Just want to say that the very first thing I do with a scan from my POP 2 is to clean up the RAW data. Once I've cleared out the useless data, I move on to where your video starts off. Note that the current version of Revoscan STILL has bugs in its point selection tools. For example, if you lasso-select a group of points for the purpose of deleting them, even if you have the 'select through' option turned on, the lasso doesn't always pick up all the points you've lassoed. You have to repeat the lasso/delete operation two or more times to get everything.
Thanks for the comment. Are you able to clean up the points before the Fusion step? Also with the Miraco support I have a beta version of revoscan and haven't noticed the lasso issue. I imagine that will get sorted but if not I will mention it.
@@LearnEverythingAboutDesign _Are you able to clean up the points before the Fusion step?_ Yes. It seems Revopoint rightly treats the raw point data as sacrosanct, so the ability to edit it is slightly hidden away in the '...' menu beside each model in your project model list. The lasso issue occurs in the raw data edit, for sure, but since I do all my geometry cleanup on the raw data, I rarely use the selection tools in other 'stages', so not sure if it occurs there, too.
@@imacmill Great info thanks! I do like the ability to step back to the raw data, but it is frustrating when all the trims/deletes are gone. I imagine editing the complete raw data is destructive.
@@LearnEverythingAboutDesign Ya, raw data edits are forever, so I choose carefully what I delete. I'm not sure if it's even reasonable to do it all, but I just don't want to risk 'bad' raw data having a negative influence on the rest of the tool chain. Maybe it won't?
@@imacmill yeah it kind of depends. So if you were scanning something perfectly flat and you had some noise on the points, the Fusion, Isolation, and overlap "should" take care of that for you. You can always copy the entire project before you begin the edit so you have an untouched version. I am not sure its accurate enough to really be able to tell/quantify if hacking the raw data has an impact. It will certainly process faster because sometimes doing Fusion on all the scan can be time consuming.
It's always cool to watch a pro using his tools at work - no matter if it's sheet metal or high performance software. Thanks for posting these infos based on a lot of experience. 🔥
Best tutorial I’ve seen so far!!! Thank you. Please do more of these. Especially for car parts. I’m not scanning faces on a turntable 😂😂😂 I am scanning GTR parts, so your Starion fender is MUCH closer to the mark for my uses.
That might be the first and last time a starion and GTR are in the same sentence lol :) There is a good bit on this channel. I did a series with the starion scanning the fender and modeling a widebody. We have more scanning content in the works. What scanner are you using?
I have a Revopoint Range. I have just started my 3D scanning journey and I may not have started with the easiest parts. A R32 GTR trunk lid, and doors from a military Humvee. Apparently I’ve been taking the Belligerent Amateur approach to learning this. 😂 I plan to do more body panels for the GTR once I have the process figured out. Right now, it is not going well. Your video was very helpful so I will be trying again while playing your videos as a guide. And then 3D printing molds in sections
@@erikrummel6277 Those are 2 very different vehicles lol. The Humvee will be more difficult as its flat without a lot of unique features. You will want to use marker dots for sure or just take some blue or green painters tape, ball little pieces up as random shapes and stick them on large flat sections on the door. This works as "feature" scanning and on the flat surfaces you should be able to remove and patch those areas easy enough. R32 trunk..... might need marker dots or random feature trick again. I just recently did a 3 part series on small rc parts. while not exactly the same i did spend a bit of time talking about prep, capture, and process of the scan in the first 2 videos. id check that out next.
I'm glad i found your channel. I've been teaching myself scanning and cad for the last year and it's been a bit of a struggle bus. This video was very helpful for me since i'm working with revopoint scanners alongside fusion. I'll be checking out your other videos and just wanted to say thank you!
Awesome! I have lots of mesh based videos on this channel for mechanical parts and designing off scans/meshes. Hopefully it has what you are looking for!
Great video! I've struggled with working with scanned data in Fusion360, and this video gives a good overview on how to reduce the facets. I would love to see more on the next step in the process. For example, if you scanned a fender and wanted to design some fender flares, how would you use the scanned mesh after importing into fusion and reducing the facet count?
but note we kept the poly count fairly high. it worked ok for what we were doing and the process was slightly different with shining vs revo software. In general you can reduce the count as much as you can get away with, but for the design process I make the mesh Unselectable to speed up performance.
Thanks for the video! At 16:13 you say a dedicated mesh software would be better for this - do you have any recommendations please? I need to do boolean operations on scanned objects so wondered what software would give best results.
Well in a professional sense i think the commonly used software is Geomagic. There are free tools like MeshLab, but they are harder to use. I have plans to do some meshlab videos. Geomagic isn't cheap. I think the "essentials" version is still something like $7000. So it hard to justify spending that sort of money when a scanner in this segment is around $600-$1500. So if you are a hobby user I would look into MeshLab. I used it like 20 years ago and it hasn't changed much but its free/open source and powerful if you can figure it out.
Is that a Starion?! Any thoughts on their Revoport Mini? I have one and it’s been very frustrating to use… I come from professional tools like Faro arms etc and trying to do things at home as well. Any input on how to capture better models is greatly appreciated!
It is yes, 84 ES "flatty" On the mini, I don't have one but the sensors in the miraco i used in "near" mode are the same as the mini. I will be working on a video soon doing a detailed scan with it. Since the miraco is new the revoscan support is a little tricky so it takes a bit longer to say calibrate and things like that as they refine the software. What I have noticed with the Miraco is that it seems to do better with feature alignment than marker alignment in near mode so i would imagine the Mini is the same way. What are you trying to scan with it?
Sounds fun but it might be awhile before I could do that. I do have a video on making a pop up surround as well as one on making an intake duct for an mx5 bumper.
@@LearnEverythingAboutDesign yes i used the intake duct and tried it on the headlight and its going well almost done took me all day long ( I have a question do you have any idea how to make a compression mold for chopped carbon ??) thank you btw awesome videos 🥰❤️
@@real-player1237 I don't but easy composites has a bunch of great videos you should check out if you haven't already. ua-cam.com/video/25PmqM24HEk/v-deo.htmlsi=iKNRW5iQQjEsuzVA
I have a question, hopefully you see this comment; is there ever any issues with scaling the mesh in Fusion? I just got my POP 3, and was doing some test scans on an underwater drone I want to design parts for, and when I brought it into fusion, the mesh seemed gigantic. Do I need to specify units as millimeters, or calibrate the mesh scale somehow? Thanks for this tutorial, it was a huge help!
The units in revopoint are generally mm unless you have changed something there. So when inserting into Fusion in the insert dialog make sure its set to mm. Even if the units in the file are inch or meter or something, on the insert mesh dialog make sure its mm and the scale should be correct.
@@LearnEverythingAboutDesign I see what my problem was now, I was just opening the mesh, rather than importing it through the dialog. Thank you! Side note: any tips in terms of recreating geometries for solid modeling? I was just doing my best to fit a spline to the mesh along a slice to reconstruct the geometry, but in all honesty I'd like to do something like cut the body based on the mesh as a cutting tool; should I convert the body to a mesh and then slice? This is a new world of engineering design to me
@@Roy-K yeah, when you open a mesh it also puts the file into no history capture which changes a little bit. As for getting info from the mesh, yes you can create mesh section sketches. We have done this in a few series but you can see it in this one. ua-cam.com/play/PLBDfGh8A8kXWYEY9X4vJ5sKGh6TPzrr2h.html Basically you use Mesh Section sketch on the mesh tools, then edit the sketch and you can have it auto fit lines, splines, arcs, and circles to a mesh.
Hi! Me again. Hope you’re well? I’m in preliminary research phase and struggling to make some decision. Outsourcing scanning and reverse engineering is expensive so I’m hoping I can do as much as possible in house. Have you found the einstar to give you good enough accuracy to make well fitting splitters, skirts, diffusers etc, or would you advice biting the bullet and contract a professional company to get me the scans? Also, I’m seeing a lot of people using blender now as it creates some nice organic shapes. I’m torn between F360, Blendrr and SWs - I need to be as time efficient as possible so need to focus my learning into one program. Is Fusion up to the task for creating these parts? As always, your advice is hugely appreciated. Thank you!!
Hi, there is a bit to unpack in those questions so let me try and break it down. 1. DIY vs hiring someone. Generally when you hire someone, the gear they use will be very accurate(expensive). It will scan faster and give you a higher density accurate mesh. Is it needed? well that depends really on how you are going to make the parts from your cad files. If you are going directly to CNC molds from foam or something you might want to work with a pro. Keep in mind the difference in accuracy is something like .2mm for a hobby scanner to .02 or lower for a pro scanner when say scanning a fender. If you plan to 3d print body parts to build a mold I think .2mm is likely good enough(personally), but again it really depends on how you are making the parts. If i was going to be spending money on tooling I wouldn't want the scan data to be the weak point in that chain. 2. Software. Blender is free so bonus there, but its a "poly" modeling program. You are only working with mesh. This can be great to explore shapes, but it will be someone limited when it comes to CAD. Lets say you wanted to take some of your body parts into a simulation program for airflow, well you can't because they need to be Solid, not a mesh. If you design them in CAD they can be used for simulation. Also controlling things like hole sizes and locations for tight fitments, well that is a CAD thing. As for what software. Solidworks has better surfacing tools, but lacks freeform tools(unless you buy nPower) and doesn't really handle mesh well. Fusion surfacing tools are ok but not great, but it has freeform modeling and deals with meshes better. Freeform can snap to the mesh. Also there is a big price difference. Fusion is a few hundred a year while Solidworks is thousands of dollars. I have loads of videos on making spoilers(some surface, some forms) wide bodies. canards etc. I have a whole series on scan and widebody design. ua-cam.com/play/PLBDfGh8A8kXVvKBJ4UudhVLKlPdumAg5P.html
I’m currently looking at the new Creality otter and Raptor(laser) scanners. Finally we are getting meteoritical grade, consumer priced scanners . I’m personally hoping there’s a tidal wave of content created using scanners that are not just gimmicky that require a stupid amount of post processing. 🇦🇺🤜🏼🤛🏼🍀😎☮️☮️☮️
I have an Otter on the way and hoping to get my hands on the raptor shortly after that. in the next few weeks I should have a video with the Otter. It is going to be in a similar category to the revopoint Mini2 except the mini is blue light and the otter is IR.
it doesn't seem like paying for the design extension on fusion is worth it either in case anyone is wondering, it might be for smaller stuff like the product design online video shows with the cane grip, but for larger surfaces like fender or boats like i'm doing it isn't able to convert using the organic paid function (i have a startup license and using the trial but am still a total noob)
its looking like its probably worth spending the $ on rhino and mesh2surface for things like this as far as value to $ spent for reverse engineering stuff like this... again i'm still a total newcomer to this.... for sheet metal and stuff like that fusion still is probably better... for plasma cnc stuff like that.
There are some free quadremeshers. Autodesk used to have some available but they are no longer. Here is one you can try but honestly you would probably have better luck with some Blender add-ons. but for free its worth a try. github.com/wjakob/instant-meshes And I would agree about the Extension. Honestly even for plastic part design it isn't there yet. The biggest values in that package right now are vol lattice tool and maybe the screw boss stuff but i found the limits of it on complex shapes and still had to make them manually. I think in theory adding bosses ribs and hardware to multiple bodies at the same time is a great tool! On rhino and mesh2surface. The power surfacing stuff from nPower for solidworks is really good. I used to do work for those guys and even worked at their booth at solidworks world years ago. It has some pros over fusion forms like proper edge weighting and they have a sub level of detail that is nice to have, but $$$. With Rhino8, that is on my list. I wasn't a fan of Rhino7 for ui/ux reasons. Rhino used to use Tsplines which Autodesk bought. so the Rhino SubD modeling was lacking. I wish they would put effort into ui/ux and maybe Rhino8 has, i just haven't had the time to try it yet. There aren't a ton of great options out there for a hobby user. You can use Blender if your goal is just to 3d print parts. Its free and has some cool geometry node type things you can do. But if you need CAD then I still currently prefer Fusion 360.
in my experience using smoothing/ simplify works better on the mesh . specifically using simplify on the mesh will reduce poly count but keep harder edges rather then smoothing details
Thanks for the comment and data point! During my testing using more smoothing on the mesh ended up reducing the harder edge details on the scans. Maybe it is a case of the scan geometry making the difference? When i was working with Revopoint they suggest doing more processing on the points vs on the mesh. They said to prioritize the points over the mesh. They did tell me that the smoothing algorithm isn't straight forward in that it doesn't just look at the pure mesh but also the overall shape. One thing that is nice is the undo button :)
@@LearnEverythingAboutDesign yea the undo is a GREAT feature.. weird tho i've had the most luck with simplify the mesh gives like the best of both worlds i've simplified models and the 100k result looks better then the 3mil result from a smoothness standpoint and losing minimal details but i feel like doing anything to the point cloud tends to really soften things. however i'm sure it's model dependent
Wow your scans are super dense with points, I guess I should scan slower, doesn't help that Its only 7fps connected to my mobile device, I have a Range and I bought it for scanning large things, I tried scanning my headlight housing the otherday and it did a okay job, its gets lost pretty easy, especially when going around 90 degree corners where very little geometry is shared
yeah corners are always a tough one. The scanner really wants to be perpendicular to what its scanning. Best thing I can say is try to add some unique references to the part that you can trim out later. Little blobs of clay or tape. Marker dots don't really help because the scanner needs to be looking pretty much right at it to register and you need a ton to make that work. Scanning on a mobile device does limit the performance a bit. I was impressed with how fast the Miraco captures the data vs the Inspire being tethered to a laptop.
Yeah, last year I did a series with the ProHD. I am in talks with Shinning as I would like to get a freescan and einstar tested this year. Time is the issue. They loan their scanners out so I need to plan accordingly to have the time. Hopefully in a few months...
Just want to say that the very first thing I do with a scan from my POP 2 is to clean up the RAW data. Once I've cleared out the useless data, I move on to where your video starts off.
Note that the current version of Revoscan STILL has bugs in its point selection tools. For example, if you lasso-select a group of points for the purpose of deleting them, even if you have the 'select through' option turned on, the lasso doesn't always pick up all the points you've lassoed. You have to repeat the lasso/delete operation two or more times to get everything.
Thanks for the comment. Are you able to clean up the points before the Fusion step?
Also with the Miraco support I have a beta version of revoscan and haven't noticed the lasso issue. I imagine that will get sorted but if not I will mention it.
@@LearnEverythingAboutDesign _Are you able to clean up the points before the Fusion step?_
Yes. It seems Revopoint rightly treats the raw point data as sacrosanct, so the ability to edit it is slightly hidden away in the '...' menu beside each model in your project model list.
The lasso issue occurs in the raw data edit, for sure, but since I do all my geometry cleanup on the raw data, I rarely use the selection tools in other 'stages', so not sure if it occurs there, too.
@@imacmill Great info thanks! I do like the ability to step back to the raw data, but it is frustrating when all the trims/deletes are gone. I imagine editing the complete raw data is destructive.
@@LearnEverythingAboutDesign Ya, raw data edits are forever, so I choose carefully what I delete. I'm not sure if it's even reasonable to do it all, but I just don't want to risk 'bad' raw data having a negative influence on the rest of the tool chain. Maybe it won't?
@@imacmill yeah it kind of depends. So if you were scanning something perfectly flat and you had some noise on the points, the Fusion, Isolation, and overlap "should" take care of that for you. You can always copy the entire project before you begin the edit so you have an untouched version. I am not sure its accurate enough to really be able to tell/quantify if hacking the raw data has an impact. It will certainly process faster because sometimes doing Fusion on all the scan can be time consuming.
It's always cool to watch a pro using his tools at work - no matter if it's sheet metal or high performance software. Thanks for posting these infos based on a lot of experience. 🔥
Thanks for watching!
Best tutorial I’ve seen so far!!! Thank you. Please do more of these. Especially for car parts. I’m not scanning faces on a turntable 😂😂😂 I am scanning GTR parts, so your Starion fender is MUCH closer to the mark for my uses.
That might be the first and last time a starion and GTR are in the same sentence lol :) There is a good bit on this channel. I did a series with the starion scanning the fender and modeling a widebody. We have more scanning content in the works.
What scanner are you using?
I have a Revopoint Range. I have just started my 3D scanning journey and I may not have started with the easiest parts. A R32 GTR trunk lid, and doors from a military Humvee.
Apparently I’ve been taking the Belligerent Amateur approach to learning this. 😂
I plan to do more body panels for the GTR once I have the process figured out. Right now, it is not going well. Your video was very helpful so I will be trying again while playing your videos as a guide. And then 3D printing molds in sections
@@erikrummel6277 Those are 2 very different vehicles lol. The Humvee will be more difficult as its flat without a lot of unique features. You will want to use marker dots for sure or just take some blue or green painters tape, ball little pieces up as random shapes and stick them on large flat sections on the door. This works as "feature" scanning and on the flat surfaces you should be able to remove and patch those areas easy enough.
R32 trunk..... might need marker dots or random feature trick again.
I just recently did a 3 part series on small rc parts. while not exactly the same i did spend a bit of time talking about prep, capture, and process of the scan in the first 2 videos. id check that out next.
I'm glad i found your channel. I've been teaching myself scanning and cad for the last year and it's been a bit of a struggle bus. This video was very helpful for me since i'm working with revopoint scanners alongside fusion. I'll be checking out your other videos and just wanted to say thank you!
Awesome! I have lots of mesh based videos on this channel for mechanical parts and designing off scans/meshes. Hopefully it has what you are looking for!
@@LearnEverythingAboutDesign it's greatly appreciated, I can only beat my head off the wall for so long lol
Great video! I've struggled with working with scanned data in Fusion360, and this video gives a good overview on how to reduce the facets. I would love to see more on the next step in the process. For example, if you scanned a fender and wanted to design some fender flares, how would you use the scanned mesh after importing into fusion and reducing the facet count?
Earlier in the year i did a series on that. ua-cam.com/play/PLBDfGh8A8kXVvKBJ4UudhVLKlPdumAg5P.html
but note we kept the poly count fairly high. it worked ok for what we were doing and the process was slightly different with shining vs revo software.
In general you can reduce the count as much as you can get away with, but for the design process I make the mesh Unselectable to speed up performance.
Thanks for the video! At 16:13 you say a dedicated mesh software would be better for this - do you have any recommendations please? I need to do boolean operations on scanned objects so wondered what software would give best results.
Well in a professional sense i think the commonly used software is Geomagic. There are free tools like MeshLab, but they are harder to use. I have plans to do some meshlab videos.
Geomagic isn't cheap. I think the "essentials" version is still something like $7000. So it hard to justify spending that sort of money when a scanner in this segment is around $600-$1500.
So if you are a hobby user I would look into MeshLab. I used it like 20 years ago and it hasn't changed much but its free/open source and powerful if you can figure it out.
@@LearnEverythingAboutDesign Oh damn! Ok thanks Ill look into MeshLab.
Is that a Starion?!
Any thoughts on their Revoport Mini? I have one and it’s been very frustrating to use… I come from professional tools like Faro arms etc and trying to do things at home as well. Any input on how to capture better models is greatly appreciated!
It is yes, 84 ES "flatty"
On the mini, I don't have one but the sensors in the miraco i used in "near" mode are the same as the mini. I will be working on a video soon doing a detailed scan with it. Since the miraco is new the revoscan support is a little tricky so it takes a bit longer to say calibrate and things like that as they refine the software. What I have noticed with the Miraco is that it seems to do better with feature alignment than marker alignment in near mode so i would imagine the Mini is the same way. What are you trying to scan with it?
@@LearnEverythingAboutDesign thank you for the feedback! Looking to scan and ultimately print interior panels/parts primarily.
can you make a video on how to make a headlight intake using a 3d scan headlight ?
Sounds fun but it might be awhile before I could do that. I do have a video on making a pop up surround as well as one on making an intake duct for an mx5 bumper.
@@LearnEverythingAboutDesign yes i used the intake duct and tried it on the headlight and its going well almost done took me all day long ( I have a question do you have any idea how to make a compression mold for chopped carbon ??) thank you btw awesome videos 🥰❤️
@@real-player1237 I don't but easy composites has a bunch of great videos you should check out if you haven't already. ua-cam.com/video/25PmqM24HEk/v-deo.htmlsi=iKNRW5iQQjEsuzVA
I have a question, hopefully you see this comment; is there ever any issues with scaling the mesh in Fusion? I just got my POP 3, and was doing some test scans on an underwater drone I want to design parts for, and when I brought it into fusion, the mesh seemed gigantic. Do I need to specify units as millimeters, or calibrate the mesh scale somehow? Thanks for this tutorial, it was a huge help!
The units in revopoint are generally mm unless you have changed something there. So when inserting into Fusion in the insert dialog make sure its set to mm. Even if the units in the file are inch or meter or something, on the insert mesh dialog make sure its mm and the scale should be correct.
@@LearnEverythingAboutDesign I see what my problem was now, I was just opening the mesh, rather than importing it through the dialog. Thank you!
Side note: any tips in terms of recreating geometries for solid modeling? I was just doing my best to fit a spline to the mesh along a slice to reconstruct the geometry, but in all honesty I'd like to do something like cut the body based on the mesh as a cutting tool; should I convert the body to a mesh and then slice? This is a new world of engineering design to me
@@Roy-K yeah, when you open a mesh it also puts the file into no history capture which changes a little bit.
As for getting info from the mesh, yes you can create mesh section sketches. We have done this in a few series but you can see it in this one. ua-cam.com/play/PLBDfGh8A8kXWYEY9X4vJ5sKGh6TPzrr2h.html Basically you use Mesh Section sketch on the mesh tools, then edit the sketch and you can have it auto fit lines, splines, arcs, and circles to a mesh.
@@LearnEverythingAboutDesign That is incredible, thank you so much!
Hi! Me again. Hope you’re well?
I’m in preliminary research phase and struggling to make some decision.
Outsourcing scanning and reverse engineering is expensive so I’m hoping I can do as much as possible in house. Have you found the einstar to give you good enough accuracy to make well fitting splitters, skirts, diffusers etc, or would you advice biting the bullet and contract a professional company to get me the scans?
Also, I’m seeing a lot of people using blender now as it creates some nice organic shapes. I’m torn between F360, Blendrr and SWs - I need to be as time efficient as possible so need to focus my learning into one program. Is Fusion up to the task for creating these parts?
As always, your advice is hugely appreciated. Thank you!!
Hi, there is a bit to unpack in those questions so let me try and break it down.
1. DIY vs hiring someone. Generally when you hire someone, the gear they use will be very accurate(expensive). It will scan faster and give you a higher density accurate mesh. Is it needed? well that depends really on how you are going to make the parts from your cad files. If you are going directly to CNC molds from foam or something you might want to work with a pro. Keep in mind the difference in accuracy is something like .2mm for a hobby scanner to .02 or lower for a pro scanner when say scanning a fender. If you plan to 3d print body parts to build a mold I think .2mm is likely good enough(personally), but again it really depends on how you are making the parts. If i was going to be spending money on tooling I wouldn't want the scan data to be the weak point in that chain.
2. Software. Blender is free so bonus there, but its a "poly" modeling program. You are only working with mesh. This can be great to explore shapes, but it will be someone limited when it comes to CAD. Lets say you wanted to take some of your body parts into a simulation program for airflow, well you can't because they need to be Solid, not a mesh. If you design them in CAD they can be used for simulation. Also controlling things like hole sizes and locations for tight fitments, well that is a CAD thing. As for what software. Solidworks has better surfacing tools, but lacks freeform tools(unless you buy nPower) and doesn't really handle mesh well. Fusion surfacing tools are ok but not great, but it has freeform modeling and deals with meshes better. Freeform can snap to the mesh. Also there is a big price difference. Fusion is a few hundred a year while Solidworks is thousands of dollars.
I have loads of videos on making spoilers(some surface, some forms) wide bodies. canards etc. I have a whole series on scan and widebody design. ua-cam.com/play/PLBDfGh8A8kXVvKBJ4UudhVLKlPdumAg5P.html
@@LearnEverythingAboutDesign thank you for the info Matt. You’re extremely helpful.
You can make a body kit for BMW e87 5 doors 2008r? And load it in the description? to print on a 3D printer for personal use😊
I’m currently looking at the new Creality otter and Raptor(laser) scanners.
Finally we are getting meteoritical grade, consumer priced scanners
.
I’m personally hoping there’s a tidal wave of content created using scanners that are not just gimmicky that require a stupid amount of post processing.
🇦🇺🤜🏼🤛🏼🍀😎☮️☮️☮️
I have an Otter on the way and hoping to get my hands on the raptor shortly after that. in the next few weeks I should have a video with the Otter. It is going to be in a similar category to the revopoint Mini2 except the mini is blue light and the otter is IR.
it doesn't seem like paying for the design extension on fusion is worth it either in case anyone is wondering, it might be for smaller stuff like the product design online video shows with the cane grip, but for larger surfaces like fender or boats like i'm doing it isn't able to convert using the organic paid function (i have a startup license and using the trial but am still a total noob)
its looking like its probably worth spending the $ on rhino and mesh2surface for things like this as far as value to $ spent for reverse engineering stuff like this... again i'm still a total newcomer to this.... for sheet metal and stuff like that fusion still is probably better... for plasma cnc stuff like that.
There are some free quadremeshers. Autodesk used to have some available but they are no longer. Here is one you can try but honestly you would probably have better luck with some Blender add-ons. but for free its worth a try.
github.com/wjakob/instant-meshes
And I would agree about the Extension. Honestly even for plastic part design it isn't there yet. The biggest values in that package right now are vol lattice tool and maybe the screw boss stuff but i found the limits of it on complex shapes and still had to make them manually. I think in theory adding bosses ribs and hardware to multiple bodies at the same time is a great tool!
On rhino and mesh2surface. The power surfacing stuff from nPower for solidworks is really good. I used to do work for those guys and even worked at their booth at solidworks world years ago. It has some pros over fusion forms like proper edge weighting and they have a sub level of detail that is nice to have, but $$$. With Rhino8, that is on my list. I wasn't a fan of Rhino7 for ui/ux reasons. Rhino used to use Tsplines which Autodesk bought. so the Rhino SubD modeling was lacking. I wish they would put effort into ui/ux and maybe Rhino8 has, i just haven't had the time to try it yet. There aren't a ton of great options out there for a hobby user. You can use Blender if your goal is just to 3d print parts. Its free and has some cool geometry node type things you can do. But if you need CAD then I still currently prefer Fusion 360.
hats off :)
in my experience using smoothing/ simplify works better on the mesh . specifically using simplify on the mesh will reduce poly count but keep harder edges rather then smoothing details
Thanks for the comment and data point! During my testing using more smoothing on the mesh ended up reducing the harder edge details on the scans. Maybe it is a case of the scan geometry making the difference? When i was working with Revopoint they suggest doing more processing on the points vs on the mesh. They said to prioritize the points over the mesh. They did tell me that the smoothing algorithm isn't straight forward in that it doesn't just look at the pure mesh but also the overall shape. One thing that is nice is the undo button :)
@@LearnEverythingAboutDesign yea the undo is a GREAT feature.. weird tho i've had the most luck with simplify the mesh gives like the best of both worlds i've simplified models and the 100k result looks better then the 3mil result from a smoothness standpoint and losing minimal details but i feel like doing anything to the point cloud tends to really soften things. however i'm sure it's model dependent
Wow your scans are super dense with points, I guess I should scan slower, doesn't help that Its only 7fps connected to my mobile device, I have a Range and I bought it for scanning large things, I tried scanning my headlight housing the otherday and it did a okay job, its gets lost pretty easy, especially when going around 90 degree corners where very little geometry is shared
yeah corners are always a tough one. The scanner really wants to be perpendicular to what its scanning. Best thing I can say is try to add some unique references to the part that you can trim out later. Little blobs of clay or tape. Marker dots don't really help because the scanner needs to be looking pretty much right at it to register and you need a ton to make that work.
Scanning on a mobile device does limit the performance a bit. I was impressed with how fast the Miraco captures the data vs the Inspire being tethered to a laptop.
you should do one with the einstar and exstar software. i'm getting better scans than this with the einstar.
i should say scans that contain the needed detail with less screwing around in exstar... the second part of the process is why i'm here.
Yeah, last year I did a series with the ProHD. I am in talks with Shinning as I would like to get a freescan and einstar tested this year. Time is the issue. They loan their scanners out so I need to plan accordingly to have the time. Hopefully in a few months...
I hate the new UA-cam UI...