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

I will agree it's getting tiresome.

Flare It was tested with measurements and listening tests. That's all that really matters (as long as it's used with the stipulations I just mentioned), you don't need to be a rocket scientist to design a high performance duct. There's a reason why core velocity limits inside the duct and chuffing limits at the duct exit occur at different velocities. Your formulas don't address that, Flare It does. Reynolds number might be a principle parameter (so technically it's not completely useless - you are good at pointing out small careless errors in semantics while ignoring the main topic) but your formulas are severely inadequate to predict chuffing. You said yourself the formula was for in pipe flow, not for exiting pipe flow.

And if you read the link I posted you would know that the definition of chuffing as defined by Flare It is "Any audible noise, other than the intended signal, detected by listening with the ear close to the port".
 
Last edited:
Just in case I'm not being clear here, I'd like you to show me the mathematical formula describing the chuffing limit of a 76 mm diameter abs port with 10 mm rounded flares on each end. I'd like to see how complex that formula is. I can enter those parameters into Flare It and have accurate results based on real world testing in about 5 seconds.
 
Just in case I'm not being clear here, I'd like you to show me the mathematical formula describing the chuffing limit of a 76 mm diameter abs port with 10 mm rounded flares on each end. I'd like to see how complex that formula is. I can enter those parameters into Flare It and have accurate results based on real world testing in about 5 seconds.

Chuffing limit as defined by Flare It is audible noise other than signal - I would ask how audible and from what distance? It's like if I put my ear 4 in away from a speaker with no input to amp, I will hear hiss. But it is inaudible at listening distance say 6 feet away. In engineering, the effect of vortex generation and recirculation produced by sudden expansion jets is normally associated with pressure losses (the variable of interest in practical systems). These relations are quantified empirically and tabulated in a graph or table. Not different than your Flare It program. The independent variable to correlate the pressure loss is the Re number.

If you say the Flare It program is an empirical chuffing noise predictor based on listening tests - I will accept that is what it is. Turbulence induced jet noise though, is very accurately related to Re number.

Let's take a step back and bring the tone down. We are both diy enthusiasts who use Akabak, the Brotherhood of Akabak users is few and far in between and that should be enough to make us friends - I don't want to argue with you. I just am thoroughly trained in the science of fluid mechanics so when you make absolute claims of Re number being useless for correlating jet noise with momentum and viscous flows, it won't go un-noticed.

Let's call a truce and work on Akabak sims of these speakers instead of arguing.
:cheers:
Regards,
X
 
Sure , lets review

Nice try, but this was WAY before he built the tapped horn.


I'm not happy that you just made me skim 50 pages of this thread but I knew my memory wasn't that bad.

Ok JAG,
Had a stroll down memory lane have we? 😛 Well i suppose i just did too ...... It is a flashback from last summer !! 🙄
I went back and read quite a few pages and alright, i see what you are talking about now .... First he had air leaks from the screws that held in his bracing which he then sealed up with "PL" (good stuff , i use it too) and that was post #326 ......
Then in #333 he does indeed report "some" turbulence but it doesn't seem to take much away from his impressions of this box because he goes on to say some very positive things in his reviews and observations in posts #376 , #382 , #431, #439 ..........
Here are some juicy rave-review excerpts:

  • "Subjectively it definitely fills it's role as a "compact dj sub" in that it sounds smooth even driven hard, has decent low end "
  • "In my small room 1 dominates it if pushed hard"
  • "should be loud enough to impress and IMO if he ran them at full power and was anywhere near them would have hearing damage issues"
  • "This subwoofer (swe10s4) seems pretty balanced"
  • "Then after building the second ML-TRANSFLEX he says "Guys, it's official, 2 of these things is absolutely destructive. Same order of magnitude as the T60 (+-3 dB) and a little better on the 32-35 end as predicted by hornresp. Serious house rattle. Can't believe I built a pair for 160. Truely impressive"
  • "Brutal yet accurate with no sign of strain (this is all subjective and from 12 feet away). What surprises me most is it still generates enough output to be useful on 30 hz bass notes"
  • "Just did a "full power" lorde - Royals (DJ Slow n Throw rebass) and although I haven't compared it side by side with the T60 I wreckon it is probably louder in the 35 hz band than the T60, or definitely just as loud. Absolutely physical shake the whole house kind of loud. That unrelenting smile that comes out when you realize what you've created and how ridiculous it is compared to anything commercial you've compared it to"
  • "Still didn't hit the limiter, I just can't realistically get there, I know I'm close say maybe +2 dB but it's just absolutely insane that I don't for fear of breaking something." "Firm believer now that multiples make everything better!"
  • "if this sub didn't hit mid 30's and with authority than I wouldn't have enjoyed this song"
Nothing more was mentioned about any turbulence (after post #333), and he never used the term "chuffing" in regards to this box, furthermore he keeps describing (in a variety of words) the high output levels he was getting from these compact subs so it doesn't sound very likely that port compression was hindering their performance by much if any, and he really seemed rather impressed with their performance ......


In fact phrases like "smooth even driven hard" and "Brutal yet accurate with no sign of strain" would seem to imply an absense of chuffing and and absense of distortion ... Am i assuming too much here?



Anyway, you could be the cynical type and suggest that he was just humoring me with all of his positive public reviews, so in that case, lets talk about other designs like the Karlsonator , and Brian Steele's POC2 Tapped Horn ....

The Karlsonator has quite a constriction mid-path, but it doesn't seem to be an issue , go read the Karlsonator discussion and you can see for yourself ....... Even within this discussion you can go read post #353 where XRK discusses the lack of turbulence/chuffing with the Karlsonator , and he is very knowledgeable in regards to that design and other similar 6th order series tuned variants ... He has plenty of real world experience with them ...


Furthermore , Brian has never mentioned any issues with chuffing or port compression in his POC#2 even when asked about it he will tell you that it is not a problem with that box even though his CSA was squeezed down to 1/3rd that of the standard mid-path in a tapped horn with his "DFD" constriction ...



So once again the Karlsonator works , and Brians POC#2 TH works , SabaSpeed's ML-Transflex loaded with an Alpine works, these are real-world experiences which counter your argument and Im sure that these are not the only ones .. So what in the heck is going on here JAG???? Well, perhaps i have (at least part of) the answer for you..



So what is the difference between the real world experiences of the folks who developed "flare it" versus these other real-world experiences listed above? I can say that one obvious difference is the fact that flare-it calculates for ports that exit at the very end of the path, which is different than our above mentioned real-world examples that have their constriction in mid-path or mid-to-late path (as in Saba's ML-TRANSFLEX) but NOT at the end of the path ....... Since velocity at mid-path is innately much reduced compared to end-of-path (according to quarter wave resonant pipe theory) the pattern we are seeing here actually makes quite a bit of sense ...



Nevertheless , we have all already covered this topic last summer and there is no need to do it all over again .... It is a good thing to take air particle velocity into account when designing cabinets, and i do, more than ever (since last summer) so thanks for that JAG, truly, but lets not get hung up on this velocity topic because it has already had it's fair share of attention in this discussion ...
 
Last edited:
It's all right there in the link I posted.

The onset of "chuffing" is unmistakable and a change of 1dB in input power is usually enough to hear it clearly. Where the transition occured over a range of values, the average of "I think I can just hear it" and "I can definitely hear it" was recorded
I'm confident that anyone who repeated the tests would find their results very close to those presented here. Even if a different threshold for chuffing was chosen, the relationships found in these tests would still be present.

The main area of subjectivity is in deciding how to translate results based on barely audible chuffing, into recommendations for typical music and home theater applications.
A test was done to see how much the port velocity could be increased before the "chuffing" became audible at the seating position.
For a seat 2.5 metres from the sub in a room 5m * 4m, the velocity could be increased by 25%.

An allowance for the masking effect of content can also be used.
The actual amount is based on the performance of sub's I've built, and on existing industry recommendations.
I estimate that reasonable values for masking allowance are an additional 15% for music and 30% for Home Theatre usage.
The equations developed below are based on designing for music, so a total allowance of 25+15 = 40% is used

I always use zero percent masking because ALL chuffing is unacceptable to me. Even if you need to stick your ear 1 cm from the port to hear a barely noticeable onset of chuffing that's too much as far as I'm concerned. That is air flow provided by the driver that is NOT making bass (compression), and even worse than that it's providing an objectionable sound (distortion). It's easy to avoid so there's no real excuse to accept it. Other people don't mind so much. The Cerberus sub is one of the worst designs I've ever seen, with massive chuffing problems. (I own the driver and it chuffed with a port with 4x more area than that design.) But that design is very popular in certain circles. One of the first designs I ever built (Decware Deathbox from plans, long before I knew anything about any of this) and after all the hard work assembling it (with no carpentry skills and inadequate tools) it had chuffing issues and it drove me nuts, it was an overwhelmingly negative experience. That's one of the main reasons I started learning about audio and enclosure design. I'm VERY sensitive to chuffing.

Anyway, the first time I mentioned it I said Reynolds number alone wasn't useful for predicting chuffing. The next time I left out "alone" and that was a mistake. I imagine the formula would have to be very complex to predict the velocity limit at onset of chuffing for any give duct size with or without flare.

Anyway, truce is a good idea.
 
Last edited:
... Then in #333 he does indeed report "some" turbulence but it doesn't seem to take much away from his impressions of this box because he goes on to say some very positive things in his reviews and observations in posts #376 , #382 , #431, #439 ..........

I'm sure it performed well above the average stuff you can get at Best Buy.

Nothing more was mentioned about any turbulence (after post #333), and he never used the term "chuffing" in regards to this box, furthermore he keeps describing (in a variety of words) the high output levels he was getting from these compact subs so it doesn't sound very likely that port compression was hindering their performance by much if any, and he really seemed rather impressed with their performance ......

Audible turbulence is chuffing. If that's not what he meant I have no explanation for his description of what he heard.

Anyway, you could be the cynical type and suggest that he was just humoring me with all of his positive public reviews, so in that case, lets talk about other designs like the Karlsonator , and Brian Steele's POC2 Tapped Horn ....

The Karlsonator has quite a constriction mid-path, but it doesn't seem to be an issue , go read the Karlsonator discussion and you can see for yourself ....... Even within this discussion you can go read post #353 where XRK discusses the lack of turbulence/chuffing with the Karlsonator , and he is very knowledgeable in regards to that design and other similar 6th order series tuned variants ... He has plenty of real world experience with them ...

I've never studied that design so I can't comment. If he checked velocity at all points in the enclosure and it was ok then there's no need to worry about anything. If the driver doesn't have a lot of xmax, 1/4 Sd constriction is probably ok. If the restriction is inside the enclosure a good distance from the mouth, the enclosure will provide a tremendous amount of masking so even massive amounts of chuffing might not sound too bad. That doesn't mean that bad things are not happening, you need distortion and compression measurements to determine that.

Furthermore , Brian has never mentioned any issues with chuffing or port compression in his POC#2 even when asked about it he will tell you that it is not a problem with that box even though his CSA was squeezed down to 1/3rd that of that mid-path in a tapped horn with his "DFD" constriction ...

All comments in my last paragraph apply here. But I know a bit about this design and the dogfood mod. He did this specifically to smooth frequency response and it worked for that purpose. But the mod was literally done with cans of dog food. (And later with thin pieces of wood as a more permanent solution?) As such I kind of doubt that he tested it at high power and did distortion and compression measurements to determine ALL the effects of the constriction.

So once again the Karlsonator works , and Brians POC#2 TH works , SabaSpeed's ML-Transflex loaded with an Alpine works, these are real-world experiences which counter your argument and Im sure that these are not the only ones .. So what in the heck is going on here JAG???? Well, perhaps i have (at least part of the answer for you..

I have an answer too. Masking and lack of distortion and compression testing. I have no doubt that these designs all work. But are they optimized? Maybe, maybe not. In lieu of proper measurements these designs could at least be simulated with Akabak, but I don't have time for all that.

There's a HUGE difference between "it works" and "it has no easily avoidable issues".

So what is the difference between the real world experiences of the folks who developed "flare it" versus these other real-world experiences listed above? I can say that one obvious difference is the fact that flare-it calculates for ports that exit at the very end of the path, which is different than our above mentioned real-world examples that have their constriction in mid-path or mid-to-late path (as in Saba's ML-TRANSFLEX) but NOT at the end of the path ....... Since velocity at mid-path is innately much reduced compared to end-of-path (according to quarter wave resonant pipe theory) the pattern we are seeing here actually makes quite a bit of sense ...

I simulated the design Saba built with Akabak and IIRC velocity at the constriction was 30 m/s. Just because the constriction is not near the mouth doesn't mean that velocity can't be dangerously high. 30 m/s is too much for the size of duct at the constriction in ANY part of the enclosure IMO. Saba tested it at medium power and he heard a problem. Burying the constriction way back in the enclosure provides a tremendous amount of masking but it seems that in Saba's case it wasn't enough to make it inaudible. And since audible chuffing equals distortion and compression, IMO this is not acceptable.

One of the MAIN reasons that horns perform better than ported boxes is due to the lack of distortion and compression caused by ports. Even when a (small) horn and ported box perform ~ equally in a sim the horn will always outperform the ported box by a large margin for exactly this reason.

Nevertheless , we have all already covered this topic last summer and there is no need to do it all over again .... It is a good thing to take air particle velocity into account when designing cabinets, and i do, more than ever (since last summer) so thanks for that JAG, truly, but lets not get hung up on this velocity topic because it has already had it's fair share of attention in this discussion ...

I don't have any more to say, just wanted to bring this up again for those that missed the first round.
 
Last edited:
Imperative

Once again (in regards to resonant pipes) the air particle velocity is not the same at all points in the path ... So if the constriction is mid-path it is possible that there is no chuffing or distortion being generated to begin with .... You are correct though in saying that there would be masking (if there IS any chuffing or distortion to mask that is) ..... I am not disagreeing with you Jag , i am just saying there may be another important factor to take into account .. thats all ...
Velocity at mid-path could be considerably less than what Flare it or HR predicts for the end-of-path ...

Nevertheless yes, right, subjective reports are one thing, measurements are another....
 

Attachments

  • Pressure-vs-Velocity-regions-in-a-closed-end-quarter-wave-pipe.JPG
    Pressure-vs-Velocity-regions-in-a-closed-end-quarter-wave-pipe.JPG
    22.9 KB · Views: 245
Last edited:
UPDATED SKETCH WITH MORE DETAILS,GRAPHS, COMPARISONS AND APERTURES

GENTLEMEN,

This Karlflex can be a very practical, versatile, portable and economical alternative to other alignments even competing with a well known compact performance leader like the tapped horn... Careful choice in drivers makes this possible:key: ... This design really excels in the price/performance department with affordable drivers and cabinets used in multiples. I had mentioned all of this before but i wanted to show some comparisons to illustrate what i have been trying to describe.... 🙂 There are separate graph collages for 25hz tuned cabs and 40hz tuned cabs..

I also filled in some more detailed information on the sketch and threw in a few more ideas for apertures .....:Present:

Envision the following "herd": Utilizing one of the more extraordinary examples given here you can build 6 boxes each loaded with two surplus Rockford Fosgate 10" drivers from ApexJr for a total of 133db in halfspace or 143db in 1/8th space for only $330 spent on drivers (although im not sure what Steve would charge for shipping on those) ... Imagine 3 boxes per side (or put together in one big cluster if you like) with a 3 ohm load per channel, of course this would also have to be 1200w per channel at 3ohms to drive these to full capacity but that is a heck of a lot of sound for a relatively small amount of money ..... Combined net is also only 360 liters for all 6 cabinets! The combined cone area of around 4000 sq cm is certainly nice 😀
 

Attachments

  • 40hz-180L-TH-vs-Karlflex comparison.jpg
    40hz-180L-TH-vs-Karlflex comparison.jpg
    220 KB · Views: 259
  • Karlflex-vs-TH-Comparisons-35hz.jpg
    Karlflex-vs-TH-Comparisons-35hz.jpg
    91.8 KB · Views: 254
  • Various-Karlflex-ApertureConcepts.JPG
    Various-Karlflex-ApertureConcepts.JPG
    393.6 KB · Views: 257
  • Four_pointed_star.jpg
    Four_pointed_star.jpg
    27.4 KB · Views: 250
  • OFFSET ML-TRANSFLEX -35hz-60L-LAB12-OR-SWS-12D-24inch-150-cm 40hz or 160cm 35hz Path-Small-Front.jpg
    OFFSET ML-TRANSFLEX -35hz-60L-LAB12-OR-SWS-12D-24inch-150-cm 40hz or 160cm 35hz Path-Small-Front.jpg
    506.4 KB · Views: 244
Last edited:
Once again (in regards to resonant pipes) the air particle velocity is not the same at all points in the path ... So if the constriction is mid-path it is possible that there is no chuffing or distortion being generated to begin with ....

There's no need to guess or speculate what the velocity might be, just run the sim through Akabak. You don't even need to learn to write a script, just export your Hornresp file, import into Akabak and evaluate.

Anytime you create a pinch in the line there's going to be higher velocity, there's a reason you need to avoid this in air compressor lines, it robs power. In enclosures the highest velocity section of an unpinched line is at the end but that doesn't mean velocity can't be very high at pinched sections further back from the mouth.

Flare It doesn't simulate velocity, it just tells you what velocity is acceptable.
 
Hi Y'all,

Here are some references on the general subject of ports/chuffing:

Maximizing Performance of Loudspeaker Ports, by Salvatti, Devantier, Button

AES 112th: Vented-box geometry and low frequency reproduction: the aerodynamical approch, by Morkerken, Parzy, Pellerin and Polack

AES 116th: Sound source design in the very low frequency domain, by Pellerin, Polack, Morkerken

and maybe:

Polk, power port, Pat. 5,517,573

One very simple way to reduce port velocity is to use one of the folds recommended by bjorno, a tapped quarter wave pipe (T-QWP) without any taper. I'll attach an example for the Beyma 15P1200Nd.

Regards,

P.S.: David just released a major update of Hornresp. Time to update. 🙂
 

Attachments

Velocichat

" there's a reason you need to avoid this in air compressor lines"

Are you seriously going to compare a subwoofer cabinet to an air compressor line???

If i was pinching the path down to the CSA of a typical air compressor line or less on a high powered subwoofer system then yes, i would obviously expect some major problems :Ohno:... I cannot argue with that i suppose... .. Of course i would never even consider such an extreme pinch :no:...

In enclosures the highest velocity section of an unpinched line is at the end but that doesn't mean velocity can't be very high at pinched sections further back from the mouth.

JAG,
Right, this is very true, I agree, and it is basically related the point i have been trying to make ..... A constriction or a vent with a CSA of 300cm placed mid-path should (according to theory) generate considerably less velocity than that same vent placed at the very end of the path ... If we are to assume a linear exchange of pressure-to-velocity in a quarter wave pipe then it suggests that the velocity would be 50% of what is predicted at end-of-path if our vent is placed exactly at mid-path (50%) ...:yinyang: ... This may or may not be optimistic and i understand that it may not work out EXACTLY this way, but surely it is a general trend, simply and quite literally due to the fundamental operation of a quarter wave pipe with one end closed ..... Some Akabak experiments could perhaps tell us how the velocity will likely scale with different path placements.. 🙂 Maybe XRK would want to give this experiment a shot ...
 
Suggestion for JAG

JAG ,
Ever try open baffle designs or infinite baffle? No constrictions, paths or ports involved there ...... Performance and or portability may not be the greatest in OB or IB setups but they could work just fine in some applications and there is no chance for a "chuffing" , not even one chuff .... ever ... 😉
 
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
 
Are you seriously going to compare a subwoofer cabinet to an air compressor line???

If i was pinching the path down to the CSA of a typical air compressor line or less on a high powered subwoofer system then yes, i would obviously expect some major problems :Ohno:... I cannot argue with that i suppose... .. Of course i would never even consider such an extreme pinch :no:...

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

Usually every single inline connection in an air hose line presents a little bit of restriction so it's advisable to use as few as possible. I'm not talking about stepping on the air line and closing it off almost completely, it only takes a small restriction to cause a notable drop in pressure at the end of the line.

JAG,
Right, this is very true, I agree, and it is basically related the point i have been trying to make ..... A constriction or a vent with a CSA of 300cm placed mid-path should (according to theory) generate considerably less velocity than that same vent placed at the very end of the path ... If we are to assume a linear exchange of pressure-to-velocity in a quarter wave pipe then it suggests that the velocity would be 50% of what is predicted at end-of-path if our vent is placed exactly at mid-path (50%) ...:yinyang: ... This may or may not be optimistic and i understand that it may not work out EXACTLY this way, but surely it is a general trend, simply and quite literally due to the fundamental operation of a quarter wave pipe with one end closed ..... Some Akabak experiments could perhaps tell us how the velocity will likely scale with different path placements.. 🙂 Maybe XRK would want to give this experiment a shot ...

Maybe YOU should fire up Akabak and give it a shot. Even if you've never used it before it shouldn't take more than 10 minutes to import a file and check velocity.

You can continue to rely on theory to guess what velocity might be in any given point inside the enclosure or you can simply simulate it. Your guessing doesn't have a great track record at this point, given that only one of your designs was ever built and it had audible chuffing problems at moderate volume levels.

Also, let's not forget that a few of your designs have the constriction right near the mouth so this argument doesn't even apply to those designs.

JAG ,
Ever try open baffle designs or infinite baffle? No constrictions, paths or ports involved there ...... Performance and or portability may not be the greatest in OB or IB setups but they could work just fine in some applications and there is no chance for a "chuffing" , not even one chuff .... ever ... 😉

Ha ha, clever. XRK already made that joke (about a sealed sub) a few posts back. The truth is that sealed, IB and OB are not high performance alignments, and for subwoofers I only build performance (resonant) alignments. All those alignments have their uses but high spl bass performance is not one of them.

I've built exactly one project that had a chuffing problem, I learned why that happened, and I've never had that problem again. 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.

In the time that it took to compose these two posts to argue with me you could have simulated a design in Akabak. Instead of continuing to reply with "theory says it should be fine" when it's clearly not in real world testing, take 15 minutes to learn how to analyze your design in Akabak and learn to correlate it with acceptable limits with Flare It.

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...
 
Last edited: