A Monster Construction Methods Shootout Thread

[augerpro]

I can increase the port diameter of the Precision port to 57.9mm (from 50.8mm) to match the length of the Harman. Will check it out.

EDIT: I'll also do a shrunken Harman port to see how it scales. I need to determine the end correction for future builds so it is not iterative tuning which would require a dummy box since the mouth diameter scales up and down with the length forcing you to keep re-cutting the hole in the box.

[augerpro]

I know these are sort of data dumps, and unless you open plots in separate browser tabs to flip back and forth and comparing them, not so easy to digest. So maybe a summary of what I'm seeing so far. I should add my observations apply to any woofer duty that is not a subwoofer. For subs, it seems to me 1/2" or 3/4" plywood well braced is all you really need.
1) MDF is actually pretty good. I would use it over plywood for anything except a subwoofer
2) simply gluing two pieces of MDF together with a softer adhesive like Weicon 310M Flex Classic or Loctite PL300 works really well as a simple CLD
3) a sandwich of MDF and Nidacore or cork works well too (again glued with the 310M Flex or PL300) but I'm not sure most people will want to deal with the hassle
4) CLD bracing with 33% overlap has a small edge over solid bracing, but any bracing at all really drops that main box resonance
5) melamine foam or 100% wool batting is the way to go for lining the box
6) Resonix is a nice cherry on top if you can swing the extra cost

Things that were not impressive: XPS foam in a CLD, hard polyurethane adhesives like PL Premium, crappy plywood, coating the interior with any supposed vibration absorbing compound.

[Steve Manning]

I know these are sort of data dumps, and unless you open plots in separate browser tabs to flip back and forth and comparing them, not so east to digest. So maybe a summary of what I'm seeing so far. I should add my observations apply to any woofer duty that is not a subwoofer. For subs, it seems to me 1/2" or 3/4" plywood well braced is all you really need.
1) MDF is actually pretty good. I would use it over plywood for anything except a subwoofer
2) simply gluing two pieces of MDF together with a softer adhesive like Weicon 310M Flex Classic or Loctite PL300 works really well as a simple CLD
3) a sandwich of MDF and Nidacore or cork works well too (again glued with the 310M Flex or PL300) but I'm not sure most people will want to deal with the hassle
4) CLD bracing with 33% overlap has a small edge over solid bracing, but any bracing at all really drops that main box resonance
5) melamine foam or 100% wool batting is the way to go for lining the box
6) Resonix is a nice cherry on top if you can swing the extra cost

Things that were not impressive: XPS foam in a CLD, hard polyurethane adhesives like PL Premium, crappy plywood, coating the interior with any supposed vibration absorbing compound.

I'm being lazy and not going back through things ...... is the MDF that you've been testing, the big box store variety? I'm curious since I've been getting what's considered "cabinet grade" for a while. It's a finer structure and denser.

[augerpro]

Yeah nothing special as far as I can tell

[sfdoddsy]

As someone who is currently building something slightly more complex than my usual variation on a flat actively EQed open baffle, I tip my hat for the effort on this. And, of course, your many previous endeavours.

But I’m a bit hung up on practically.

The differences in the various materials, braces and linings are certainly measurable. As an objectivist leaner this is good.

This site is ground zero for totally over the top builds. Massively thick walls. Intricate bracing. CLD upon CLD.

Your numbers back up this approach.

But arevwe falling down the same rabbit-hole which has consumed folk over at AudioScience where they obsess over vanishingly small differences in SINAD?

Again, I am coming from open baffle land where the only good box is no box.

[augerpro]

I agree there is a point of diminishing returns, but I think we are far from that when some of cabinet responses are nearly as high as the driver itself! And a lot are 10-20dB down, which most would still consider an issue.

[Bear]

Yeah nothing special as far as I can tell

This would really be the cherry on top for this topic: does extra money spent on materials actually improve performance? Since you have a fairly dominant strategy identified based upon the tests, it would be interesting to see premium MDF and Baltic Birch vs the ordinary stuff (I believe that you were using box store plywood). Given that BB and cabinet-grade MDF are both considered the go-to materials for cabinet builders and woodworkers, it would be interesting to see if there was a sonic performance benefit to the aesthetics/construction benefits.

[augerpro]

I didn't use premium MDF, but in the first round did use expensive Russian Birch ply. It performed well, but nothing that would ever make me buy it again when the real gains were going to even a simple CLD construction.

[Concillian]

This would really be the cherry on top for this topic: does extra money spent on materials actually improve performance? Since you have a fairly dominant strategy identified based upon the tests, it would be interesting to see premium MDF and Baltic Birch vs the ordinary stuff (I believe that you were using box store plywood). Given that BB and cabinet-grade MDF are both considered the go-to materials for cabinet builders and woodworkers, it would be interesting to see if there was a sonic performance benefit to the aesthetics/construction benefits.

Extra money for materials first needs a definition of what is important so you can create a metric. So far it looks like the assumption Is dB relative to fundamental.

I'm sure there are secondary metrics like energy storage/CSD or even frequency of fundental panel resonance (and design consideration to vary fundamental resonances for each panel to avoid summation.) But if we start at just dB from fundamental vs. $, the pl300 with MDF looks very attractive for panel resonance, then home depot insulation for damping internal reflections (I can't find r6.7 denim anymore, looks like mineral wool now makes sense) Everything else is MUCH higher $ for small benefit, though if resonix works with the CLD panels, its still expensive, but looks to offer a pretty consistent 5dB less, which should be noticeable and is very easy to just stick on.

the bonus of the CLD construction is that if you do want higher aesthetic wood, like BB, you only need a 1/4 inch skin with MDF on all the rest

[augerpro]

I'll post full results this weekend, but just wanted to share some early results of my tapped port for dampening the port resonances. The port is based of my Precision port clone. I added square taps around the centerline. First plot compares results with taps open, then wrapped with wool batting, then duct tape over the wool to create a more or less sealed cavity. Second plot compares the final taped version vs the standard Precision port. The resonance at @1.3khz is the predicted first resonance. The broader resonance above that to about 3.5khz is a little unaccounted for.

[augerpro]

Here are the results of the new port test. Here I wanted to investigate a few things. One was abandoning the Kef method of flexible center sections to reduce pipe resonances, and go to tapping the center of the port and then dampening that energy. I also wanted to investigate whether the performance of the Harman port was inherent, or whether it was simply due to the average port diameter being larger than the others (thus it being longer than all the others also). I did this two ways.
First by downsizing the Harman port so it matched the overall length of my Precision port clone. Port ID ended up at 1.756". Since I hadn't yet determined an end correction factor, I had to make an educated guess and the final port tuning was high by about 2-3hz. Not bad.
Second, I upsized my Precision port clone so that the overall length was the same as the original Harman port. Port ID is 2.28". Tuning was virtually identical.
I also measured an actual 2" Precision port to see if surface finish made a difference. If you recall I 3D printed all of these to remove the surface finish variable, since that is known to have a measurable impact on performance.
Additionally I added a 2.38V harmonic distortion plot to better capture typical listening conditions.

Harman port (shorty)















Precision port (jumbo)















Precison port (factory smooth finish)













Tapped port

I should note that I heard (and measured) a "flapping" sound during the harmonic distortion tests. No doubt the taped cavity being pumped back and forth. So treat these HD results as preliminary until I can print the dampening cavity, so it is sealed with hard material.

[augerpro]

Some observations...

The Harman port is superior. It's a bit ahead in distortion performance for the original vs jumbo Precision port, and more definitely in the shorty Harman vs the original Precision port. Behavior at the very highest levels is where the difference is pretty large.

Surface finish matters. Quite a bit. The harmonic distortion of the actual smooth finish Precision port is cleaner than my clone, and is really noticeable at the highest power. Not enough to catch up with the Harman though, but close. Close enough that smoothing the Harman print is well worth the effort. I've seen if you print with ABS you can smooth the finish with acetone.

For some reason the F1 was not quite as smooth in the distortion plots this time around so it is not as obvious when chuffing starts when just looking at F1. The higher harmonics are still indicative though. Because of the difference in F1, power compression vs the earlier testing is a little harder to see.

[augerpro]

Just wanted to share a fantastic paper. Harman's must be based on it. I actually like this paper's method better than Harman's for exactly the reasons that I diverged from Harman: normalizing based on tuning, not length, and later on length but with varying diameter so tuning was still the same. Curiously Roozen's worst performer (D) superficially looks like mine, BUT Harman did several variants where the changed ratio of length vs wall curvature radius. The one I selected was called N=0.5, which was sort of in the middle. At the high end of N there was much more curvature, and at the low end it looked pretty similar to Roozen's E. So I'm guessing his D was more like the N=1 port from Harman's paper. Still he specifies the optimal contour as like Harman (and his own D) but where curvature from port center to exit is not faster than 6 degrees. I'm going to try that, and tapping the port at 1/2 and 1/4 length.

[augerpro]

EDIT: N is actually NFR in Harman's paper

[augerpro]

I hope to have some new testing soon of new ports and CLD cabinet construction. I'll have one port using the Roozen profile mentioned from the paper above. Assuming that does not outperform the Harman inspired port (if it does I'll use the Roozen profile instead) I'll have a new hybrid combining the Harman profile with taps at 1/2 and 1/4 wavelength that will be filled with stuffing and sealed using some plastic strips to absorb the primary port resonances. I dub thee: AugerPort!!

[neuro]

Excellent ongoing work Brandon. First reaction to the AugerPort was "lossy port, why would someone do that?" Reading on, I see the plan and am very curious for the result.
Since you are printing ports, how about other cross sections, maybe a Koch snowflake or something else that should make standing waves difficult.

Here is a TY video I came across that those following this thread will probably like: https://www.youtube.com/watch?v=n3wlmaENJio . Its a simple enclosure comparison between several woods and two PLA infill densities.

[augerpro]

I believe the resonance is set up by the impedance change at the ends of the port, so it can't be prevented just with the shape. Have you see someone do that?

[neuro]

I was imagining that one source of resonance is between the parallel walls of the port. A triangular port could be a simpler way to test if parallel walls are an issue.

[---k---]

Brandon,
Very cool work. This is the type of testing and experimentation that I really enjoy reading about. I've consumed both your thread here and at DIYaudio. I think it is really great work.


I want to make sure I understand your method and graphs. In this graph, you've 1) taken the perless driver and mounted it on the top of the box pointed out towards you and then measured the "bare driver" result. then 2) you turned the same driver around pointed into the box and drove the response at the same voltage level as you did pointed out. 3) you measured the response of 1) and 2) and graphed. And, so at like 220hz, the volume of sound being resonated off the side and rear is only 5db lower than the bare driver.



That is really hard for me to grasp. I know it is only at that one frequency or so. But, if you're standing 5 feet away, and run a frequency sweep does it sound similar? Like you just turned the volume knob down 5db? Is it even less than this, since you got six surfaces radiating and each is down 5db? My logarithmic math is in my brain cellar, but if you start adding both sides and back together, wouldn't it exceed the bare driver? What am I missing here?


Some other thoughts I have:

Within categories, as more of builder and not a designer, I find the results to be difficult to find a clear, obvious, run away winner. The differences tend to be very small, and at some frequencies one material will be ahead and then at others another will be. I wonder if it could be more obvious by increasing the graph scale some - maybe just changing the max db to 110db? I also wonder if there is any way you could produce a diff graphs, where the results are just the differences between the two. I also think about whether it would be accurate to integrate the areas under the curves to get a total spl measured across all frequencies and then if it would be fair to compare those results.

I'm getting older, less time, and generally lazier. I've long known about the benefits of CLD, and your testing proves it. I enjoy reading about it, but won't undertake it in my next build. I'd love to see testing on some of the less time consuming methods, tricks. A couple that I've wondered about: 1) Glues. I typically use Tightbond III. I've seen Jon use structural epoxies for joints. Hummm... 2) Bracing. You used oak dowels. I often use plywood or MDF scraps for my bracing. Does this make a difference. What about a oak/MDF combo? 3) Minwax wood hardener. https://www.minwax.com/wood-products...-wood-hardener This was something I started to experiment with years ago, but never formally tested. It will absorb in MDF good and creates a hard exterior layer. I used it on some box somewhere I've forgotten and swore it improved the knuckle test. Cheap and easy to do. I wish to believe that it would be like laminating 5/8" MDF laminated to 1/8"HDF. 4) What about Jon's bamboo cutting board baffles.

I'd be interested in seeing overlay graphs comparing 1/2" to 3/4" material. Looking at your graphs, in the lower frequencies it looks like the 1/2" is better. And maybe an overlay of 1/2" w/ brace vs 3/4".

It'd be interesting to construct one or two more of the easy boxes, like the 1/2 ply or 3/4 ply or mdf boxes and see how repeatable the tests are. Even just doing the same box three days in a row, moving the box and mic in between would be interesting.

Thanks again for doing this. It is interesting.
If you're up for testing some of these, I can put some money into your gofundme.

[augerpro]

Great questions Ryan! Hopefully I can answer some satisfactorily. Regarding the bare driver vs in-box, the method is more involved than that, but the end product is as you understand it, the relative SPL levels are as shown in the plots. From where I sit it sounds like hash with a midrange tone overlaid on it. So it sounds pretty much how the plot looks. It surprised some people, but not those who have read the previous papers from the BBC and other well known researchers from the 60's and 70's. The explanation as I understand it is that at the panels primary resonance it is basically transparent. I'm no material engineer but that sounds about right to me! As far as how all the panels together are radiating and then summing at some measurement/listening position, I just can't say for sure since I'm not measuring that. But I have seen papers that show the basic character of the panels is essentially what is measured farther away where all the panels output has summed together. And as I said, just from my uncritical listening when doing the measurements, it sounds like there is a resonance problem for sure that needs some solving.

Regarding the frequencies shifting for the primary resonances, one thing that needs to be recognized is relative stiffness really defines where the peaks land. In the first round where I'm just getting baselines of simple materials, you can see MDF (for a given thickness) shifts the frequencies lower, which makes sense since it is not as stiff as plywood. Same with 3/4" vs 1/2" of any given material, the 1/2" shifts the primary resonances lower as it flexes more easily. So you have to make that mental correction. Bracing does something similar, but it actually drops the SPL of the main resonance also. And that is what I would look for overall, have the worst peaks dropped in level? That is the better material.

That does bring up a discussion though about what happens when you have a favorite material/method and then add bracing. I plan to measure that soon. For example, one thing that may come out of this is say you had a pure midrange, only to be used >300hz. You may like the 1/2" MDF best as its primary resonances are near or outside the lower cutoff of the passband and maybe you should NOT add bracing where it would push those resonances up into passband. The construction technique might be different for different passbands. We already know this on subwoofers, simple plywood boxes of 5/8" plywood with heavy bracing are what is called for and CLD would probably have little benefit.

I should have, either here or on my website comparisons of the same material at different thickness, and different materials at the same thickness. The differential SPL is something Soundeasy can do and I have thought about. Once some best practices emerge i'll try that.

I use Titebond III for building the boxes too, and will continue to do so. But if it is CLD, you really need something like the PL300 or Weicon for the constrained layer. Different adhesives for different jobs.

I've always used oak dowels because they are very stiff, but from my testing I don't think the difference is significant anymore. For example the first CLD braces (95% overlap) are really not much different than the oak control. I can't imagine an MDF brace would be any better than the CLD brace just due to stiffness! Now there may be something to be said about at least using dissimilar materials, say, plywood bracing on an MDF box or vice versa so the natural resonances of the materials work against each other. Might have to check that out at some point.

I will circle back to measuring best practices vs some standard methods when I get there, sort of a summation of the work as it is easy to lose the thread of incremental improvement here and there. So far though I do think some promising methods have emerged, the simple CLD with PL300 or Weicon and MDF behaves pretty well. I need to see how adding bracing changes that picture now. If is still good, I would say at this point that method with simple bracing and Resoniz added would be pretty dang good and not terribly expensive. And easy to build. Just build an outer box as normal, then attach the inner material one panel at a time to the inside. Pretty on the outside, ugly on the inside but who cares?

I do want to investigate other materials like the bamboo wood, but the gofundme for this project has not even paid for 1/3 of what I've spent so far and that stuff is expensive. My next round (setting up today ectually) is using materials I already had on hand.

[augerpro]

Missed your repeatability question. This project has been by far the most sensitive to changes between measurements. Usually I have virtually no change, but then I always use very standardized methods. With the box measurements it is critical you are aimed at the same spot each and every time. I mark the exact center with a circle just bigger than the tip of the microphone so I can get good repeatability. Still the area above @ 1.5khz will wiggle a little between setups. But nothing that changes the character, and most importantly the area below that stays pretty stable. As I mentioned in thread I was forced to change environments between the two rounds of port testing and there was a bit of a change to the HD because of it, but it will be the same going forward. But yeah, repeatability took some planning and experiments up front, this wasn't a typical measurement environment that I would use designing speakers that are quite stable for me.

[Steve Manning]

Great questions Ryan! Hopefully I can answer some satisfactorily. Regarding the bare driver vs in-box, the method is more involved than that, but the end product is as you understand it, the relative SPL levels are as shown in the plots. From where I sit it sounds like hash with a midrange tone overlaid on it. So it sounds pretty much how the plot looks. It surprised some people, but not those who have read the previous papers from the BBC and other well known researchers from the 60's and 70's. The explanation as I understand it is that at the panels primary resonance it is basically transparent. I'm no material engineer but that sounds about right to me! As far as how all the panels together are radiating and then summing at some measurement/listening position, I just can't say for sure since I'm not measuring that. But I have seen papers that show the basic character of the panels is essentially what is measured farther away where all the panels output has summed together. And as I said, just from my uncritical listening when doing the measurements, it sounds like there is a resonance problem for sure that needs some solving.

Regarding the frequencies shifting for the primary resonances, one thing that needs to be recognized is relative stiffness really defines where the peaks land. In the first round where I'm just getting baselines of simple materials, you can see MDF (for a given thickness) shifts the frequencies lower, which makes sense since it is not as stiff as plywood. Same with 3/4" vs 1/2" of any given material, the 1/2" shifts the primary resonances lower as it flexes more easily. So you have to make that mental correction. Bracing does something similar, but it actually drops the SPL of the main resonance also. And that is what I would look for overall, have the worst peaks dropped in level? That is the better material.

That does bring up a discussion though about what happens when you have a favorite material/method and then add bracing. I plan to measure that soon. For example, one thing that may come out of this is say you had a pure midrange, only to be used >300hz. You may like the 1/2" MDF best as its primary resonances are near or outside the lower cutoff of the passband and maybe you should NOT add bracing where it would push those resonances up into passband. The construction technique might be different for different passbands. We already know this on subwoofers, simple plywood boxes of 5/8" plywood with heavy bracing are what is called for and CLD would probably have little benefit.

I should have, either here or on my website comparisons of the same material at different thickness, and different materials at the same thickness. The differential SPL is something Soundeasy can do and I have thought about. Once some best practices emerge i'll try that.

I use Titebond III for building the boxes too, and will continue to do so. But if it is CLD, you really need something like the PL300 or Weicon for the constrained layer. Different adhesives for different jobs.

I've always used oak dowels because they are very stiff, but from my testing I don't think the difference is significant anymore. For example the first CLD braces (95% overlap) are really not much different than the oak control. I can't imagine an MDF brace would be any better than the CLD brace just due to stiffness! Now there may be something to be said about at least using dissimilar materials, say, plywood bracing on an MDF box or vice versa so the natural resonances of the materials work against each other. Might have to check that out at some point.

I will circle back to measuring best practices vs some standard methods when I get there, sort of a summation of the work as it is easy to lose the thread of incremental improvement here and there. So far though I do think some promising methods have emerged, the simple CLD with PL300 or Weicon and MDF behaves pretty well. I need to see how adding bracing changes that picture now. If is still good, I would say at this point that method with simple bracing and Resoniz added would be pretty dang good and not terribly expensive. And easy to build. Just build an outer box as normal, then attach the inner material one panel at a time to the inside. Pretty on the outside, ugly on the inside but who cares?

I do want to investigate other materials like the bamboo wood, but the gofundme for this project has not even paid for 1/3 of what I've spent so far and that stuff is expensive. My next round (setting up today ectually) is using materials I already had on hand.

Hey Brandon, I don't know if you've looked at Wilson Audios construction methods at all, but they tend to go with different types of material for different driver ranges. Of course they don't publish much of the data, understandably, but going through their videos you can pick out some good info. Falls very much in line with what your looking at.

[---k---]

Brandon,
Thanks for the detailed response. It really helped clarify things. I totally get it.

And thanks for coming back and responding to the repeatability. I know how hard repeating measurements can be. Sounds like you got a plan for that. My real question, and how do I say this without sounding like an ass, I was more wondering about the construction side. I'm more of a rough carpenter than a finished carpenter. I know that no two boxes of mine would be the same. So, and this is just a scientific curiosity that anything, I'd be interested in seeing how close the result would be from three assumed identical boxes... variations in plywood? variations in how perfect the joints were cut, or variations in amount of glue or how it was clamped. I'm guessing they'd be pretty small with your construction (are you cnc'ing the panels?), but its something that just sticks out to me and I'd like to see proven. BUT,

BUT, I know this is probably 2x to 3x them amount of undertaking I'm imagining. and I'm assuming pretty big. Its unreasonable of me to ask. But, if you were testing something else where the opportunity presented, it may be interesting. Like, if you did another round of CLD and built 3 exterior boxes and before you installed the inner panels, test them... whatever. Just a thought.

[augerpro]

It's something I've thought about and I just don't have a real good answer. For example, that is one reason I tried to make sure I used a certain trowel so the constrained layer thickness would be the same for each build. And I apply them like tile were you push across the "grain" of the beads so the beads collapse and expel air. I'm making sure to communicate those details so if anyone uses one of the methods they can be sure to get as close as possible to how I did mine. Hopefully that is enough.

[---k---]

Totally. Don't take anything I say as negative to what you're doing. It's more than I'm doing. If I really motivated, I got two .25cuft knockdown boxes on the way for rebuilding some rear speakers. I could easily do my own experiment. But, I probably won't. So, I'll donate to you and maybe someday these questions will get answered. :T

[augerpro]

Much appreciated Ryan!!

[augerpro]

Started another round of testing. I hope to have them all done within a week. For now here are the port results. I tested two models: the my Augerport and another based on the Roozen paper attached earlier.

The Roozen port is similar to the Harman - it appears that Harman got the constant curve contour from the Roozen paper - but with a much shallower flare as he specifies an angle of <6 degrees from center of the pipe to the exit.

My Augerport is in 3 variants. 1 = taps are aligned, 2 = taps are offset, and 3 = taps offset and bigger. I used wool batting to damp the taps energy.





First the Roozen results:

















Now the Augerport variants. Augerport 1:















Augerport 2:















Augerport 3:















Some comparisons:

[augerpro]

Couple observations so far...Augerport 3 really knocks down the port's 1/2 and 1/4 wavelength resonances. At this point I though it was a slam dunk - until I started harmonic distortion testing. Performance degraded to similar to the Precision port clone and worse than the short version Harman on which the Augerport is based. Chuffing is noticeable and comes on soon. I slightly chamfered the taps and I think this was the wrong thing to do. I should just leave a clean hole. Larger holes were obviously better for knocking down the resonance but it seemed like I heard chuffing sooner than the other two variants even though the HD plots look similar.



What do you guys think I should try for taps now? I'm temped to do a slot, maybe broke up into 3 segments. Maybe increase the void/wall ratio. Has to be something clean though. Also, I only put the one line of taps for the 1/4 wavelength from one end, I think this should damp the wave, but maybe I should try taps at 1/4 and 3/4 length?



The Roozen also excited me at first with its best-in-class sensitivity. I thought I might switch the Augerport from the Harman profile to this one. But harmonic distortion was really unimpressive.

[tktran]

Fantastic work on the AugerPort. Count me in as someone who would buy an AugerPort Pro XL.

Maybe I missed it, and not to derail this thread- but have you seen studies on how different various passive radiators compare?

[augerpro]

Fantastic work on the AugerPort. Count me in as someone who would buy an AugerPort Pro XL.

Maybe I missed it, and not to derail this thread- but have you seen studies on how different various passive radiators compare?

It would be interesting to compare with passive radiators. Down the road I'll have to check that out. The overall HD and chuffing performance of the Harman profile is so good, and the tapped port's ability to damp pipe resonances, that I REALLY want to figure out how to combine them together. Once I finalize the Augerport I'll post the .f3d file and anyone can create their own version to print.

[augerpro]

If anyone wants to play with the Harman-inspired port, I've attached it. It is based on the NFR=0.5 example from their paper I linked to earlier, probably the first post. You can see you only have freedom with the port diameter and length, the exit size will scale up and down depending on those choices.

The end correction is 1.15, I know that seems weird but it behaves as a port with a larger diameter, thus end correction over 1 to get proper tuning. Some softwares use a preselected EC, so be sure to manually set the value to 1 when modeling, then apply the 1.15 correction to whatever length you settle on for a given tune. That's one thing I like about Soundeasy, it applies no EC, leaving that to the designer to know what to use.

While modeling you might decide you need, say, a 3" port for air velocity reasons. If you use that for the Harman you will have a rather long and large port (EC over 1 remember), but the cool thing is if you shrink the diameter by 6% you will have a port the same length as a more ordinary example, but still with superior distortion/chuffing performance.

Surface finish matters a lot for harmonic distortion, best results will be by printing in ABS then doing the smoothing trick with acetone vapor.

Once you've settled on a port diameter and length, open the .f3d file and in "sketch 1":
1) set the port radius
2) set the port length

Enjoy!

PS: when I settle on a final design for the Augerport will post the .f3d file and anyone can use it for DIY purposes.

[augerpro]

Toying with different size taps

[augerpro]

Versions 4 and 5 of the AugerPort are in!

AugerPort 4 results:

















AugerPort 5 results:

















Comparing the AugerPort 5 versus the Harman contour it is based on shows the huge improvement in knocking down the pipe resonances:



The issue with all of these is much worse harmonic distortion. To add some data points I did a simple single tone spectrum analysis and you besides the harmonic distortion products, you can see broadband noise in from 500-5000hz that is not present in the Harman, and broadband noise in between the first 4 harmonics that only become an issue in the Harman at higher voltages. Making progress so I'll keep working the tap shape.

[ergo]

Such interesting work and results. Thanks Brandon again for providing the data.

Looking at the cutouts and wall thickness I'm wondering if you'd make these holes by replicating "punch of donut shapes" on circle and then cutting into pipe with each edge of donut just passing the material from inside out.... it would perhaps create these slits that have potentially smoother air travel through the port?

[augerpro]

Can you illustrate what you mean?

[ergo]

A quick mockup reveals that the donut needs to be pretty big and my description of donuts in circle was not the best... but anyhow, a slot with edges that should appear less sharp for wind passing through was what I had in mind...

[augerpro]

I chamfer and roundover the backside of the taps:

[augerpro]

Tried a new CLD box using 1/4" Hardie backer board, a type of cement and paper composite board. Similar to earlier simple CLD boxes I did, the Hardie board was the internal layer, 1/4" MDF as the outer layer, glued together with Loctite PL300. I reran harmonic distortion for some earlier boxes that tested well because I couldn't remember how my test rig was setup. This time I included the THD% so you can just compare a value instead of the mental gymnastics of finding the difference between two wiggly lines.

[augerpro]

Tried a couple more ports designed to dampen the pipe resonance. While I've made some great examples that really do the job well, all of them suffer from much increased harmonic distortion and chuffing. Future designs will look into the newest Polk port that uses closed pipe in the center of the port that is tapped in the center to absorb that resonance. Polk claims it actually improved the HD performance even though that wasn't a goal. We'll see.

AugerPort 6
















AugerPort 7

















Comparing to best performer AugerPort 5 and the Harman port they are based on:

[tktran]

Hi Brandon,
I thought I’d bring your attention to Lars Risbo’s (Purifi) post over at DIYAudio, where he shows a FEA simulation and practical implementation of suppressing port resonances using a damped stub in the middle of the port:

https://www.diyaudio.com/forums/multi-way/352063-exploring-purifi-woofer-speaker-builds-post6699757.html

[Zvu]

The problem i see with implementation of damping resonances that way is that Lars measured distortion performance on regular straight (read bad) pipe. Harman port is kilometers ahead of that kind of port. Since distortion of straight pipe port is naturaly very high, the differences in distortion profile are are small to non existent and the resonance is successfully solved. Harman port, on the other hand, has very low distortion and what ever you do to it's profile makes it audibly and measurably worse.

[augerpro]

Huh, I never would have thought a big tap the same diameter as the port wouldn't disrupt airflow and increase harmonic distortion and chuffing, but his measurements do seem to support that. I'll have to look into it. As we speak I have a Polk inspired port on the printer, so that may be another path forward.

[augerpro]

Over the next few weeks I'll be posting more testing, but I wanted to share the two ports I'm working on. One is inspired by current Polk thinking and uses a resonator suspended along the the axis of the port. The second was inspired by a post from Nissep on microperformations of a sound absorber/resonator. They define microperforation as < 1mm, mine are .75mm.

[augerpro]

New port designs were measured today. AugerPort 8 is similar to my earlier designs, except that instead of large-ish holes in the wall, here I used microperforations, defined as being <1mm in diameter. Mine are .75mm. The hope was that with no large holes the chuffing and harmonic distortion would be closer to an untapped port. Similarly, AugerPort 9 is inspired by Polk's new X-port, which uses a suspended closed-end pipe along the port axis with taps in the center to again absorb the half-wave resonance of the port. Polk claims chuffing was actually improved.



AugerPort 8 results:





















So damping of the pipe resonances was good, 2nd best to AugerPort 5. The hissy sort of chuffing was better too than prior attempts, but there was a distinctive "burrrr" sound at all power levels. You can see in the single tone spectrum plots how lit up it is. I don't see how I can improve this situation, so that's disappointing.

Next AugerPort 9:





















As far as damping the port resonance, this design did absolutely nothing. The resonator diameter is 1/4 that of the port, but watching Polk's video again it looks more like 1/3 or even 1/2 the diameter of the port. I'm going to resize to 40% of port diameter and try again.

Some comparisons, including with AugerPort 5, which had great damping, but poor chuffing as all these tapped ports do.

[augerpro]

Next are some new brace schemes. One is "matrix" style suggested by others. The other is similar to Kef's CLD brace where the place the damping material between the box and brace. I used the dowel brace here with 3M VHB tape as the constrained layer. But given the muted results vs Kef's simmed results, I wonder if using something like a window frame brace with the VHB tape would be better? More contact across the panel to absorb more energy? Also included are some comparisons with earlier tests: regular oak dowel bracing and CLD bracing where oak braces are overlapped 33% and glued together with Weicon 310M Flex.

Matrix bracing:





CLD using VHB, Kef-style:



Some comparisons: