They look like they fit the form fairly well. Considering it was modelled from photos and a couple of dimensions I'm surprised it needed only a little trimming!

It's a little hard to see where I need to adjust the model. Would you be able to sketch on my 3D viewport images where it needs modification? That last photo looks as though the top lip doesn't extend forward enough.

I'll pm you my email address. It's a public holiday here on Mon but I can tweak it Tuesday and get you an update.

If anyone needs a test model for the GV lip I will have my NA bumper off the car and can modify a prototype and/or take pictures as needed. The top part is smashed but still pretty close on the bottom of the bumper.

Madjak and I have been working on it a bit. He is pretty busy right now though.

yeah sorry guys... been a bit busy trying to get my track car sorted for weekend of motorsport next weekend. After that I'll have some spare time again.

I am going to be doing full 3d scan of my car in about a month (using 4 kinect v2 sensors). I can post the files from that so that people can build a more accurate model.

Could I do a scan of the air inlet with a single kinect?

You could but I don't find the kinect v1 to be accurate enough for much in terms of critical measurements and parts fit. The data is too noisey.

I meant a single v2 kinect

My dream is a nextengine 3d scanner and a decent 3d printer, are there any of the latter in the <$500 range that are capable of printing parts this size reliably?

I built a early rep-rap printer that was largely useless. It was the steel allthread with printed clamping nuts, except they were extruded plastic versions. Sloppy piece of crap couldn't lay more than 1/2" before going all lopsided...

I personally think the up mini is the only printer worth its money in that range

I'm moderately happy with my Printrbot | Affordable high resolution 3D printers. I think I need a heated bed though.

You can do photo-modelling quite accurately with a good setup. So you place markers on the surface you want to track, and then you take a series photos in a circle around the surface and use software to either photo-model or track the markers.

I use an application called Syntheyes to track marker information and can get accuracy around the 1mm mark for most surfaces. I use this at work to place 3D accurately into helicopter or drone footage, or to modify objects moving in the video footage, like people's heads.

When I get some spare time, I'm going to make some 3D printed canyards for the front spoiler. I'll use this technique to get an accurate surface model of the front spoiler and wheel arch and then model over the top in 3D for the canyard. This process should be fairly accurate and reduce the number of iterations required to get the model correct.

I tried using a 3d webcam, a laser level, a printed grid, and a free program called david3d. If I had worked with it more I imagine it would have worked great!

https://www.miataturbo.net/attachmen...ine=1432623407

https://www.miataturbo.net/attachmen...ine=1432623407

This was what seems like ages ago though, half a decade. I had great plans to build a rotary table and a servo or gear driven motor for the laser sweep.

No go on a car though, though they had a structured light scan option. If only I had a projector!

So how did those duct adapters work out for you guys? Did they hold up over the summer?

Not sure. Madjack got real busy. I'm not very good with solidworks. They were close but I haven't used them yet.

I've had some ducting sitting on my bench for 2 months waiting to be installed I just haven't had a chance to check fitment. I actually jacked my car up yesterday to have a look as I just need to tweak the R-lip ones to fit past my custom radiator mounts.

I'm going to trim my radiator hanger so I can check the fitment properly, but it looks like they would fit a stock car nicely.

The GV-lip version needed some mods. Aidandj, I can have a look at it this week.

Also for 3D scanning, try Agisoft photoscan. It doesn't work too well with reflective surfaces though. Might have to get your car really dirty!

nice work. Once finalized, would these parts be used as printed, or used for a foundation to mold CF/FG?

Can someone print me a set ^_^

Did these ever get finalized? I will probably take a stab at it soon with an expanding foam pour followed by a 3d scan. I'll share the scan volume when I do.

Also, on the subject of scanning- For the cost photogrammetry is the way to go. I have worked with data from high end lidar systems, kinect scans, and good photoscan setups. The photoscans are the best option for relatively small parts where super tight tolerances aren't needed.. Agisoft photoscan works very well if you prep surfaces, take careful photos, and spend some time editing the point cloud. This example is about 100 photos and I spent 20 minutes cleaning up the point cloud. You can get better results with more time editing the point cloud and masking your photos, but for dimensional reference this is plenty.

https://www.miataturbo.net/attachmen...ine=1465315608

Clean model built from scan reference:
https://www.miataturbo.net/attachmen...ine=1465315608

This is amazing. How can I get into this line of work?