Alpine driver and a deliberately undersized vent (in this case the throat chamber's port Ap1/Lpt) i was able to simulate the velocity at end-of-path , then 20cm inside of path , then 45cm inside , then 60cm , and so on readjusting vent length and chamber size at every step to maintain size and FB .... . The reductions are apparent albeit seemingly conservative, for example -10% at 40cm , -17% at 60cm , and -25% at 80cm (all rounded numbers) ..... I will run this experiment again with a larger port to see if the numbers scale the same way, and if they do then i have new guideline to use for designing (after verifying with Akabak of course) ..

Again, I applaud your creativity, but using the right tool will give you the right answer in a couple of minutes. I have no idea how well this "fudged" method will work to predict velocity but I'm guessing it's not accurate at all.

However with all of the above said , there are some factors that HR does not take into account when calculating velocity such as box losses from panel flex and such (which will reduce the peak in velocity near FB) , and the use of stuffing which will also reduce the velocity peak .... It is good to keep in mind the HR simulates as a concrete box , or a lossless box made of stone and therefore the peaks tend to be exaggerated in sim ....... This phenomenon has been demonstrated repeatedly by builders who compare their real measurements to simulations ...

In these newer offset designs like the Karlflex there is space to add filling/stuffing in the S1 chamber , which will not only knock down particle velocity slightly , but can also reduce an amplitude response peak at FB (if needed) , as well as reduce the peak in group delay and tighten up impulse response ....

Truth is that our software won't be capable of telling us what the velocity figure is with precision, it can only give us an approximation based upon the information it has available and it's algorithms ....

Not only that, but the highpass filter (which should always be used with resonant systems, especially subs) will reduce the velocity near Fb (which is usually the biggest problem by far) by a tremendous amount. Akabak can simulate velocity with the filters in place, Hornresp can't.

Velocity is never going to be as high as predicted. The problem though, is when audible issues make it though into the built product so we have to be vigilant in doing what we can to reduce it. As I've mentioned, you can get the same output with designs that do NOT have a restriction in the line, so that's why I'm so adamant that these designs should be analyzed properly.

JAG, I am going to be a decent person and apologize for calling you petty, but honestly how do you expect people to react when you assume and publicly accuse them of being irresponsible?

No need to apologize, I really don't care about personal comments at all, the only reason I pointed it out is to emphasize that it does not further the conversation to a mutually acceptable conclusion.

But I will respond to the information that's posted. From my perspective, it takes 2 minutes to evaluate the sims in Akabak, failing to do so when a possible issue has been brought up is willfully negligent.

Now that you have expressed an interest in learning to do this, I will help you and will no longer be calling you irresponsible as long as you follow through.

The truth is that i respect you and all of the other big brains here on this website ...

I am definitely no expert on fluid mechanics, and it seems that both you and X are knowledgeable on the subject, however your (JAG's) personal preference for velocity limits seem impractically low for these particular designs, yet on the other extreme end of the spectrum XRK's calculations suggested that even at velocity figures of 40+ m/s we aren't even remotely close to approaching mach1 and the related non-linearities.... After some research on other forums and websites i am finding that most people consider their maximum allowable figure to be somewhere in the middle of this spectrum and so i am generally shooting for that as well ....

Depending on port size and shape there can be massive chuffing at 40 m/s. I think XRK was talking about CORE limit acceptable velocity, which is much different than acceptable velocity at the duct exit, which is CHUFFING limit with his percentage of mach comments. Core limit and chuffing limit is extremely different, and it's why flares are effective. This means you can use a small duct and your core limit will be fine, but you will have chuffing unless both ends are flared. Otherwise you can make the whole duct large but unflared and have the same effect but that's not a conservative use of enclosure volume.

Rules of thumb don't take port size and shape into consideration. If you want to follow a rule of thumb the safest bet is to not exceed 10 m/s. That is MJK's standard, as well as many others. I always try to use this standard too, although sometimes it's very hard.

There's another problem here. Sometimes people will exceed xmax by a fair bit in the real world. I do it all the time, I am very hard on subs, especially when testing, finding their limits, showing them off, enjoying it loud, etc. In cases like that, velocity will be a lot higher than it was when simulated at xmax.

Other people are happy to design with a 27 m/s rule of thumb at xmax, I've seen some people that are happy with even higher velocity in their sims. IMO and IME this is WAY too high.

Compression starts LONG before chuffing is an audible problem.

I'm no master at fluid dynamics, and thankfully no one has to be to design a high performance duct. Akabak and Flare It is all you need.

I am also pondering that a vent may behave differently with different alignments, after all can we really expect a vent to work on a 4th order cabinet the exact same way it would on an 8th order cabinet? Maybe , maybe not .... HOPEFULLY our software would give us a general idea, but that's all we can really expect , an approximation just like any of these virtual models ... ...........................................................
I personally feel pretty good about the fact that other closely related 6th order designs do not seem to suffer from any apparent chuffing or port compression, and those designs have been built by highly competent and capable people who would likely notice these sorts of problems (if there were any such problems to notice) ....

A duct is a duct. Velocity is velocity. Some designs with the duct INSIDE the chamber can provide massive amounts of masking so you might not hear audible problems but distortion and compression measurements will always show if there's a problem or not. The program material can provide a lot of masking too. Turning the box around so it's not facing you and sitting far back can provide a lot of masking too. Additionally, some people find chuffing a lot more offensive than others. To me it's like nails on a blackboard, and the fact that I KNOW it's a result of compression and distortion makes me hate it even more. A lot of people play content that doesn't go as low as Fb (like girls and guitars music) and will never have a velocity problem simply due to that. A lot of people don't ever turn it up loud enough to have velocity problems. Everyone will have their own subjective impressions of the same thing.

For those reasons and more, I put almost no weight at all on subjective reviews unless they point out a particular problem. How loud is loud? It's different for everyone. How clean is clean? How "fast" and accurate is "fast" and accurate? How do you quantify "weight", "punch", "slam", etc? These are all useless words that mean nothing. How much experience do these subjective reviewers have? Are they just impressed because their sub is kicking up a massive room resonance? If they were testing the same sub blind and didn't know which sub it was would they have the same impression? If they tested the same sub a year later would they have the same impression?

BUT chuffing is not subjective, it's an audible phenomenon that everyone can hear. Some people might not mind it as much as others, but if someone says they hear it at moderate volume that's a HUGE red flag for me.

I don't care if people like their bandpass and Karlson and Karlflex and whatever else they have. Until I test it myself (or at least simulate it) I can't judge if it's good or not.

Subjective reviews mean nothing UNLESS they point out something that is not subjective like the presence of chuffing.

Anyway , *end of rant* Yes, Akabak, it is in my future, it is imperative and it is inevitable ...



<3

Here's how you start. Unfortunately I'm not near my Akabak computer at the moment so this is purely from memory. If I make a mistake XRK can correct me.

Install Akabak.
Export a Hornresp file by clicking Export - Akabak script from the inputs window.
Start Akabak.
Import the file, or just place it in the scripts folder.
When the script is displayed, click Inspect from the menu, then Velocity.
Set the power level you want in the input box, select the node you want to look at from the drop down box.
Click OK or Analyze or whatever the option is.
Check the graph.

That's all there is to it.

For bonus points -

Install Flare It.
Fill in the Port Diameter and Flare Radius boxes.
Click Calculate.
Make sure both the red line (AND the blue line in the graph IF there is an abrupt discontinuity at that point in the line) indicate a HIGHER acceptable velocity than the highest velocity from the Akabak graph.

Once you get used to it you should be able to do it all within a couple of minutes.

For extra bonus points -

Learn to manipulate the imported script in Akabak to allow you to simulate dual mouth designs and other designs that Hornresp can't simulate.

Unfortunately this last step is a bit more work, but that's why you get extra points for it.

 

[GM]

TL.app can't simulate tapped horns so you can't use that feature as it doesn't provide accurate results.

Hmm, its response plot looks a little odd compared to HR & AkAbak, but when I punch in the dims I want, it calcs a 'close enough' equivalent.

GM

 

[Matthew Morgan J]

Wacky fudge and the filthy frikkin interwebz.

"The velocity thing is only one issue. IIRC I've seen one of your designs with dual mouths at different points in the design."

Right , there is that dual mouth aperture you mentioned, and there is also a need to simulate an offset driver within a very large throat chamber for another design , and i would also like to experiment with modifications of the DCR script and 6th order scripts, attempt to make hybrids etc etc ....
If i can become proficient in the use of Akabak i know it creates many more possibilities and lessens the need for workarounds, i agree that it is time to stop procrastinating ...

Again, I applaud your creativity, but using the right tool will give you the right answer in a couple of minutes. I have no idea how well this "fudged" method will work to predict velocity but I'm guessing it's not accurate at all.

OH YES , about that "fudged" method as you say , there was a vital detail that i failed to mention in my previous post ... In an ML-Transflex model (with the throat chamber being huge, even larger than the s1-s5 sections combined , sort of like a tapped-ML-TL) THE POINT AT WHICH HR MEASURES "PARTICLE VELOCITY-----> TAPPED THROAT" IS THE S2 POINT so you have to reduce both s1 and s2 CSA to match Ap1's area then eliminate the distance between s1 and s2 which is no problem because that is something that should already be done in an ML-Transflex model ("L12" slider set to 0.10 in "Loudspeaker Wizard" ) .....

It is one of the wackiest workarounds i have used thus far but it wasn't something i expected to use very often, i just wanted to get a general idea of how much reduction in particle velocity i can expect with the vent/constriction placed at various points along the path ........................ The results do seem to follow the pattern that you might expect but who knows just how accurate these results are until verified with Akabak, which i fully plan to do ...

Anyway , if anyone wishes to witness this nutty experiment for themselves just let me know and i can produce three or four Hornresponse ML-TP export files with the vent at different points along the path so it is easy enough to see the changes in the Tapped Throat's estimated velocity ...

As I've mentioned, you can get the same output with designs that do NOT have a restriction in the line

Well yes, that is true, but there is this related issue of physical dimension challenges (especially) with some of these modern drivers which have very low VAS figures and very strong motors combined with resonant pipe cabinet alignments such as a tapped-pipe (a plain ole transflex) calls for dimensions to be made so skinny (yet long) that the shape can be impractical or simply impossible due to the fact that the driver would not properly fit into the line ..... I have run into this issue more than a few times when figuring out the actual physical dimensions of a cab ... I personally prefer quarter wave resonant cabinets and i also prefer compact cabinets so i guess it is no surprise that i would encounter such an issue...... A solution is to use some amount of mass-loading (a constriction) to physically shorten the line while maintaining net size, just enough to make everything fit properly, and that is what started me down this path to begin with ....................................
Nevertheless this mass-loading journey began due to the physical feasibility difficulties but after some amount of modeling i began to recognize the patterns of other interesting effects that some strategically placed mass loading can have upon performance such as alterations in damping behavior and bandwidth, and IMO these effects can be useful under the right circumstances..............................

Depending on port size and shape there can be massive chuffing at 40 m/s. I think XRK was talking about CORE limit acceptable velocity, which is much different than acceptable velocity at the duct exit, which is CHUFFING limit with his percentage of mach comments. Core limit and chuffing limit is extremely different, and it's why flares are effective. This means you can use a small duct and your core limit will be fine, but you will have chuffing unless both ends are flared. Otherwise you can make the whole duct large but unflared and have the same effect but that's not a conservative use of enclosure volume.

Ok , yes , and i do recall you talking about the core velocity and the beneficial uses of flares last summer ..... This concept is intriguing, and i am excited to know that using some amount of series k-aperture action on the front of a Karlflex cabinet creates a flare which effectively eases velocity !! Of course you are also saying that the flares need to be placed on BOTH sides of the vent to perform ideally right? , and i do have a simple internal layout tweak in mind to accomplish an internal flare but it is not likely to be EXACTLY like the external flare since one will be k-aperture based and the other will not ..... Slightly mismatched flares will have to suffice ...

By the way, about this Mr Karlson guy: over time I am recognizing (more about) just how absolutely brilliant some of his ideas are! His K-aperture allows for some tapped effect (which is a 6th order series tuned feature, and in the Karlflex case an optional parallel tuned feature as well!) , while ALSO allowing an effective flare/expansion at end of path while also establishing a front chamber and a path extension without taking up too much space! AND all of the above without entirely wiping out the upper bandwidth response!! This guy was a fully ingenious acoustical superstar!!!!!!!!!! I really appreciate his concepts more than ever as i am developing a better understanding of them ..

For those reasons and more, I put almost no weight at all on subjective reviews unless they point out a particular problem. How loud is loud? It's different for everyone. How clean is clean? How "fast" and accurate is "fast" and accurate? How do you quantify "weight", "punch", "slam", etc?
I don't care if people like their bandpass and Karlson and Karlflex and whatever else they have. Until I test it myself (or at least simulate it) I can't judge if it's good or not.

Subjective reviews mean nothing UNLESS they point out something that is not subjective like the presence of chuffing.

I do understand the difference between subjective reviews and hard data, and i try to respectfully employ critical thinking wherever i can but at the same time wouldn't it be unwise to disregard the opinions or impressions being offered to the public by some seasoned veterans or highly capable and experienced enthusiasts?.....These people are not noobs, they are not green, so their statements carry weight and i tend to listen and take note...
JAG, You are surely correct about the ultimate value in actually getting your hands on the cabinet and testing it for yourself but if that is not possible then i will gladly partake in some second hand reviews coming from an honorable source, and if i am not confident about the source then i (or anyone) can (and probably should) take it with a grain-of-salt just like anything else on the FILTHY FRIKKIN INTERWEBZ!!!!!!! ...






I am installing Akabak now.

 

[freddi]

the original Karlson cabinet (finished in the summer of 1951) plays pretty clean - moreso than some horn cabinets. I purchased a Yorkville USC1 and ran a few sine tests of it vs a one sheet K-type tuned with two 4" id ducts (not flared) - USC1 had 20dB more H2 at only 20vrms. This K-type wasn't as smooth as K15 which had a reflective front board. The rear section of K15's board acted as a mild choke and filter.

when K15 was subjectively compared to a 3.6 cubic foot reflex tuned to around 50Hz and driven at 36Hz/20vrms, the Karlson rectangular vent and main aperture had a lot less chuffing than the reflex and less cone excursion. K15 in its upper front chamber has ~Sd CSA IIRC so is not 'pinched".


I think I could take a K12 of the distributed slit vent variety and play bass guitar into it and the listener would not perceive any pitch shift or distortion - but if I ran sine at say 60Hz and 1 watt level into the same cabinet, there would be distortion stronger than the fundamental (F3 is higher than 60) and a series of (loud) harmonics extending way up. It would blow out a candle a foot away from the aperture at 1 watt. A single 10 or so sq.in. vent would fix the sine distortion. If a given regular coupled-cavity K-type with a single or double vent subjectively sounds somewhat heavy/'thick" then a few small holes added to the reflector wil make the sound "faster". (crazy subjective stuff)


how does music type and instrument type matter with regards to subjective bass distortion? when does music approach sine conditions? I would assume a classical jazz trio of upright bass, acoustic drumkit and piano to be different than some synthesized modern bass which may have some reasonably pure sine sweeps within.

it'll be real interesting to see Akabak simulations of MMJ's proposals.

 

[Fiddler9]

i'm not sure if this would work perfectly

 

[just a guy]

Hmm, its response plot looks a little odd compared to HR & AkAbak, but when I punch in the dims I want, it calcs a 'close enough' equivalent.

GM

A long time ago I did a comparison of an identical front loaded horn in Hornresp and TL.app. The results were so close they would almost overlay exactly if you put them on the same graph.

Quite a while later I compared an identical tapped horn in both programs and the results were much worse. I don't remember exactly how bad it was but I clearly remember thinking it was not acceptable at all.

It's entirely possible that he could have fixed that since I last did a comparison, but I did specifically ask if he had worked on it in the TL.app thread and got no response so I just assumed he hadn't.

I'll have to do another comparison of an identical tapped horn with both programs when I get some time. Can't do it this morning.

 

[just a guy]

Right , there is that dual mouth aperture you mentioned, and there is also a need to simulate an offset driver within a very large throat chamber for another design , and i would also like to experiment with modifications of the DCR script and 6th order scripts, attempt to make hybrids etc etc ....
If i can become proficient in the use of Akabak i know it creates many more possibilities and lessens the need for workarounds, i agree that it is time to stop procrastinating ...

Akabak has some aspects that are not as friendly as Hornresp, just so you know. Other than the steep learning curve, there's no schematic (Hornresp's schematic is really helpful to verify you are not doing something stupid and not realizing it), script writing and editing is tedious (compared to Hornresp's sliders) and it's just generally not as user friendly.

Well yes, that is true, but there is this related issue of physical dimension challenges (especially) with some of these modern drivers which have very low VAS figures and very strong motors combined with resonant pipe cabinet alignments such as a tapped-pipe (a plain ole transflex) calls for dimensions to be made so skinny (yet long) that the shape can be impractical or simply impossible due to the fact that the driver would not properly fit into the line ..... I have run into this issue more than a few times when figuring out the actual physical dimensions of a cab ... I personally prefer quarter wave resonant cabinets and i also prefer compact cabinets so i guess it is no surprise that i would encounter such an issue...... A solution is to use some amount of mass-loading (a constriction) to physically shorten the line while maintaining net size, just enough to make everything fit properly, and that is what started me down this path to begin with ....................................

Pick a design of yours (or do a new one) with any driver you like. As long as the restriction isn't too small (which I would consider cheating) I should be able to match it in output and tuning frequency in an equal size with a design that doesn't have a restriction.

Ok , yes , and i do recall you talking about the core velocity and the beneficial uses of flares last summer ..... This concept is intriguing, and i am excited to know that using some amount of series k-aperture action on the front of a Karlflex cabinet creates a flare which effectively eases velocity !!

It's the size of the mouth that determines velocity, not the shape so much.

Of course you are also saying that the flares need to be placed on BOTH sides of the vent to perform ideally right?

Ideally, yes.

I do understand the difference between subjective reviews and hard data, and i try to respectfully employ critical thinking wherever i can but at the same time wouldn't it be unwise to disregard the opinions or impressions being offered to the public by some seasoned veterans or highly capable and experienced enthusiasts?.....These people are not noobs, they are not green, so their statements carry weight and i tend to listen and take note...

Even if they are highly capable seasoned veterans, you need to know them pretty well for a subjective impression to have any real weight. For example, I have a buddy around the block. We have listened to dozens of subs together in his listening room. We have talked extensively about the things we listened to. We share a common frame of reference. I know his room, I know what he means when he uses certain words (or doesn't use certain words) and I know what aspects he considers important and what he doesn't.

Subjective opinions have to be interpreted and without having a common frame of reference it's impossible to interpret them. It's like reading subjective opinions on the taste of a recipe on the internet - is it useful? Who knows, any given person might have the same subjective opinion while others might disagree strongly.

This is why we need measurements.