1/2 - 20 Moderator long and short versions

[subscriber]

Thanks  for figuring out how to do that edit, Bob,

That adjustment of yours is definitely much more appropriate now in terns of the outer wall strength.  It is possible that the first baffle - below where the break occurred in Jared's photos should be thicker.  It is slightly tricky to be sure if that is the first obvious conical baffle near the top in your section image; or part of the unusual "holy" design further up.

I went with ~3.2mm at the entry and ~1.6 at the exit.  Made the exit cap 4mm.  Can be "easily" re-adjusted now that I have the edited pieces in Tinkercad.  Sliced in half here and a random height cross-section.   The layer view looks good in Cura, it should be printable.  Let me know if anyone sees any issues.

[TorqueMaster]

Thanks  for figuring out how to do that edit, Bob,

That adjustment of yours is definitely much more appropriate now in terns of the outer wall strength.  It is possible that the first baffle - below where the break occurred in Jared's photos should be thicker.  It is slightly tricky to be sure if that is the first obvious conical baffle near the top in your section image; or part of the unusual "holy" design further up.

I went with ~3.2mm at the entry and ~1.6 at the exit.  Made the exit cap 4mm.  Can be "easily" re-adjusted now that I have the edited pieces in Tinkercad.  Sliced in half here and a random height cross-section.   The layer view looks good in Cura, it should be printable.  Let me know if anyone sees any issues.

Beefing up the first baffle (both halves) is probably doable too, let me see.  The weird pattern of staggered "half-baffles" adds to the fun, lol.  At first I was confused -- why just the first baffle -- then realized, duh, it gets the brunt of the incoming blast.

Regarding importing STLs that are too big --
"SW actually warned me that the file has too many points and lines.  Asked if I was sure I wanted to open it.  Now I know the answer is "no". "

Meshmixer is free  -- import -- -- select the object(ctrl-a) -- edit -- reduce (the mesh) -- plenty of options but shape-preserving and percent seem to work pretty well.  Triangle count should be on the lower right edge of screen.   It also has an "Inspector" to look for mesh errors.  -- export.   I will say I have never compared the exterior dimensions of the remeshed STL to the original -- I assume it would be preserved to the limits of STL format resolution, so some simple tests would be a good idea before trusting it with real models.

[subscriber]

Thanks for the heads-up on Meshmixer, Bob

I notice that your modified file is less than half the bitcount of the original.  Yet, I see no roughness or loss in the file.  We are not printing machined accurate and true cylinders to the nearest 0.001" here.   The printers are not capable of that, so why feed them such finely detailed STLs?  I know that I suffered form the "more resolution is better" mentality, until you straightened me out.

[TorqueMaster]

Oh, yeah, I noticed that too.  The original STL was said to have come from Tinkercad and was about 4MB and 88k triangles.  After importing to Tinkercad, editing it, and exporting, it is down to about 2MB and 44k triangles.   I did no further compression, Tinkercad just does what it does, no settings to adjust.  Yeah, for printing purposes, it's still more than adequately defined.

[bduares]

The designer is trying to get on GTA forums. He just sent me this message:

Blackmagic12 sent you a new message:

Hello again.

I tried to join jour forum, but i am still waiting for approval.

In the meantime, you can check out this link in tinkercad: https://www.tinkercad.com/things/aLtasJvxM9F-lyddemper-12-20-unf

It should now be made public.

And another variant if you are interested: https://www.tinkercad.com/things/9RfW7xavFO3-lyddemper-pyramid-core-225-degree-twist-12-20-10mm-hole

[subscriber]

I followed Bob's instruction on how to reduce the original long mod STL file resolution using Meshmixer.  At 50% resolution reduction, Solidworks could open the file as a surface model (not solid).

SW operations are very sluggish, but when I turn off lines and add some grey, I can see what is going on.   Images below.

It may be moot, since Bob worked on the file to strengthen the walls, but at least I have a new tool to edit large STL files.  Even though that will still not go easily, because SW wants to snap to the facets, and establishing proper XYZ planes is work.

[subscriber]

Jared,

I can see that tinkercad file.  It seems to be a variant of the one you posted.  The front cap seems to have a series of holes, rather than just a central bore.  The flutes have features that might be small holes.

[bduares]

Yes, I told him about my issues and he remade it a well as another version. He also says he applied to GTA but is still waiting for approval

[Blackmagic12]

So I decided to shoot a few shots off with the Sumatra and try out these 3 moderators.
First the long one from Subscriber. 2 shots, still quite loud, but otherwise uneventful. It was expected. Can't really quiet the supersonic beast, and it is pushing 70+ fpe.
One after the other with the PLA+ and PETG moderators. Both had impressive explosive deconstructions. Very strange. The first one I printed of that design is still just fine and so far the quietest I have found for the Sumatra.

I admit I had been playing with slicer settings before printing the other 2. It is possible my slicer setting modifications made something weaker. I am almost certain I had used 100% infill, but it is a possibility I did not. So I reset my slicer settings for both the printer and the filament and am printing a new batch.

 
To answer your question, I usually print with whatever I have in the machine at the moment. Right now I have a roll of white PETG in it because I was testing a roll out of a 10 pack case I just picked up the other day. The grey one is PLA+, a stronger and more flexible PLA than the old stuff.

If I want something black I usually just spray paint it a couple of coats, like I did the last moderator I put on my Sumatra. They paint very nice.

Hi.

I have strengthened the failure point bduares experienced.

Regards.
Blackmagic12

[bduares]

Thank you Kent. I am printing your new version now. This one has the little holes in the end cap. That may relieve some of the pressure as well.
Jared

When this one finishes printing I will be doing tests. It will give me 4 moderators to test.
1) The original file printed in 15% infill (didn't remember to change the default Orcslicer infill)
2) The original in 100% infill
3) Bob's fixed version in 100% infill
4) Kent's fixed version in 100% infill

I may even set up one of my good cameras and record the tests as videos. I'm pretty sure the 15% infill will not survive the test. The original file I predict a 50/50 chance of it surviving. So some of you might get some amusement out of the Sumatra blowing apart moderators.

I also just ordered another 3D printer, an Elegoo Neptune 4 Pro. When it gets here I will be able to print that larger moderator design of Subscriber's and test it out.

[TorqueMaster]

Thank you Kent. I am printing your new version now. This one has the little holes in the end cap. That may relieve some of the pressure as well.
Jared

When this one finishes printing I will be doing tests. It will give me 4 moderators to test.
1) The original file printed in 15% infill (didn't remember to change the default Orcslicer infill)
2) The original in 100% infill
3) Bob's fixed version in 100% infill
4) Kent's fixed version in 100% infill

I may even set up one of my good cameras and record the tests as videos. I'm pretty sure the 15% infill will not survive the test. The original file I predict a 50/50 chance of it surviving. So some of you might get some amusement out of the Sumatra blowing apart moderators.

I predict it depends on how thick your walls, top, and bottom are. 
My slicer defaulting to 2mm on those shows very little infill, and not in terribly critical areas. 
If yours are thinner than that, then the 15% infill may cause it to be weaker in critical areas and more likely to fail.

I usually use 3-4mm walls/top/bottom on moderators, (and 100% infill) so there little chance a critical area will not be solid.

The fixed version appears to have constant ~2.0 mm thick walls, rather than the previous ~1.6mm.  At 70 FPE... I'd say it depends on your filament choice and how well the layers fuse together.  That's always true, but even moreso given the power level.  Will a 25% increase in thickness be enough?

 So 1) and 2) are the same model, same printer, same filament?  they should both fracture as before.
 3) I expect has the best chance to survive (though I did not strengthen the 1st baffle)
 4) Somewhat thicker walls give it a better chance, I don't think the extra escape holes will help much.  I'll say 50-50.

I predict at least 2 of them will be fun to watch in the videos

[bduares]

I printed in .2mm layer height with top shell layers of 5 and bottom shell layers of 4, 100% infill in a new roll of Kingroon PETG that is leaving the dryer roller at 22% RH.

I also did the last 2 in Gyroid for more strength. The 1st two are in Concentric.

I am not 100% sure the 2 that blew up were at 100% infill since my slicer defaults to 15% and I may or may not have changed it. I am fairly sure they were 100% though. The slicer settings I had changed to try to get some speed may have been what caused the failure. My original print of this moderator is still working fine, done with no settings changed. This time I am doing it with no fancy changes. I deleted the filament changes and went back to default and then made it 100%. Just to be sure, I deleted Orcaslicer, reinstalled the new version that came out today, put back in my printer and filament temps, and just changed the pattern to Gyroid and the infill to 100%.

[bduares]

From the look of the new file printing the baffles aren't offset halves anymore. They look more bowl shaped. That may help a bit with strength as well.

[subscriber]

Welcome to the forum, Kent

Bob's method of setting the walls to several layers almost make the infill percentage moot.    Certainly, with a low wall surface layer count and 15 percent infill things might get a bit dodgy, one a 1.6 mm wall.

If Kent's new 2 mm part wall thickness survives at 100 percent infill that will be impressive; but it won't tell us how much margin is in the system.  Anyway, Bob and I have a series of test prints with thinner walls to see when a given airgun power blows them up.  Using Bob's 30 FPE Avenger the test prints are amazingly robust.  I am attaching one series of such prints.  I just noticed these files are for dry firing muzzle blast impact.  I will post the pressure vessel version in a bit.

Jared, when you have a print of my larger mod and are ready to test it, could you also test the shorter one along side for sound?  Thanks

[subscriber]

Single chamber mod wall thickness test prints having a 12 CC expansion chamber.
0.4 to 1.2 mm wall thickness:

[subscriber]

Single chamber mod wall thickness test prints having a 12 CC expansion chamber.
1.4 to 2.0 mm wall thickness:

[subscriber]

I suppose a picture would help - for the above attached files.  The version shown here has the thinnest test wall thickness at 0.4 mm.

The 7.5 mm on the front cap is the baffle bore diameter as printed.

The 12 CC is the volume of the chamber.  We started with 24 CC chambers but could not burst even the 0.4 mm wall.  12 CC is approximately the volume of a 22 inch long .22 barrel bore.

[bduares]

Jared, when you have a print of my larger mod and are ready to test it, could you also test the shorter one along side for sound?  Thanks

Will do.

Here are the test subjects that, if I am lucky, I will be testing tomorrow

L-R
Old file 15% infill, Old file 100% infill, Bob's fix 100% infill, Kent's remake 100% infill, Subscriber's "smaller" can. Kent's remake also has tiny bumps all over it like miniature Braille. Interesting.

I just noticed that the three on the left all have a visible seam line in the top portion that reaches to the top of the grooves. Kent's remake no longer has that seam

[Struckat]

A friend printed the original two file in this thread.

I tested both @33 fpe. The long one works as well as my Mod 40. The small one has all the small hole around the exit which will need to be plugged.

These will be used on sub 20 fpe and should work nicely. Thank you.

[subscriber]

Here are the test subjects that, if I am lucky, I will be testing tomorrow....

Thanks; that should be interesting.