Responses to common questions/statements: - Once the part is verified to fit, I will be releasing the STL for free. I have no desire to profit from this, and Myles had no desire to profit from it either. This is the same Myles that frequently makes 3D printed parts for the Mighty Car Mods boys. He is a generous legend. We are all just car enthusiasts using 3D printing/scanning to enjoy the hobby and help each other. - Plenty of requests to also try scanner x or y. I don't have an iPhone to try that and I previously compared a Kinect scanner to the Revopoint in that video. The Revopoint is much better so there was no point including the Kinect sensor here. - As I mentioned in the video, this part will need to be painted before it goes in the car. The resin will become more and more brittle in the sun so will need to be protected. If it still turns out to be too brittle, as some commenters have suggested, a mold can be made to create the part from another technique if necessary. - Some people are upset that I didn't show the part in place. This is simply a matter of logistics. I've only met the owner of the car a couple of times and we don't live near each other. In case anyone has forgotten, we are still in a pandemic and it's not as straight forward as rocking up somewhere to try the part. It's unlikely I will see the car in person any time soon, but I will update with a pic in the description when I can. - Plenty of people saying to just model the part. As the video stated, I could have modelled it, but it would need several iterations with test fitting in the car before it was ready. As the point above explains, this was not possible. The scanned part still needed clean up, but I am 100% confident the overall form will fit the dash. It's just a matter of minor refinements where necessary for the 'clip' portions on the underside which I can do remotely. Besides, Myles had his scanner and offered for me to try it. This project came up and I'm sure most would agree the comparison between scanning methods is what the video interesting.
Making a silicone mold of the original part and casting resin would probably still be the most accurate compared to scanning, but now you can share it as a file, and the accuracy isn't critical. Another very precise but time consuming way to 3d scan small objects (though it's easier with conductive objects because you can have an automatic trigger) is with a cnc or a dedicated mechanical scanner and plotting the points. (Basically an AFM but bigger) You could probably make that with an fdm 3d printer by gluing a needle on a bl touch and making a grid pattern and noting at which point it triggered. Not sure if the accuracy would be that great though.
Good on you mate. Plenty of internet experts and critics; very few blokes like you who not only make something new, but share it for free. Keep up the great work!
Considering how rare of a part that is, i hope you release an stl so other people can finish their datsuns too. Really amazing to see a damn near replica of a part that no longer exist.
@@Waitwhat469 nah. I doubt there’s that many cars still around to justify such extremes. I assume most owners of such classic cars are of classic age too, so I doubt their understanding of Kickstarter and similar online things
If that was my dato and it was the final bit i needed after a 10 yr 100k+ resto i'd pay a lot more than 50 or 60 us or a modest fee.... this would be a dream come true and i'd happily pay a premium to some one who just took the initiative to go to the trouble to spend prolly 20+ hours with specialist tools.... f^ck i'd prolly give my left nut.
The thing I like the most about your videos is that even if I don't fully understand the video's subjects or explanations, I still get some sort of inspiration for something else. In either case, your videos turn out to be extremely helpful regardless.
_You skipped a step that I always perform. I completely remodel the component in CAD, using the scanned mesh as a guide._ _As for painting the final part, I would plate it with satin chrome, to draw attention to them, and as a flex on other Datsun owners._
That technique is what Myles suggested before he started the scanning. I agree it would likely give the best result but in this I'm hoping it's not necessary.
@@TeachingTech I would like to see the final product. With a good finish I'm sure it will be fine. How concours is a Datsun today? At some point someone at one of the auto aftermarket manufactures will have injection tooling made. Then it will be remodeled.
Nah, the guy who wants that part wants it to look exactly like it just left the Datsun factory. The fellow enthusiasts who know enough to be impressed by his car having perfect vent grilles will notice them without gaudy plating ruining the factory appearance.
And this kids is why 3D scanning is less costeffective for such simple parts than modeling them in CAD. The time he spent on editing the mesh and reverse engineering the part is probably much greater than just spending 30min modelling the part in CAD accuratly. And then you even would be able to mill this part in a 5 axis CNC from almost any material without the risk of the resin part getting brittle over time in the interrieur of the car and breaking eventually.
Yep. To right. i could have reverse engineered that part without 3D scanning it easily. Some of the parts i've had to do over the years without a 3D Scanner make this thing look a breeze. And thats coming from someone that built his own 3D Scanner 12 years ago.
Watch whole video ;) That scanner is used exactly for reverse 3d engineering professionally. this time they didn't override 3d scan geometry (which save you time measuring object) with cad model, since i guess 3d scan was precise enough for just one part, but they could easily done that, if needed (in case for some larger scale production etc). It is mix of technologies, that works great together. Imo, on the other hand, photogrametry is most usable on larger scale objects like building, statues, terrain etc, especially with drones etc (where lidar solutions costs by factor 10 or 100). I'm in the making of my father's old country house in 3d model, which will be probably prototyped in 3d print, which i will use after as painted plaster casts, ends as a gift for relatives, since house is going to be sold.
Indeed this part is a little simple for scanning but it wouldn't take much more complexity for a part to tip over into scanning being ideal. Also would your model be a perfect copy or just a good enough look alike. There is merit to getting the visible geometry perfect.
I once measured a part with smooth geometries like this one completely manually, and after 2 iterations it was perfect. I think this is often also an option! The technique is a hassle but at least it does not cost 14k!
Honestly, I suspect you can get a result that's even cleaner that way. You can pick up the measurements, the curves and radiuses and so on with measuring and once you start feeding that into CAD you will get pretty close. Of course, it would require some design skill, but you'd arrive at a clean design that I wager would be visually indistinguishable from the original. Still, this was an interesting video about what you can do with 3D scanning, no question about that.
That's how a made a replacement top cover half for the remote for my 2008 DTS. The replacement models I found on the internet didn't look like factory so I measured and drew it from that. Had to tinker with some of the curves, but ended up quite nice.
This is great. I did the same for my father in law. He has a 1982 Kawasaki 1300 and we replicated the air box cover. Used a cheaper Einscan scanner but as mentioned a few times in the comment section it becomes a resource intensive exercise and the cost of the scanner can simply not be justified unless you have a daily use case for it. That being said great video and really hope to see hobby 3D scanners to get better soon.👍🏽
I don't know if you're familiar with SuperfastMatt but he does a tremendous amount of car retrofitting work on his channel and recently used the same exact scanner in a showdown with some other methods. Great to see a video from you on the same topic :)
This is exactly where the magic of 3d modelling and printing kicks in. Being able to design or replicate, then print a durable part from nothing is amazing to me.
He should definitely start selling these vent parts... I know how hard it is to come across rare out-of-production pieces… and when they do come up, it’s insanely expensive.
This video makes me wonder, has anyone successfully got a digital 3d model (be it an STL, OBJ, STEP file) from a manufacturer who no longer produces these? For instance, a broken handle of a trash can and then get the digital file from the manufacturer so you can print it yourself.
I saw that vent in the thumbnail and immediately knew what it was because I'm also restoring a roadster. There used to be a guy on Instagram who made these, but these cars are so uncommon it's pretty hard to have much commercial success in making these parts.
@@matthewlafleur5147 Well, there's that, but there's quite a bit of time investment that goes into quality 3d scans. I'm working for a company that sells a few competing ranges of professional scan tools to the Einscan. While it totally depends on how accurate you need the model to be, how big the part is, how small it's features are, and what material it's made of, you can fully expect the full process to take a couple hours minimum. So even stocking any sort of file inventory isnt really sustainable. It's safer to do it by commission, unless you really love a car and want to put the work in without potential reward. You can totally form a business around this process, but it's nowhere near as cheap as most hobbyist people want to deal with. If you're restoring a whole car, the hardware and software investment is a lot cheaper than relying on someone to do it for you at $100/hr. I am talking, however, producing a perfect and parametric CAD model from the scans to allow for editing and proper printing. As we saw in this video, the shortcut is to scan, merge, fix, and smooth the mesh and just use it raw which is always inaccurate and lumpy even with $40k scanners.
Have you ever wondered why some companys reproduce some car parts that you can buy on ebay pretty cheap, while others don't? Well its related to copyright. There is a reason why that vent doesn't exist anymore.
Some time ago my dad was working on restoring an old jaguar and had a similair problem. Although we were unable to find the original part anywhere, so I modeled a replacement by measuring the imprint it had left on the leather of the car, and my looking at pictures of it on the internet. It came out pretty good, and is a testament why 3d printing can be awesome
I cant wait until 3d scanners are more affordable and give more precise scans for projects just like this. I can think of so many custom parts that would be fun to make. Would really be nice if there was a huge repository of 3d scanned parts and full car scans to work off of.
If you want to scan stuff like that I find that hanging it off fishing line and doing a full spherical photo shoot with someone holding a white blanket behind makes very good results. A good soft flash light like the vloger circle thingy illuminates it very well. The more pictures the better the detail. Also no auto settings on camera :) Meshroom is pretty powerfull.
That would have violated my 80% rule, but I agree it would have been effective. I quite liked experimenting with Meshroom and I'll be looking for an excuse to use it again in future.
@@TeachingTech I'm only using meshroom, with my smartphone. Once tweaked the settings your good to go. I always use sift and akaze with force cpu ticked off ;) I'm still trying to learn how to use CCTAG, With this you can adjust the size and orientation off the object in the app. Worth looking at it ;)
This kind of thing is exactly why I got into 3D printing, structure light, and modeling. Recreating long lost or critical items from raw materials. Well done!
@Techcing Tech I love 3D printing, it was videos like this that made me start my own 3D printing side business, we started off with 2 ender 3 pros and now have 7 FDMs and 5 resin printers, it's great to be living in the 4th industrial revolution 😀
Love the Videos and all the info, I am only 3 months in to 3d printing and the accompanying software, I have just finished a 1960 Lambretta Spanish Speedo face, now working on some Vespa lights , Thank you for all the insight , Dave Uk.
I mentioned to a friend I was going to do the same for my Mk1 Escort dash vent. Next minute he's already measured and designed the vent, he's also printed and sold some sets since..... I need to keep my mouth shut. Scanning looks to save a bunch of time. Great video.
That's awesome! This is exactly why I love 3D Printing so much. I hope you guys made that 3D file freely available so other people who own that car can use it!
It will be released for free. I am compensated for my time from RUclips revenue. I am lucky enough to do this full time, I don't need to get greedy beyond that.
I LOVE IT! I've got a one-off classic car from the 70's and finding certain components have proved to be almost impossible. A huge reason why I bought a Prusa a couple of years ago was for this very thing. Excellent job. I look forward to much more.
Oh i totally agree my father used to have a '59 Volvo 544 that was missing knobs and interior parts one of these 10 years ago would of gone a long way to restore it. It's since rusted away sadly he couldn't work on it anymore and the lack of parts caused it to sit until the motor fell out when my son and I tried to move it.
Cool video! I actually do this for my restoration clients. I scan parts like engine heads, or old swedged tubing (motorcycles) then I 3D print SLA molds that I inject wax into. Then I send the wax to foundries. This part you did here would have been one of the parts I would’ve just remodeled by using images and calipers. Its not that complex, however 3D scanning is a great way to avoid printing prototypes the tweaking the model. Another tip, don’t by the expensive scanning spray, Krylon Dulling spray is $5-6 and does the same thing, you just have to wipe it off afterwards.
Great Work! THIS is the application/work flow that many of us, who primarily use 3D printers as a TOOL for making functional parts, really want to see develop. I can't wait for higher capability scanners to move along the development / cost reduction curve, to the point where they become widely affordable (say, at around the 400-500 price point).
I know that it's not the point of this video, but for this particular part a contour gauge and set of calipers would have resulted in a near-perfect part. For larger objects the scanners are far more suitable of course
Geomagic's other software is capable of producing perfect CAD data, this was the easy/ cheap way out of the process. It's (expensive) package exists to easily let you sketch, modify, and create parametric solid data that can be exported (with feature history) to your big CAD programs, like solidworks. Creating basic STL meshes off the raw data is something that a lot of scanners have included within their software and is a definite waste of the program's cost. It's one of those things where if you're going to spend $15-$20k on a scanner, you may as well have the $5k package that lets you go all the way with it. It's something i'm just getting into at work. It's pretty fascinating, but in no way intuitive to just jump into. So there's that option for companies that want to produce accurate parts faster than by hand measuring.
I agree completely, this isn't that complicated and would only take a few hours to model by hand if you had the part. No lumpiness to sort out either. But the scanner tech is going to be a lot more useful for complicated things like the other examples shown. If anything, the vent is too simple a part to use as a demonstration, you could have just 2-part (or even open-top 1-part) silicone molded it, which is probably how it was made originally.
Photogrammetry with a bigger amount of knowledge and experience can handle such as small part pretty well. Its important to make any side of an object visually different much as possible to get away from destortion and noise in the photo scan. And of course not a crappy smartphone camera. The higher the image quality is, the higher the amount of detail is in the scan. Expecially sharpness. Smartphone cameras likes to wash out images so they look "nicer". But this is the oposite of what photogrammetry needs.
That's exactly what I thought. I understand that he's trying to show the process of 3d scanning but I believe that part is not the best example. The part in question is easy to 3d model, so no need for 3d scan. That been said this is a great video showcasing the capabilities of 3D scanning technology
While I love 3d printing, there are other low tech options for reproducing the unobtainable. For this particular part a simple 2 part silicone mold could have been made. Some shapes are harder to reproduce by mold making due to internal features, overhangs etc. I know this was just an example of the process, but also moldmaking can be very cheap even with using high end mold making materials. So even if you went with the mid-range scanner your way over a budget moldmaking.
While normally I would agree, with this part I wasn't willing to risk pouring anything around it. It was on loan to my best friend's uncle so there was no way I was going to do anything to compromise it. I was even wary about using the canning spray in case it reacted badly with the plastic.
Just for saving cost, it could be worth just taking some measurements and making the 3d model yourself. It might take a few revisions, but this part seems like it would be pretty easy to model
@@laxbrobrandon yeah, measure, model, print, test fit, if everything is good print multiples for moldmaking so stronger/different materials can be produced.
Meanwhile, the 2nd-year Japanese employee that designed the vent cover is scratching his head in wonder. I only have the highest regard for this channel, but remember how imprecise manufacturing can be in the real world. I woulda banged it up in SolidWorks (or Fusion 360) after a weekend of trying to understand what the designer was thinking… including that slight radius on the top.
If I had persevered learning even basic Fusion 360 I would have saved a lot of my time over the last 4 years making re-designs of small parts. I'm "mentally stuck" back at Tinkercad. Certain car and motorcycle trim and access covers look simple until you start really examining them. No two sides or edges are parallel or have a consistent size of radius anywhere. And the backside has different details & part no. lettering that owners will insist on. It really calls for "lofting" skills and I got frustrated and gave up for the time being. Anyway all of these things are good to keep my old grey matter bubbling.
For all those saying the resin printed models will become brittle over time, he now has a resin object that can be used to create a mold without risk of damaging "The Precious". This can then be used to generate more stable models.
I think since this piece is also getting painted (and maybe primed) I think FDM would also have been nice. Especially leaving open "slots" at the back face of the pins to push some tiny steel wire in it (like paperclips) to help with rigidity.
Spot on! I now have the means to produce as many versions as necessary for other methods. For instance, I could resin print in wax like and send it off for investment casting in metal.
Sir, are you a mind reader? =) This is exactly what I needed. Im building a 1951 Land Rover hybrid and Im in the middle of designing defroster vents that fits the style of the vehicle. Thank you for the inspiration!
I would've taken 3 profile pictures, bring them into Fusion, calibrate to size, and draw the part using splines and tweaked them until perfect. Ten minute job, but...that's just me.
@@novo740 I don't know why that's a requirement for a redesigned part vs a scanned part. You can compare either to the original part, and they're both likely to need some refining once they meet the car
I think he tried to show that technology like this is possible. I would like to see you modeling for instance an engine block this way. But I agree. As someone who works as reverse engineer with FARO arm scanner I think this is the type of thing that is too small and easy to be worth scanning if you do have to outsource it.
I've tried that several times to attempt to design a case for my earbud case that looks better and is personalized, but the pictures never line up right even when I had the camera on a tripod and the scale object never moved. I don't know why. Any tips on making it work, or is this all theoretical on your part?
Interestingly, the shorter the wavelength of the laser (blue) the higher the resolving power and definition of the scan, as long as the imaging sensor has a high enough resolution to accommodate the returned reflection detail!!! Remember when they switched from DVD red laser to Blu-ray blue laser? Because higher information density per unit of area!!!😁👍🏻
Red and blue light have wavelengths of 0.7 and 0.4 of a micron respectively, and are about 50x smaller than the resolution of the models, so it really doesn't matter yet. It matters for DVD and Blu-ray because those really do operate at the nano-scale. 3D printing is still orders of magnitude larger.
I would suspect on this case we're dealing a lot more with the reflection properties of materials on blue light vs. any other colors. The wavelength is way, *WAY* beyond the maximum attainable precision of the tool after all.
Nice! I've done a bunch of 3D scanning and reproductions (not auto) using my Einscan (but not this fancy blue laser version, more of a prosumer one), and I've certainly never used this reverse engineering tool. That's very cool to see that such a thing exists. My experience is get a giant scan out of Einscan, then do a bunch of mesh reductions to make the files reasonably sized to edit in Meshmixer. Then meticulous and careful mesh editing to remove scanning mistakes and what not. It is more of an art than a science. And lots of saves in case Meshmixer crashes on a long operation or you edit something you didn't want to.
@@TeachingTech indeed. And it is too bad Meshmixer has been basically deprecated by Autodesk. They say oh you can use F360, yeah, nah, you try to load a STL with 16mil triangles into F360. It isn't fast in Meshmixer, but you can still do stuff with it, in F360, good luck.
I appreciate the educational aim of this project- but in my opinion it's rare that a 3D scan will produce a clean enough object to print, with the exception of an artistic part like a sculpture. 3D scanning a part like this should be used as reference geometry to make a perfect, clean CAD model for printing like you show myles doing at the end. Scanning makes geometry that would be difficult or impossible to measure with calipers very easy to replicate.
@@timwrigley102 I think in some instances it is not enough. For example if you wanted to mirror this grill to fit mirrored holes. I fully agree with Seth, it is far better to not just print but remodel. It also features the possibility of changing a parametric model at the end freely, making multiple design iterations a bliss. At the end, remodeling this part using the scan as reference combines the best of two worlds for a little additional work.
I don’t think anyone will disagree with you, but I think the videos intent was to determine the feasibility. It would be great if a phone could make a perfect 1:1 straight to printer scan, because that’s the future of 3D printing, but instead we learned it takes a $15k scanner and some clean up in CAD. A part like that exceeds the 80/20 approach to learning CAD (or at least mine). So, instead of paying someone to make a bunch of complex CAD models, I can rent one of these for a day and exercise my limited CAD abilities cleaning up and tweaking the last 5% of the model.
There are software packages that allow your traditional parametric CAD data to be produced off of the scan mesh, but it's another $5k give or take. It more or less allows for cross sections to be taken of the mesh data, it then makes construction sketches to your tolerance with the section plane. You can then convert it to a proper sketch as is, or redraw it to simplify it. Add your dimensions and constraints in. Then follow a similar workflow to solidworks to create your geometry. Rinse and repeat. You wind up with the capabilities to make hair-pulling features with ease. When everything is said and done, the part and it's feature tree can be exported to the big CAD programs for delivery or further modification/ analysis. This is kind of a basic explanation of all this, but I think you get the gist. The whole process is time consuming, though much less so than guessing and test printing for sure. It just has a very large upfront investment. And recurring for maintenance and software. It's a big enough process that despite having the tools available for free at work, I dont often go out of my way to go that route. If you look a a part and think 'this is going to be a b---" then it's the tool for the job. You have to factor in whether the scan time is worth it ontop of you modeling, assuming you could just fly though the design process with it's help.
Instead of the specialty spray you can use Dr Scholl's aerosol foot spray. It won't dissolve on its own, but it wipes right off. It would be best for the laser or structured light scanner, and would help some for photogrammetry, though adding a sputter texture is best in that case.
I researched some home remedies for scanning spray such as talk in IPA but in the end I was unwilling to risk the part with somewhat of an unknown formula. Thanks for the tip though.
I've been considering starting a print farm for just basic needs I'm still learning using a cad program and creating objects from scratch but mostly I've been focused on learning the FMD printer for now I don't have a usable work space as yet for the trickier resin printers but this video has given me some ideas for 3D scanning which would be a huge help.
While I haven't watched the video yet past the precious, It seems like a standard solution would've been to make a 2 part silicone mold and make an epoxy casting.
@@S_Roach Care to explain why? Molding in silicone can be a pretty gentle process and TT did spray it multiple times with scanning spray for his attempt
@@RubixB0y Depending on how brittle it is, I could see accidentally pulling one of those tines off, trying to separate the mold from the original. I could be wrong. I've never had the occasion to do silicone molding, but separating the mold Looks, to me, to be an opportunity to snag the original if you're not careful.
@@S_Roach With best practices, a mold release spray is used to prevent any possible sticking even though silicone dosen't stick to non-porous surfaces strongly. When demolding, the silicone can be significantly softer than its full cure durometer and can be peeled away from the form bit by bit. I agree that plastic with UV damage will be more brittle and obviously accidents can happen just like dropping it, I just don't believe the molding process is any more rough than general handling.
ive done this before for my own car haha! i lost a hyper specific part (that luckily there was two of) and was able to fully recreate it in blender taking measurements with a calliper and then 3d printing it, it was perfect and felt like the future!
3D printing is extremely useful, and it's funny how many people don't fully comprehend how useful 3D printing is. I've seen a few automotive projects where 3D printing was involved, and a lot of times, you're left going "uh, I didn't think of using 3D printing to do that." 3D printing replacement parts for vehicles that you can no longer obtain, or you can obtain, but you need a very large amount of money to pay for it. Some times it's more of a part that doesn't exists. I made a part for my motorcycle that is a mount that hold a plug for plugging my motorcycle into a battery maintainer/charger. I did it because where the battery is located on the bike, to access it, I have to use a few tool to unbury it enough to attach the charger's cables. But now I have a plug port right on the frame of the bike where I can just plug in, and it's out of sight unless you're looking for it.
I do hope Miles is going to keep the files for the vent - he may get some business out of it . I know there will be lots of posts stating that they can model it in CAD faster and they probably can but with such a rare part the manual handling has to be kept to the absolute minimum - you wouldn’t be very popular if you broke one of those delicate fins off when trying to measure it as it could very well be the only complete unit in existence! I also hope you will post up a few photos of the vent when installed and of the finished car . Amazing technology ! Thanks for sharing !
I wish there was a contact link to Miles, I have an old wagon I want to try to get converted to electric and with updated internal features like heated seats, backup camera, etc. but to do that I would need to do a TON of work with measuring things or have someone with the right scanner have a go at it.
Myles did this as a favour. The part will be released for free, neither of us are looking to profit. Myles' business name and contact details are purposely withheld, he is not looking to use this video as a marketing tool.
I've watched your videos for a long time and I'm surprised to click the Subscribe button because I thought I was subscribed a long time ago anyway your 3D printing videos I've always been good I was getting burnt out on some of your content but when I saw this integration with cars and restoration really drove me to this video keep it up
Amazing process. It is obvious that you get what you pay for. I imagine with enough time and effort you could get results from any of the scans, but as a business, the cost savings on the pro scanner are worth it.
I agree. To my knowledge, Myles uses his scanner a lot in producing parts for the his business. I imagine a lot less time is required with the vehicles and time in money.
Don't know for other methods but for photogrammetry when scanning a non texture single color object it really helps "adding" texture to it .. the software can analyze it better.. i saw people add paints/draw.. myself i added random light/laser texture and it works great
Thank you for this interesting comparison of scanning devices! Beside choosing the right scanning method, I have encountered the problem of choosing the right printing material. In my experience resin based materials offer good detailing, however minor stability especially in the long run under tough conditions, like in an car. This might be worth a closer look! Best regards, and keep up the good work 👍🏻
I agree it may not be suitable long term. We will see how painting goes to protect from UV but if that is insufficient, we now have the mans to make copies and explore other options.
I'm printing light sockets for a Mahindra 4x4 bumper. Not old but my first time with PETG. After a few settings and a retraction tower I think I have it perfect.
I think you should have tried an iPhone as a budget option (face scanner over lidar for resolution) as a lot of people have this to hand and is actually pretty good
That was seriously interesting , thanks Techy . I'm currently working on a 22 year old motorhome , which despite having a multitude of plastic parts . I've been fortunately enough to either replace or repair on the whole . Though in a couple of instances i have taken to carving styrene . I've been using a plastic welder for the repairs , reinforcing with a stainless mesh . Some mechanical parts have been seriously difficult to source , though , just requires a bit of thinking outside the box . I had been thinking it would handy to have a 3d printer for trim , but i'm not very clever with cad programs . Brutal truth , i'm more of a doer , than thinker , if you catch my drift .
Hey, interesting tip! If you hold your resin print under hot water (around 40-50°C) the support release much easier, but you have to do it before post-curing.
Yes!! I'm doing some parts for my dad's MG and another club members MGA. Most of the parts we have original examples to scan and work from but there have been a few pieces I'm making from old pictures and measurements taken from where the part originally sat. I've even made replacement electrical connectors that have long since been discontinued. Its amazing how many little clips, brackets, and do-dads that we can make with our 3D printers for restoration projects. Never really thought about doing it as a full time business though.. I do it mostly for the challenge but if I can earn a few doll hairs to support other hobbies I'm tickled pink..
I would try to persevere with photogrammetry, but in reality I would take photos of each axis with a ruler in shot, use as a background for Blender and then trace, build and modify a mesh such as a torus to start with, subdivide and extrude the prongs and modify until it matches the profile and scale of the images. Yes, it's a lot of polygon bashing, but the results are optimal and probably worth it to create a master file for mass production.
Would like to see quick follow up vid with the vents completed in the car when done. :) Sounds like the 20K scanner is awesome as, we might even get that sort of tech bubbling down into the home maker in 10-20 years (depending on many factors of course) ^_^
THIS IS EXACTLY WHAT THE AUTO ENTHUSIAST WORLD WAS PROMISED15 YEARS AGO, when commercial 3D protrudes came out I'm paraphrasing a snippet from a hot rod/car craft article "Soon, obsolete parts will be available in the form of a file sent to you, and you will print your own 1971 hemi E body trim bezel"
Well done. I've been using abs to print replacements for old Saabs. It's also great to make radio bezels and similar since aftermarket and custom setups often have fitment problems and look terrible.
We use spray meant to find cracks in parts, cheap and easily cleanable. It essentially creates a dust on top of it. Probably exactly the same as 3d scanning spray, but much cheaper as it's used in industry to find cracks in parts.
I work for a shop doing high end custom builds and restorations. Among other smaller options, we've got a bigrep studio that allows us to do a lot of shit that just wouldn't be possible otherwise. 3d printing is invaluable for modern restorations.
I've started doing this for my business Andes up with two mid range scanners but was surprised just how difficult it was to scan objects and just how much work goes in to fixing the models after a scan. I honestly thought the advertisement of 0.1mm accuracy was meaning it wouldn't miss any small detail, but instead it adds weird bumpy textures and misses entire parts if they aren't the right colour.
This is a cool way to do this. Obviously as others are saying it could be measured and modeled but this is just an example. Also, alot of people have zero 3d modeling experience and don't print anything they didn't grab off thingiverse
Responses to common questions/statements:
- Once the part is verified to fit, I will be releasing the STL for free. I have no desire to profit from this, and Myles had no desire to profit from it either. This is the same Myles that frequently makes 3D printed parts for the Mighty Car Mods boys. He is a generous legend. We are all just car enthusiasts using 3D printing/scanning to enjoy the hobby and help each other.
- Plenty of requests to also try scanner x or y. I don't have an iPhone to try that and I previously compared a Kinect scanner to the Revopoint in that video. The Revopoint is much better so there was no point including the Kinect sensor here.
- As I mentioned in the video, this part will need to be painted before it goes in the car. The resin will become more and more brittle in the sun so will need to be protected. If it still turns out to be too brittle, as some commenters have suggested, a mold can be made to create the part from another technique if necessary.
- Some people are upset that I didn't show the part in place. This is simply a matter of logistics. I've only met the owner of the car a couple of times and we don't live near each other. In case anyone has forgotten, we are still in a pandemic and it's not as straight forward as rocking up somewhere to try the part. It's unlikely I will see the car in person any time soon, but I will update with a pic in the description when I can.
- Plenty of people saying to just model the part. As the video stated, I could have modelled it, but it would need several iterations with test fitting in the car before it was ready. As the point above explains, this was not possible. The scanned part still needed clean up, but I am 100% confident the overall form will fit the dash. It's just a matter of minor refinements where necessary for the 'clip' portions on the underside which I can do remotely. Besides, Myles had his scanner and offered for me to try it. This project came up and I'm sure most would agree the comparison between scanning methods is what the video interesting.
Good of you guys to release it for free! 👌 Car people!
If you need a stronger part I'd suggest printing it in pla and cleaning that up for casting
Fuck the fauxdemic.
Making a silicone mold of the original part and casting resin would probably still be the most accurate compared to scanning, but now you can share it as a file, and the accuracy isn't critical.
Another very precise but time consuming way to 3d scan small objects (though it's easier with conductive objects because you can have an automatic trigger) is with a cnc or a dedicated mechanical scanner and plotting the points. (Basically an AFM but bigger)
You could probably make that with an fdm 3d printer by gluing a needle on a bl touch and making a grid pattern and noting at which point it triggered.
Not sure if the accuracy would be that great though.
Good on you mate. Plenty of internet experts and critics; very few blokes like you who not only make something new, but share it for free. Keep up the great work!
Pair of legends, Myles has been helping sort some of these kind of issues for us since twenty twelve!
I though that straight away when i saw Myles on screen. such a small world
OH MY GOD ITS YOU MCM LOVE YOU GUYS HAVE A GREAT AMAZING DAY
Myles did the bezel for the 240z right and some parts for blackchops?
where can i find more of this miles guy.im also into 3d printing for the auto industry :D
Considering how rare of a part that is, i hope you release an stl so other people can finish their datsuns too. Really amazing to see a damn near replica of a part that no longer exist.
@@WhiteG60 Agreed as there was a lot of work with a $20k tool which makes it reasonable to me to charge a modest fee to help recoup overhead!
@@hcpookie Maybe a kickstarter with stretch goal of opensource stl?
@@Waitwhat469 nah. I doubt there’s that many cars still around to justify such extremes. I assume most owners of such classic cars are of classic age too, so I doubt their understanding of Kickstarter and similar online things
If that was my dato and it was the final bit i needed after a 10 yr 100k+ resto i'd pay a lot more than 50 or 60 us or a modest fee.... this would be a dream come true and i'd happily pay a premium to some one who just took the initiative to go to the trouble to spend prolly 20+ hours with specialist tools.... f^ck i'd prolly give my left nut.
@@hcpookie what overhead? He was given the tool free to use.
The thing I like the most about your videos is that even if I don't fully understand the video's subjects or explanations, I still get some sort of inspiration for something else. In either case, your videos turn out to be extremely helpful regardless.
_You skipped a step that I always perform. I completely remodel the component in CAD, using the scanned mesh as a guide._
_As for painting the final part, I would plate it with satin chrome, to draw attention to them, and as a flex on other Datsun owners._
Now that’s a flex!
That technique is what Myles suggested before he started the scanning. I agree it would likely give the best result but in this I'm hoping it's not necessary.
@@TeachingTech I would like to see the final product. With a good finish I'm sure it will be fine. How concours is a Datsun today? At some point someone at one of the auto aftermarket manufactures will have injection tooling made. Then it will be remodeled.
Nah, the guy who wants that part wants it to look exactly like it just left the Datsun factory. The fellow enthusiasts who know enough to be impressed by his car having perfect vent grilles will notice them without gaudy plating ruining the factory appearance.
Vapour deposition is the only way to finish automotive trim pieces properly, paint will always be a cheap looking and shabby option
And this kids is why 3D scanning is less costeffective for such simple parts than modeling them in CAD. The time he spent on editing the mesh and reverse engineering the part is probably much greater than just spending 30min modelling the part in CAD accuratly. And then you even would be able to mill this part in a 5 axis CNC from almost any material without the risk of the resin part getting brittle over time in the interrieur of the car and breaking eventually.
Yeah...good luck with less effort on making those fins, Mr CNC
Yep. To right. i could have reverse engineered that part without 3D scanning it easily. Some of the parts i've had to do over the years without a 3D Scanner make this thing look a breeze. And thats coming from someone that built his own 3D Scanner 12 years ago.
Watch whole video ;) That scanner is used exactly for reverse 3d engineering professionally. this time they didn't override 3d scan geometry (which save you time measuring object) with cad model, since i guess 3d scan was precise enough for just one part, but they could easily done that, if needed (in case for some larger scale production etc). It is mix of technologies, that works great together.
Imo, on the other hand, photogrametry is most usable on larger scale objects like building, statues, terrain etc, especially with drones etc (where lidar solutions costs by factor 10 or 100). I'm in the making of my father's old country house in 3d model, which will be probably prototyped in 3d print, which i will use after as painted plaster casts, ends as a gift for relatives, since house is going to be sold.
Indeed this part is a little simple for scanning but it wouldn't take much more complexity for a part to tip over into scanning being ideal. Also would your model be a perfect copy or just a good enough look alike. There is merit to getting the visible geometry perfect.
@@izoyt i did what teh whole video. That in no way disproves the points made. And neither does th rest of you post i'm afraid.
I once measured a part with smooth geometries like this one completely manually, and after 2 iterations it was perfect. I think this is often also an option! The technique is a hassle but at least it does not cost 14k!
Honestly, I suspect you can get a result that's even cleaner that way. You can pick up the measurements, the curves and radiuses and so on with measuring and once you start feeding that into CAD you will get pretty close. Of course, it would require some design skill, but you'd arrive at a clean design that I wager would be visually indistinguishable from the original. Still, this was an interesting video about what you can do with 3D scanning, no question about that.
That's how a made a replacement top cover half for the remote for my 2008 DTS. The replacement models I found on the internet didn't look like factory so I measured and drew it from that.
Had to tinker with some of the curves, but ended up quite nice.
This is actually how I got into 3d printing. I needed to make door handle trim for my Plymouth duster. I designed my own though.
Wow, with all that cleanup work, and the lower complexity of the part, it almost makes sense just to have modeled it from scratch.
The part is curved, so good luck modeling it
@@vidznstuff1 That just makes it a little bit more difficult, I don't know why everyone pretends like it would make it wholly impossible.
@Nobody Noone You are correct. I always remodel the component in CAD, using the scanned mesh as a guide.
@@vidznstuff1 It would be easy to recreate a flat version, then apply a curve to it.
This is great. I did the same for my father in law. He has a 1982 Kawasaki 1300 and we replicated the air box cover. Used a cheaper Einscan scanner but as mentioned a few times in the comment section it becomes a resource intensive exercise and the cost of the scanner can simply not be justified unless you have a daily use case for it. That being said great video and really hope to see hobby 3D scanners to get better soon.👍🏽
I don't know if you're familiar with SuperfastMatt but he does a tremendous amount of car retrofitting work on his channel and recently used the same exact scanner in a showdown with some other methods. Great to see a video from you on the same topic :)
The ability to design and make custom car parts or recreate unobtainable ones is what got me into 3D printing in the first place. 😁
This is exactly where the magic of 3d modelling and printing kicks in. Being able to design or replicate, then print a durable part from nothing is amazing to me.
He should definitely start selling these vent parts... I know how hard it is to come across rare out-of-production pieces… and when they do come up, it’s insanely expensive.
This video makes me wonder, has anyone successfully got a digital 3d model (be it an STL, OBJ, STEP file) from a manufacturer who no longer produces these? For instance, a broken handle of a trash can and then get the digital file from the manufacturer so you can print it yourself.
I saw that vent in the thumbnail and immediately knew what it was because I'm also restoring a roadster. There used to be a guy on Instagram who made these, but these cars are so uncommon it's pretty hard to have much commercial success in making these parts.
@@nickdrom I think that's where 3D printing this really shines. They can easily be made as ordered, with almost no overhead.
@@matthewlafleur5147 Well, there's that, but there's quite a bit of time investment that goes into quality 3d scans. I'm working for a company that sells a few competing ranges of professional scan tools to the Einscan. While it totally depends on how accurate you need the model to be, how big the part is, how small it's features are, and what material it's made of, you can fully expect the full process to take a couple hours minimum. So even stocking any sort of file inventory isnt really sustainable. It's safer to do it by commission, unless you really love a car and want to put the work in without potential reward.
You can totally form a business around this process, but it's nowhere near as cheap as most hobbyist people want to deal with. If you're restoring a whole car, the hardware and software investment is a lot cheaper than relying on someone to do it for you at $100/hr.
I am talking, however, producing a perfect and parametric CAD model from the scans to allow for editing and proper printing. As we saw in this video, the shortcut is to scan, merge, fix, and smooth the mesh and just use it raw which is always inaccurate and lumpy even with $40k scanners.
Have you ever wondered why some companys reproduce some car parts that you can buy on ebay pretty cheap, while others don't? Well its related to copyright. There is a reason why that vent doesn't exist anymore.
Some time ago my dad was working on restoring an old jaguar and had a similair problem.
Although we were unable to find the original part anywhere, so I modeled a replacement by measuring the imprint it had left on the leather of the car, and my looking at pictures of it on the internet.
It came out pretty good, and is a testament why 3d printing can be awesome
I cant wait until 3d scanners are more affordable and give more precise scans for projects just like this. I can think of so many custom parts that would be fun to make. Would really be nice if there was a huge repository of 3d scanned parts and full car scans to work off of.
The Pop is as affordable as they get
That's a cool handheld, we use gom triple scan at work for checking parts.
If you want to scan stuff like that I find that hanging it off fishing line and doing a full spherical photo shoot with someone holding a white blanket behind makes very good results. A good soft flash light like the vloger circle thingy illuminates it very well. The more pictures the better the detail. Also no auto settings on camera :) Meshroom is pretty powerfull.
That would have violated my 80% rule, but I agree it would have been effective. I quite liked experimenting with Meshroom and I'll be looking for an excuse to use it again in future.
@@TeachingTech I'm only using meshroom, with my smartphone. Once tweaked the settings your good to go. I always use sift and akaze with force cpu ticked off ;)
I'm still trying to learn how to use CCTAG,
With this you can adjust the size and orientation off the object in the app. Worth looking at it ;)
This is just fucking unbelievable. It blows my mind that I can think of something, draw it, print it, and then it exists. It's amazing.
Don't even own a 3d printer and found this extremely interesting.
In my 60s but I love technology so I found this to be pretty cool.
This kind of thing is exactly why I got into 3D printing, structure light, and modeling. Recreating long lost or critical items from raw materials. Well done!
A great comparison of current 3d scanning options available today. Very informative... Thank you.
@Techcing Tech I love 3D printing, it was videos like this that made me start my own 3D printing side business, we started off with 2 ender 3 pros and now have 7 FDMs and 5 resin printers, it's great to be living in the 4th industrial revolution 😀
Love the Videos and all the info, I am only 3 months in to 3d printing and the accompanying software, I have just finished a 1960 Lambretta Spanish Speedo face, now working on some Vespa lights , Thank you for all the insight , Dave Uk.
I mentioned to a friend I was going to do the same for my Mk1 Escort dash vent.
Next minute he's already measured and designed the vent, he's also printed and sold some sets since.....
I need to keep my mouth shut.
Scanning looks to save a bunch of time. Great video.
I learned long ago that if there is something I need that is not available from any source, there are many others with the same need.
That's awesome! This is exactly why I love 3D Printing so much. I hope you guys made that 3D file freely available so other people who own that car can use it!
No he should sell it. His labor has value.
It will be released for free. I am compensated for my time from RUclips revenue. I am lucky enough to do this full time, I don't need to get greedy beyond that.
I LOVE IT! I've got a one-off classic car from the 70's and finding certain components have proved to be almost impossible. A huge reason why I bought a Prusa a couple of years ago was for this very thing.
Excellent job. I look forward to much more.
Oh i totally agree my father used to have a '59 Volvo 544 that was missing knobs and interior parts one of these 10 years ago would of gone a long way to restore it. It's since rusted away sadly he couldn't work on it anymore and the lack of parts caused it to sit until the motor fell out when my son and I tried to move it.
Cool video! I actually do this for my restoration clients. I scan parts like engine heads, or old swedged tubing (motorcycles) then I 3D print SLA molds that I inject wax into. Then I send the wax to foundries. This part you did here would have been one of the parts I would’ve just remodeled by using images and calipers. Its not that complex, however 3D scanning is a great way to avoid printing prototypes the tweaking the model.
Another tip, don’t by the expensive scanning spray, Krylon Dulling spray is $5-6 and does the same thing, you just have to wipe it off afterwards.
Awesome, this kind of videos are the best. I feel pretty good when I finish a piece that I couldn't find anywhere else
Worked a treat, I was interested in the Einscan series but they said not suitable, wanted to scan larger objects.
Great Work! THIS is the application/work flow that many of us, who primarily use 3D printers as a TOOL for making functional parts, really want to see develop. I can't wait for higher capability scanners to move along the development / cost reduction curve, to the point where they become widely affordable (say, at around the 400-500 price point).
3D printing can definitely save older vehicles especially with parts that are discontinued. This is the so freaking amazing
I know that it's not the point of this video, but for this particular part a contour gauge and set of calipers would have resulted in a near-perfect part. For larger objects the scanners are far more suitable of course
Geomagic's other software is capable of producing perfect CAD data, this was the easy/ cheap way out of the process. It's (expensive) package exists to easily let you sketch, modify, and create parametric solid data that can be exported (with feature history) to your big CAD programs, like solidworks. Creating basic STL meshes off the raw data is something that a lot of scanners have included within their software and is a definite waste of the program's cost. It's one of those things where if you're going to spend $15-$20k on a scanner, you may as well have the $5k package that lets you go all the way with it.
It's something i'm just getting into at work. It's pretty fascinating, but in no way intuitive to just jump into. So there's that option for companies that want to produce accurate parts faster than by hand measuring.
I agree completely, this isn't that complicated and would only take a few hours to model by hand if you had the part. No lumpiness to sort out either. But the scanner tech is going to be a lot more useful for complicated things like the other examples shown. If anything, the vent is too simple a part to use as a demonstration, you could have just 2-part (or even open-top 1-part) silicone molded it, which is probably how it was made originally.
Photogrammetry with a bigger amount of knowledge and experience can handle such as small part pretty well. Its important to make any side of an object visually different much as possible to get away from destortion and noise in the photo scan. And of course not a crappy smartphone camera. The higher the image quality is, the higher the amount of detail is in the scan. Expecially sharpness. Smartphone cameras likes to wash out images so they look "nicer". But this is the oposite of what photogrammetry needs.
That's exactly what I thought. I understand that he's trying to show the process of 3d scanning but I believe that part is not the best example. The part in question is easy to 3d model, so no need for 3d scan. That been said this is a great video showcasing the capabilities of 3D scanning technology
While I love 3d printing, there are other low tech options for reproducing the unobtainable.
For this particular part a simple 2 part silicone mold could have been made. Some shapes are harder to reproduce by mold making due to internal features, overhangs etc.
I know this was just an example of the process, but also moldmaking can be very cheap even with using high end mold making materials.
So even if you went with the mid-range scanner your way over a budget moldmaking.
While normally I would agree, with this part I wasn't willing to risk pouring anything around it. It was on loan to my best friend's uncle so there was no way I was going to do anything to compromise it. I was even wary about using the canning spray in case it reacted badly with the plastic.
Just for saving cost, it could be worth just taking some measurements and making the 3d model yourself. It might take a few revisions, but this part seems like it would be pretty easy to model
@@laxbrobrandon yeah, measure, model, print, test fit, if everything is good print multiples for moldmaking so stronger/different materials can be produced.
Awesome project, well done. Many times making the model is the hardest part and a good tool makes all the difference
Thanks for the time you put into this. Inspirational. Educational. Please keep up the good work.
Meanwhile, the 2nd-year Japanese employee that designed the vent cover is scratching his head in wonder. I only have the highest regard for this channel, but remember how imprecise manufacturing can be in the real world. I woulda banged it up in SolidWorks (or Fusion 360) after a weekend of trying to understand what the designer was thinking… including that slight radius on the top.
If I had persevered learning even basic Fusion 360 I would have saved a lot of my time over the last 4 years making re-designs of small parts. I'm "mentally stuck" back at Tinkercad. Certain car and motorcycle trim and access covers look simple until you start really examining them. No two sides or edges are parallel or have a consistent size of radius anywhere. And the backside has different details & part no. lettering that owners will insist on. It really calls for "lofting" skills and I got frustrated and gave up for the time being.
Anyway all of these things are good to keep my old grey matter bubbling.
Had a '69 MGB in high school. Access to a 3d printer would have been an absolute game changer.
For all those saying the resin printed models will become brittle over time, he now has a resin object that can be used to create a mold without risk of damaging "The Precious". This can then be used to generate more stable models.
I think since this piece is also getting painted (and maybe primed) I think FDM would also have been nice. Especially leaving open "slots" at the back face of the pins to push some tiny steel wire in it (like paperclips) to help with rigidity.
Spot on! I now have the means to produce as many versions as necessary for other methods. For instance, I could resin print in wax like and send it off for investment casting in metal.
Sir, are you a mind reader? =) This is exactly what I needed. Im building a 1951 Land Rover hybrid and Im in the middle of designing defroster vents that fits the style of the vehicle. Thank you for the inspiration!
I would've taken 3 profile pictures, bring them into Fusion, calibrate to size, and draw the part using splines and tweaked them until perfect.
Ten minute job, but...that's just me.
Cool, and you can do that. Would make for a pretty shit video though.
There is a caption that says that he doesn't have access to the car for test fits.
@@novo740 I don't know why that's a requirement for a redesigned part vs a scanned part. You can compare either to the original part, and they're both likely to need some refining once they meet the car
I think he tried to show that technology like this is possible. I would like to see you modeling for instance an engine block this way. But I agree. As someone who works as reverse engineer with FARO arm scanner I think this is the type of thing that is too small and easy to be worth scanning if you do have to outsource it.
I've tried that several times to attempt to design a case for my earbud case that looks better and is personalized, but the pictures never line up right even when I had the camera on a tripod and the scale object never moved. I don't know why. Any tips on making it work, or is this all theoretical on your part?
Interestingly, the shorter the wavelength of the laser (blue) the higher the resolving power and definition of the scan, as long as the imaging sensor has a high enough resolution to accommodate the returned reflection detail!!! Remember when they switched from DVD red laser to Blu-ray blue laser? Because higher information density per unit of area!!!😁👍🏻
Red and blue light have wavelengths of 0.7 and 0.4 of a micron respectively, and are about 50x smaller than the resolution of the models, so it really doesn't matter yet.
It matters for DVD and Blu-ray because those really do operate at the nano-scale. 3D printing is still orders of magnitude larger.
I would suspect on this case we're dealing a lot more with the reflection properties of materials on blue light vs. any other colors.
The wavelength is way, *WAY* beyond the maximum attainable precision of the tool after all.
More classic car content please!
Now those are the practical print stories I like to hear of. Great work!
You are a boss, well done on making the last part. You are truly a tech priest of the heights order
Nice! I've done a bunch of 3D scanning and reproductions (not auto) using my Einscan (but not this fancy blue laser version, more of a prosumer one), and I've certainly never used this reverse engineering tool. That's very cool to see that such a thing exists. My experience is get a giant scan out of Einscan, then do a bunch of mesh reductions to make the files reasonably sized to edit in Meshmixer. Then meticulous and careful mesh editing to remove scanning mistakes and what not. It is more of an art than a science. And lots of saves in case Meshmixer crashes on a long operation or you edit something you didn't want to.
I think that part of the software is where much of the cost comes from. Meshmixer is powerful but as you say it can be quite unreliable.
@@TeachingTech indeed. And it is too bad Meshmixer has been basically deprecated by Autodesk. They say oh you can use F360, yeah, nah, you try to load a STL with 16mil triangles into F360. It isn't fast in Meshmixer, but you can still do stuff with it, in F360, good luck.
Einscan would be a wonderful addition to a modern maker space.
I appreciate the educational aim of this project- but in my opinion it's rare that a 3D scan will produce a clean enough object to print, with the exception of an artistic part like a sculpture. 3D scanning a part like this should be used as reference geometry to make a perfect, clean CAD model for printing like you show myles doing at the end. Scanning makes geometry that would be difficult or impossible to measure with calipers very easy to replicate.
Way easier to pull a silicon mold. Then you could pop out as many resin copies as you wanted.
@@timwrigley102 I think in some instances it is not enough. For example if you wanted to mirror this grill to fit mirrored holes.
I fully agree with Seth, it is far better to not just print but remodel. It also features the possibility of changing a parametric model at the end freely, making multiple design iterations a bliss. At the end, remodeling this part using the scan as reference combines the best of two worlds for a little additional work.
Correctomundo
I don’t think anyone will disagree with you, but I think the videos intent was to determine the feasibility. It would be great if a phone could make a perfect 1:1 straight to printer scan, because that’s the future of 3D printing, but instead we learned it takes a $15k scanner and some clean up in CAD.
A part like that exceeds the 80/20 approach to learning CAD (or at least mine). So, instead of paying someone to make a bunch of complex CAD models, I can rent one of these for a day and exercise my limited CAD abilities cleaning up and tweaking the last 5% of the model.
There are software packages that allow your traditional parametric CAD data to be produced off of the scan mesh, but it's another $5k give or take.
It more or less allows for cross sections to be taken of the mesh data, it then makes construction sketches to your tolerance with the section plane. You can then convert it to a proper sketch as is, or redraw it to simplify it. Add your dimensions and constraints in. Then follow a similar workflow to solidworks to create your geometry. Rinse and repeat. You wind up with the capabilities to make hair-pulling features with ease. When everything is said and done, the part and it's feature tree can be exported to the big CAD programs for delivery or further modification/ analysis. This is kind of a basic explanation of all this, but I think you get the gist.
The whole process is time consuming, though much less so than guessing and test printing for sure. It just has a very large upfront investment. And recurring for maintenance and software.
It's a big enough process that despite having the tools available for free at work, I dont often go out of my way to go that route. If you look a a part and think 'this is going to be a b---" then it's the tool for the job. You have to factor in whether the scan time is worth it ontop of you modeling, assuming you could just fly though the design process with it's help.
Instead of the specialty spray you can use Dr Scholl's aerosol foot spray. It won't dissolve on its own, but it wipes right off. It would be best for the laser or structured light scanner, and would help some for photogrammetry, though adding a sputter texture is best in that case.
I researched some home remedies for scanning spray such as talk in IPA but in the end I was unwilling to risk the part with somewhat of an unknown formula. Thanks for the tip though.
Your always on the right time
I've been considering starting a print farm for just basic needs I'm still learning using a cad program and creating objects from scratch but mostly I've been focused on learning the FMD printer for now I don't have a usable work space as yet for the trickier resin printers but this video has given me some ideas for 3D scanning which would be a huge help.
Great project 👍
Thanks for sharing your experiences with all of us :-)
Thank you for the walk thru of the various approaches.
Thank you very much for this and your time
While I haven't watched the video yet past the precious, It seems like a standard solution would've been to make a 2 part silicone mold and make an epoxy casting.
Depending on just how sun damaged the part was, that could have been a risky move.
@@S_Roach Care to explain why? Molding in silicone can be a pretty gentle process and TT did spray it multiple times with scanning spray for his attempt
@@RubixB0y Depending on how brittle it is, I could see accidentally pulling one of those tines off, trying to separate the mold from the original.
I could be wrong. I've never had the occasion to do silicone molding, but separating the mold Looks, to me, to be an opportunity to snag the original if you're not careful.
@@S_Roach With best practices, a mold release spray is used to prevent any possible sticking even though silicone dosen't stick to non-porous surfaces strongly. When demolding, the silicone can be significantly softer than its full cure durometer and can be peeled away from the form bit by bit. I agree that plastic with UV damage will be more brittle and obviously accidents can happen just like dropping it, I just don't believe the molding process is any more rough than general handling.
… it’s a 3D printing channel
That’s probably why he 3D printed it
ive done this before for my own car haha! i lost a hyper specific part (that luckily there was two of) and was able to fully recreate it in blender taking measurements with a calliper and then 3d printing it, it was perfect and felt like the future!
3D printing is extremely useful, and it's funny how many people don't fully comprehend how useful 3D printing is. I've seen a few automotive projects where 3D printing was involved, and a lot of times, you're left going "uh, I didn't think of using 3D printing to do that." 3D printing replacement parts for vehicles that you can no longer obtain, or you can obtain, but you need a very large amount of money to pay for it. Some times it's more of a part that doesn't exists. I made a part for my motorcycle that is a mount that hold a plug for plugging my motorcycle into a battery maintainer/charger. I did it because where the battery is located on the bike, to access it, I have to use a few tool to unbury it enough to attach the charger's cables. But now I have a plug port right on the frame of the bike where I can just plug in, and it's out of sight unless you're looking for it.
I do hope Miles is going to keep the files for the vent - he may get some business out of it .
I know there will be lots of posts stating that they can model it in CAD faster and they probably can but with such a rare part the manual handling has to be kept to the absolute minimum - you wouldn’t be very popular if you broke one of those delicate fins off when trying to measure it as it could very well be the only complete unit in existence!
I also hope you will post up a few photos of the vent when installed and of the finished car .
Amazing technology ! Thanks for sharing !
I wish there was a contact link to Miles, I have an old wagon I want to try to get converted to electric and with updated internal features like heated seats, backup camera, etc. but to do that I would need to do a TON of work with measuring things or have someone with the right scanner have a go at it.
@@MegaDman42
Try sending Michael a message .
I don’t even want to guess what scanning cost - I imagine it wouldn’t be cheap .
Myles did this as a favour. The part will be released for free, neither of us are looking to profit. Myles' business name and contact details are purposely withheld, he is not looking to use this video as a marketing tool.
Love how your introduction music/sequence is still showing homage to your first upload.
Incredible work! I love to see care and effort put into restoring historical cars!
I've watched your videos for a long time and I'm surprised to click the Subscribe button because I thought I was subscribed a long time ago anyway your 3D printing videos I've always been good I was getting burnt out on some of your content but when I saw this integration with cars and restoration really drove me to this video keep it up
Myles is awesome, what a legend :)
Amazing content, so great to see how quick scanning technology has advanced!
Amazing process. It is obvious that you get what you pay for. I imagine with enough time and effort you could get results from any of the scans, but as a business, the cost savings on the pro scanner are worth it.
I agree. To my knowledge, Myles uses his scanner a lot in producing parts for the his business. I imagine a lot less time is required with the vehicles and time in money.
Don't know for other methods but for photogrammetry when scanning a non texture single color object it really helps "adding" texture to it .. the software can analyze it better.. i saw people add paints/draw.. myself i added random light/laser texture and it works great
Thank you for this interesting comparison of scanning devices! Beside choosing the right scanning method, I have encountered the problem of choosing the right printing material. In my experience resin based materials offer good detailing, however minor stability especially in the long run under tough conditions, like in an car. This might be worth a closer look! Best regards, and keep up the good work 👍🏻
I agree it may not be suitable long term. We will see how painting goes to protect from UV but if that is insufficient, we now have the mans to make copies and explore other options.
I'm printing light sockets for a Mahindra 4x4 bumper. Not old but my first time with PETG. After a few settings and a retraction tower I think I have it perfect.
Car restoration is about to get a lot more interesting with all this new tech! Love it!
This tech is transforming the classic car world !
I just got the pop 2 premium package for 735 using a coupan, and so far I’m happy with my first scan results. Cant wait to try scanning everything
This is exactly why 3d printing and scanning is an amazin tech.
I think you should have tried an iPhone as a budget option (face scanner over lidar for resolution) as a lot of people have this to hand and is actually pretty good
I've used meshroom and a phone cam quite often to get a baseline to clean up
Check out Superfast Matt's recent video for the iPhone Pro comparison.
Really?
I don't have an iPhone so to buy one for this would definitely not be budget for me.
That was seriously interesting , thanks Techy . I'm currently working on a 22 year old motorhome , which despite having a multitude of plastic parts . I've been fortunately enough to either replace or repair on the whole . Though in a couple of instances i have taken to carving styrene . I've been using a plastic welder for the repairs , reinforcing with a stainless mesh . Some mechanical parts have been seriously difficult to source , though , just requires a bit of thinking outside the box .
I had been thinking it would handy to have a 3d printer for trim , but i'm not very clever with cad programs . Brutal truth , i'm more of a doer , than thinker , if you catch my drift .
Hey, interesting tip! If you hold your resin print under hot water (around 40-50°C) the support release much easier, but you have to do it before post-curing.
shall be named "the precious" from now on. had me rolling!!
Yes!! I'm doing some parts for my dad's MG and another club members MGA. Most of the parts we have original examples to scan and work from but there have been a few pieces I'm making from old pictures and measurements taken from where the part originally sat. I've even made replacement electrical connectors that have long since been discontinued. Its amazing how many little clips, brackets, and do-dads that we can make with our 3D printers for restoration projects. Never really thought about doing it as a full time business though.. I do it mostly for the challenge but if I can earn a few doll hairs to support other hobbies I'm tickled pink..
yess!!!! we need more complete datsuns on the road!
I would try to persevere with photogrammetry, but in reality I would take photos of each axis with a ruler in shot, use as a background for Blender and then trace, build and modify a mesh such as a torus to start with, subdivide and extrude the prongs and modify until it matches the profile and scale of the images.
Yes, it's a lot of polygon bashing, but the results are optimal and probably worth it to create a master file for mass production.
Or do the same in a real mechanical CAD Program to get a real parametric CAD model instead if a mesh
@@hansdietrich83 That would work too, but I would use Blender because I know it best.
Would like to see quick follow up vid with the vents completed in the car when done. :) Sounds like the 20K scanner is awesome as, we might even get that sort of tech bubbling down into the home maker in 10-20 years (depending on many factors of course) ^_^
THIS IS EXACTLY WHAT THE AUTO ENTHUSIAST WORLD WAS PROMISED15 YEARS AGO, when commercial 3D protrudes came out
I'm paraphrasing a snippet from a hot rod/car craft article
"Soon, obsolete parts will be available in the form of a file sent to you, and you will print your own 1971 hemi E body trim bezel"
Nice, this is great for restoration!
Excellent as always, Michael. Thank you for making great content like this, the wealth of practical knowledge within is invaluable.
Well done. I've been using abs to print replacements for old Saabs. It's also great to make radio bezels and similar since aftermarket and custom setups often have fitment problems and look terrible.
Thanks for this great video !
We use spray meant to find cracks in parts, cheap and easily cleanable. It essentially creates a dust on top of it. Probably exactly the same as 3d scanning spray, but much cheaper as it's used in industry to find cracks in parts.
Even if it's a #short video, I'd love to see the finished project (both the vent in the dash, and the car itself)
I work for a shop doing high end custom builds and restorations. Among other smaller options, we've got a bigrep studio that allows us to do a lot of shit that just wouldn't be possible otherwise. 3d printing is invaluable for modern restorations.
Awesome walkthrough and your lucky with friends with more equipment. Now make other datsun owners happy :)
My friend owns this Datsun . It's a convertible...he digs it . He repainted it red .
This right here is the power of 3D printing and the untapped possibilities to come. Star Trek early stages of replication.
Wow that scanner is amazing
Amateur scanner. You should see the scantech simscan.
make sure to paint the printed part with a uv resistant clear coat if you aren't painting it so that the resin doesn't degrade as fast.
I've started doing this for my business Andes up with two mid range scanners but was surprised just how difficult it was to scan objects and just how much work goes in to fixing the models after a scan. I honestly thought the advertisement of 0.1mm accuracy was meaning it wouldn't miss any small detail, but instead it adds weird bumpy textures and misses entire parts if they aren't the right colour.
3D scanning is a tricky technology but thins are improving vey fast.
Brilliant! Great use of the tech. Fabulous car too, I'm envious.
S15 dash scanned awesome 👌
This is a cool way to do this. Obviously as others are saying it could be measured and modeled but this is just an example. Also, alot of people have zero 3d modeling experience and don't print anything they didn't grab off thingiverse
Generally we scan and then load the model as a reference to parametrically model the part to print.
Aww I was hoping to see the finished version on the cars ^^. Hope you do a followup video on this. Cheers.
Awesome work mate
I used autocad recap in college. Take lot of pictures and it puts them together. Seemed to work good if you do enough pictures and they are good