Porous fill moderators, any success yet?

[WobblyHand]

Here is the data plotted in the time domain three different ways.  First is linear amplitude vs time.  This is what we see in Audacity.  Second is the magnitude (absolute value) vs time.  Third is the Power vs time, where the power is 20*log10(abs(signal)). We can see some echoing, particularly at t=28.5ms.  That is due to reflections off the walls of the room.  The time delay between the main impulse and the secondary one, corresponds to the time delay of the room dimension.  We also can see that the sound power is decreasing exponentially over time.  This is not unexpected, but is interesting to see.  Time scale is in milliseconds, simply to make it easier to display.  At least with this low power shot, the sound event is attenuated by 40 dB within 1/8 of a second.  Obviously, we want to reduce both the peak and if possible, the exponential time constant.

[WobblyHand]

I was thinking about the sound issue.  Large diameter fiberglass HVAC tubing might work - that is if I can find some anymore.  Maybe 8" pipe insulation.  So far can't find anything at reasonable cost.  (Found something at $200, which is more than I want to spend on this.). Might have to go with blankets.  It seems we got rid of our old carpets, which would have worked ok as well.

[WhatUPSbox?]

Within Audacity, there is the contrast tool that gives you the RMS volume difference between two time slices. I was using it to look at differences between shot and quiet and shot to shot, and how that varied as a function of the duration. I also used the playback function for different duration samples. It helps illustrate how the perception of loudness falls off for the short durations. At the timescale of airgun shot (say <50ms) the ear appears to be more of an energy collector that a peak detector. A 1ms sample that contains the peak sounds quieter than the next 25ms.

A few youtube vids collecting quantitative data on some sound management options. Unfortunately, they tend to be audio focused and are looking for flatness as much as absorption.

I'm sure you are ready to build one of these


Some early material data


And
   

[WobblyHand]

When I was working one of the labs had a "DIY" mini-anechoic chamber for testing radars.  It wasn't good for precision measurements like the big facilities, but you could do a lot of testing in them.  About 6 feet long, maybe 2.5 ft x 2.5 ft, filled with carbon loaded pyramidal foam.  Those chambers were acoustically quite dead.

I visited the national testing lab in Beijing when I tested our millimeter wave automotive radars for Chinese compliance.  Now that was a big test chamber.  You could fit a jet fighter in there.  (They probably did.). No pictures were allowed there - and they watched you.  Curiously, they used the latest and greatest Rohde & Schwartz spectrum analyzers.  (Over $300K per copy)  I set up the tests there.  The absorber made for a very quiet room.

It wouldn't be as hard to set up an acoustic chamber, compared to a radar one.  Unfortunately, I don't have that kind of free space that I could take up.  I'm more thinking of lining a big sonotube with absorber.  Or fiberglass pipe split insulation.  Something relatively cheap and compact.

[rsterne]

I have little interest in moderators, since they are illegal in Canada.... but the idea of a variable density Gyroid fill is fascinating.... If you had four concentric tubular sections, the densest at the bore, doubling in volume and half the infill density as you go outwards, would that not be about the correct expansion?.... I drew up such a configuration, with a 6mm bore and a 40mm OD, and ended up with the following cylinder diameters....

Outer.... 6mm bore, 40mm OD, (ends up with a 28mm ID), 10% infill, overlaid by:
Next.... 28mm OD with 19mm ID, 20% infill
Then.... 19mm OD with 12mm ID, 40% infill
Inner.... 12mm OD with 6mm ID, 80% infill



I used the "Modify Settings for Overlaps" in Cura to change the infill of the inner cylinders where they overlap the largest one.... I don't know how these would hold together if you don't "connect infill lines", and/or without any perimeters (these are all without), but it certainly looks like an interesting baffle material if it prints OK....



I looked at the bottom layer with and without the infill lines connected....

Without....



With....



With only the outer two cylinders connected (looks promising)....



And here are the first 10 layers with only the outer two cylinder infill lines connected....



I you think the inner portion is too dense, you could use (from outer to inner) 20%, 30%, 40% and 50%.... The first layer (outer 2 cylinders with infill connected) would look like this....



The first 5 layers would look like this....



I think that looks workable?....

Bob

[WobblyHand]

Pretty neat Bob.  Yeah, it seems there needs to be some more work on slicer settings.  I don't know if PrusaSlicer has something similar, but I will go look for it.  It's worth playing with for sure.  I didn't know you could have modify settings for overlaps.

Variable density gyroid fill is interesting.  Wish there was a programmatic way to control the density in a simple fashion, but your shell idea seems to be a good way of doing it.

I was printing with TPU and found it a bit sticky in the sense it wanted to stick to the standard moderator barrel adapter.  I'm going to machine an aluminum insert for the moderator so I don't have to fight it when attaching it and detaching it from the barrel.  I figure I will go with a 3/4-13 thread on moderator, and a 1/2-20 on the internal threads of the adapter.  The adapter looks like the attached picture.  Might even print it in PETG for testing, why not.

[rsterne]

You can't "stack" multiple settings, you start with a basic setting (in this case 20% for the outer, which is actually the whole thing), then use concentric tubes, modifying each one to a higher infill (30%. 40%, 50%) as they get smaller.... Those settings then replace the 20% infill in each of the smaller tubes, changing from one to the next automatically.... Pretty slick!....

Bob

[subscriber]

Very nifty idea, Bob

If the tubes overlap, that means their patterns overlap and are additive, right?  If that might cause air passages to choke off then the tubes could be configured to just touch.

I think you are right about your first model being a bit dense around the center bore.  But then, as you said, that is a matter of choosing the most appropriate infill density; whatever that is.

If you can use a simple cylinder or tube to demarcate the gyroid infill, I presume it is possible to use more complex shaped, such as the "tube" shown below?  STL attached.

[WobblyHand]

@subscriber, you might consider downloading a slicer of your choice and playing with it some to find out some of the options.  They are a lot easier to use than a CAD program, although they are more limited.  That's all I have done to (slowly) figure out some of this stuff.

[subscriber]

Bruce, that would make a lot of sense, if I was a printer guy.

I do have a version of Cura that I run my designs through as a design verification.  I trade in STLs and STEP files, rather than pre-sliced files.  So, I live vicariously through people like you and Bob, that post threads such as this one.

[WobblyHand]

Bruce, that would make a lot of sense, if I was a printer guy.

I do have a version of Cura that I run my designs through as a design verification.  I trade in STLs and STEP files, rather than pre-sliced files.  So, I live vicariously through people like you and Bob, that post threads such as this one.

I'm not a printer guy.  Actually 3d printers are a pain in the behind.  Like just now, when I had a print fail.  Some piece of unpurged PETG came out and globbed on my TPU print.  The print head hit the blob which was above the current Z height, (0.86mm to be exact) and knocked over my print after 5 hours.  The print was on layer 609 out of 742.  Pretty disappointing. 

[subscriber]

Well, Bruce; I am so far behind you when it comes to printing, I depend on those who have fought their way through it.

I am reasonably able to generate concepts in my head and then translate them to 3D CAD parts.  I am self-taught using Solidworks, and at the beginning it was a struggle.  It takes motivation to gain proficiency with any of these tools.  I am impressed with Bob's history of mastering 3D printing that is documented on this forum.  Ditto Mike in TX (miksatx); especially considering his printer started out with hardware failures.

You seem skilled with printers from my perspective, although I did not witness the process.  Ditto for the other Bob (TorqueMaster).  I have the greatest confidence in his prints with respect to critical dimensions, by shear volume of functional prints he has produced of my designs. 

I try to be useful.  To do that I need to understand other disciplines enough to communicate with the experts in those areas.  I do not need to match their skill.  If I am talking to someone in Greece who understands English, I do not need to learn to speak Greek   

[rsterne]

sub, my first try overlapped the tubes, and the slicer picked one and ignored the rest, so I ended up with 20% in the outside and 40% for the rest.... I had reasoned (from past experience) that allowing a perimeter on a tube would print a solid wall on both the inside and outside of each tube, so that was a no-go as it would prevent airflow past the wall.... Therefore, all the tubes have no perimeter (wall), except where you choose to use the "connect infill" function, which prints a line from the end of the infill line to the start of the next one, so a "partial wall" (but with holes to permit airflow to the next layer out).... I don't think you need that on the inner tubes, but I think it is a good idea on the outer ones where the infill lines are quite a distance apart (see the infill pic above labelled "Without")....

For the drawing you posted in Reply #27, I could load it as the "main body" at the lowest infill ratio, and then add progressively smaller tubes concentric with it, of more dense infills.... I think all of the pyramids would be the more dense, and probably divide the outer tube into two densities, if not just leave it at the lowest one.... Do you envision that large a bore, or smaller, closer to the calibre?.... The pyramids could also have a "ring" (short length of wall) near where they join, to add support if necessary, but it would take a lot of work....

Bob

[subscriber]

Bob,

Strengthening the outer wall of the low density gyroid with a solid "skin" is probably a good idea to make the part less fragile.

The "baffle" I posted was created as an alternative to the stock Talon pistol baffle spacers, stamped from foam rubber, shown below.  Not the baffle with the caliber sized hole, but the cloverleaf shaped expansion chamber section.  So, technically, not a baffle; but a baffle spacer.

The reason for me asking was I would like to know if the gyroid infill could be made to follow such a complex angular surface, as a single part.  Making it as part of a multi-density layered foam section was not really my goal; although someone might see utility in that.

[rsterne]

I downloaded your file, and used 20% Gyroid fill for the outer section and 40% for the pyramids.... Here are the results....

First Layer....



Section through top of lowest pyramids....



Complete, view from top....



I don't know what the structural integrity would be like, it would be print and see what happens.... Time about 3.5 hrs.... I was not able to add a single wall to the outside, but could probably do it by adding a very thin tube.... or print one out of PLA and glue the TPU foam baffle into it....

Bob

[subscriber]

That looks brilliant.  Thanks Bob.

I think that making the pyramids denser should help their durability.  I am impressed that you can differentiate them from the rest of the part.  I assume you created a tube outer at the lower density to achieve that?

Do you have and AGN account?  If not, do you mind if I post your images there?

[rsterne]

I don't have an AGN account, please feel free to post them there.... I loaded your file and told it 20% Gyroid fill and no walls or top and bottom.... Then I superimposed a 35mm OD cylinder over the center with pyramids and told the slicer to use 40% infill for the cylinder and merge the meshes....

Bob

[subscriber]

Thanks, Bob

[denovich]

You found the slicer settings I would recommend (using geometric shapes to modify portions of a print.)

I eventually changed to a solid print with different baffle designs... as the TPU itself seemed to embody much of what the mesh was trying to achieve.

[subscriber]

denovich,

I think the actual baffles should be impervious to air.  Their function is to forcefully redirect air flow.  Coating the surface with gyroid foam to to absorb reflected sound waves or add turbulence would seem useful, but should not require the foam layer to be very deep.  Coating the cylindrical portion of the muffler inside with foam could fulfil both these functions.  Making the layer deeper may act as lossy expansion volume where the air lingers, and is damped going in and out.

That said, I wonder if a foam layer could act as a lossy "wall" to a secondary open expansion volume, rather than filling most of the expansion volume with foam.  I suspect that muffler foam is like salt on your food.  A small amount in the right places may work better than using as much as you can fit in.

Your results with TPU would seem instructive.  Certainly, a slightly flexible structure that leaves open most of the air expansion volume may be the ticket.

The design shown below was derived from my oval offset Huben pistol muffler.  The latter works rather well, and early indications are that the cylindrical derivative is equally good.  Note the perforations that feed secondary internal expansion volumes.  The perforations are rounded on the inlet side and sharp on the reverse flow side, for differential flow resistance.  If gyroid foam could improve this design, I suspect it would be just surface treatments, rather than stuffing most of the open space.

Perhaps gyroid foam would come into its own when used inside something like a three chamber hair curler style Weihrauch muffler.  And then not over filling the space with dense foam.