Jumping on the 3D printer bandwagon
- Thread starter BigAl205
- Start date
Hopefully not too much - I'd only be doing sealed or Variovent anyway. My main concern would be that I was thinking of doing a tongue-in-groove sort of joint between each section, and I'd have to find some sort of adhesive to bond them permanently, whatever material I end up with.
Has anyone tried to thermo-form a 3D printed object, after printing it?
And I'm not talking about anything fancy - just heating it up with a heat gun and bending/tweaking something that you printed.
Here's my application:
I just picked up two of those Volvo/Dynaudio domes to experiment with.
(The black rings are just screwed on from behind, for shipping - they come off)
And previously, I bought a set of Mirage onmidirectional speakers for my home - very similar to these:
View attachment 14888
Those speakers use normal, small speaker drivers, bounced off reflectors, so the sound is radiated in a 360 degree dispersion (well - they tilted the mid at a 15 degree angle, too). And as a result, you can locate these speakers near - and on - walls.
I was going to play with how well these would work below the corners of an angled windshield - not exactly like "a wall".
My thought was to do some measuring of the ratio of that Mirage dome tweeter diameter/radius vs the reflector diameter/radius, and extrapolate that into 3D printing a larger version of that for the Dynaudio domes.
And then I was thinking I could do additional modifications, maybe bending the back/side edge of that reflector (maybe not) by using heat to bend it after the fact.
Should I maybe keep it thin to allow for that?
Since this is just an experiment, how thick do you think I should print these dishes, so they'd be good reflectors but still thin enough to heat form?
And I'm not talking about anything fancy - just heating it up with a heat gun and bending/tweaking something that you printed.
Here's my application:
I just picked up two of those Volvo/Dynaudio domes to experiment with.
(The black rings are just screwed on from behind, for shipping - they come off)
And previously, I bought a set of Mirage onmidirectional speakers for my home - very similar to these:
View attachment 14888
Those speakers use normal, small speaker drivers, bounced off reflectors, so the sound is radiated in a 360 degree dispersion (well - they tilted the mid at a 15 degree angle, too). And as a result, you can locate these speakers near - and on - walls.
I was going to play with how well these would work below the corners of an angled windshield - not exactly like "a wall".
My thought was to do some measuring of the ratio of that Mirage dome tweeter diameter/radius vs the reflector diameter/radius, and extrapolate that into 3D printing a larger version of that for the Dynaudio domes.
And then I was thinking I could do additional modifications, maybe bending the back/side edge of that reflector (maybe not) by using heat to bend it after the fact.
Should I maybe keep it thin to allow for that?
Since this is just an experiment, how thick do you think I should print these dishes, so they'd be good reflectors but still thin enough to heat form?
Last edited:
Any chance one of you with a 3D printer could print something for me? It's a small bracket, I'd happily pay. Looking to get this printed: https://www.thingiverse.com/thing:3983396
Thanks.
Thanks.
So...
To really complete my dream of using a 3D printer for car audio fabrication - I really REALLY want to be able to scan parts, and modify them in digital space, or remove panels and scan surfaces and make parts that mate up tight.
I bought an Xbox Kinect V2 (the one for the Xbox 360 - supposed to be "the good one"), but I've failed to get it to work with my fairly high-powered laptop, my desktop that the 3D printer is hovering above, and I've even bought a new laptop (which I'll promptly return) that I also can't get it to work with. I should have wondered when I saw there was a "configuration verifier tool"... and discovered that Microsoft discontinued support years ago, closed the forums... and I couldn't have struggled more or tried more to get it working. Time to throw in the towel.
So I'm going to explore "photogammetry" - has anyone here tried it?
Seems theoretically could be easier - take lots of photos all around the object, then use one of these software options:
https://all3dp.com/1/best-photogrammetry-software/
Anyone tried any of these, or any other 3D scanning options?
To really complete my dream of using a 3D printer for car audio fabrication - I really REALLY want to be able to scan parts, and modify them in digital space, or remove panels and scan surfaces and make parts that mate up tight.
I bought an Xbox Kinect V2 (the one for the Xbox 360 - supposed to be "the good one"), but I've failed to get it to work with my fairly high-powered laptop, my desktop that the 3D printer is hovering above, and I've even bought a new laptop (which I'll promptly return) that I also can't get it to work with. I should have wondered when I saw there was a "configuration verifier tool"... and discovered that Microsoft discontinued support years ago, closed the forums... and I couldn't have struggled more or tried more to get it working. Time to throw in the towel.
So I'm going to explore "photogammetry" - has anyone here tried it?
Seems theoretically could be easier - take lots of photos all around the object, then use one of these software options:
https://all3dp.com/1/best-photogrammetry-software/
Anyone tried any of these, or any other 3D scanning options?
I'm giving 3DF Zephyr a shot now.
First attempt -
Took photos of a Hot Wheels car on my counter... 50 pics. Not great lighting. I used my flash on the drivers side of the car, none on the passenger side.
Imported them into 3DF Zephyr and it rejected all but 14 of them - but still created a 'not awful' 3D image, and that software was pretty simple.
I just downloaded a free photo optimizer ("one click" with a bulk import option) to see if I can use something like that as a step to help.
Sent from my LM-G820 using Tapatalk
First attempt -
Took photos of a Hot Wheels car on my counter... 50 pics. Not great lighting. I used my flash on the drivers side of the car, none on the passenger side.
Imported them into 3DF Zephyr and it rejected all but 14 of them - but still created a 'not awful' 3D image, and that software was pretty simple.
I just downloaded a free photo optimizer ("one click" with a bulk import option) to see if I can use something like that as a step to help.
Sent from my LM-G820 using Tapatalk
Did some playing yesterday, trying to get a Hot Wheels Trans Am scanned.
First set of pics - phone camera, overhead lights on kitchen island, and phone flash on -
The software rejected a bunch of my 50 photos, so it worked, but the end product was pretty rough.
Second attempt - I downloaded photo editing software that worked in bulk (more like "in sequence"), cropped all the pics real tight to the car and clicked "auto optimize" for each...
Actually got worse results. Rejected more pics. I think I exacerbated any blurryness by zooming in.
Third attempt - new pics, camera on "manual mode" (as a video advised) and I only took 25 pics. Same setup - kitchen island, overhead light, phone flash.
Actually got worse results again. I think I need to use the manual camera controls - but I'm no photographer.
Fourth attempt - Took a quick 25 second video around the car, as I saw the software had a "import video" feature. I felt like I rushed it, but what the heck.
Definitely didn't get good results - all that does is take frames every "x" seconds (you specify), rather than taking the best/clearest frames.
Fifth attempt - tried to take a longer, slower, steadier video - but failed. I am going to pick up a handheld phone gimbal I think (and go back to photos) - focus seems to be the key, whether photo or video, so I think that will help.
My first run was the best and proved two things:
1) it works!
2) it's super dependant on the quality of your pics.
Sent from my LM-G820 using Tapatalk
First set of pics - phone camera, overhead lights on kitchen island, and phone flash on -
The software rejected a bunch of my 50 photos, so it worked, but the end product was pretty rough.
Second attempt - I downloaded photo editing software that worked in bulk (more like "in sequence"), cropped all the pics real tight to the car and clicked "auto optimize" for each...
Actually got worse results. Rejected more pics. I think I exacerbated any blurryness by zooming in.
Third attempt - new pics, camera on "manual mode" (as a video advised) and I only took 25 pics. Same setup - kitchen island, overhead light, phone flash.
Actually got worse results again. I think I need to use the manual camera controls - but I'm no photographer.
Fourth attempt - Took a quick 25 second video around the car, as I saw the software had a "import video" feature. I felt like I rushed it, but what the heck.
Definitely didn't get good results - all that does is take frames every "x" seconds (you specify), rather than taking the best/clearest frames.
Fifth attempt - tried to take a longer, slower, steadier video - but failed. I am going to pick up a handheld phone gimbal I think (and go back to photos) - focus seems to be the key, whether photo or video, so I think that will help.
My first run was the best and proved two things:
1) it works!
2) it's super dependant on the quality of your pics.
Sent from my LM-G820 using Tapatalk
So all the above, by the way, was me jumping in and not even doing a tutorial... And now I see there's no excuse, these are all totally short:
General overview / basic use / demo: (worth watching for anyone curious)
https://youtu.be/17UTelgZqBg
And basically the "how to take pictures for this, you dummy":
https://youtu.be/E06kgYBftak
So...
I'll be trying again after doing the steps differently - like they say I actually SHOULD do them
General overview / basic use / demo: (worth watching for anyone curious)
https://youtu.be/17UTelgZqBg
And basically the "how to take pictures for this, you dummy":
https://youtu.be/E06kgYBftak
So...
I'll be trying again after doing the steps differently - like they say I actually SHOULD do them
I see those Haribo gummies there.Some progress after taking pictures like they said and actually following the workflow like they suggest.
[video]https://photos.app.goo.gl/8Jaus1echeGf1Tdv8[/video]
It still discarded most of my photos from this set, so I think I need to try the photo calibration so it doesn't hate my phone pics.
[video]https://photos.app.goo.gl/8Jaus1echeGf1Tdv8[/video]
It still discarded most of my photos from this set, so I think I need to try the photo calibration so it doesn't hate my phone pics.
Once I un-fudge my non-existent photography skills and actually get a good looking model, that's next, for sure!
Looks like the best way to do that is just relative measurements. That could be anything from putting reference dots on the object to scanning a ruler next to the object, to just measuring any old thing on the scanned object, then scaling it up in the 3D space until those same points have the same measurement as reality.
That to me is critical, though - I want to scan my inner door surface(s) and make enclosures that hug up tight behind them, for midbass. Scale has to be perfect.
I also want to scan my naked A pillars and sail panels and print my own lower portion to hold my widebands and BG Neo3s.
...for that, I already did buy some junkyard A pillar panels if I have to do those the old school way, but I just think it would be cool to print those.
Sent from my LM-G820 using Tapatalk
Looks like the best way to do that is just relative measurements. That could be anything from putting reference dots on the object to scanning a ruler next to the object, to just measuring any old thing on the scanned object, then scaling it up in the 3D space until those same points have the same measurement as reality.
That to me is critical, though - I want to scan my inner door surface(s) and make enclosures that hug up tight behind them, for midbass. Scale has to be perfect.
I also want to scan my naked A pillars and sail panels and print my own lower portion to hold my widebands and BG Neo3s.
...for that, I already did buy some junkyard A pillar panels if I have to do those the old school way, but I just think it would be cool to print those.
Sent from my LM-G820 using Tapatalk