[dataway]

I've got the first piece ... 68 AM radio knob. Modeled in Inventor. "Life Size" CAD model, no scaling, mic'ed up an OEM piece.

Only 10,000 more parts to go ... so everyone start modeling

Good start would be finding the CAD models that the plastic model builders use to mold their models.

[242177P]

67-72, headlight, wiper, lighter, radio, 8-track, and even B-body heater control.

Billet aluminum, but *without* the ridiculous bling?

[dataway]

If you mean without the flutes and concentric rings ... probably pretty simple for someone that knows how to use a CNC lathe. With the bling ... jeez, probably five axis CNC work.

Actually the file I created for that knob (to 3D print it) would translate fine to CAM software for machining. Easy enough to remove the fancy features.

But, how to machine the notched hole for the stem on a radio I have no idea. Unless you just went with a round hole and added a tiny set screw for the stem flat. 68 of course were injection molded plastic ... and the older metal knobs were probably cast.

However if you want crude plastic knobs in various colors of the right size and shape I can provide those Or "custom" shapes, like hearts, moons and clovers

Here is a first attempt at printing one, I can do about 50% better, I was setup for printing "functional" parts at the time, not detailed parts. I'll spare you the details. The shiny black plastic did not photo well, I'll print out another before long in a silver plastic. Certainly not a piece a person would be proud to have in their vehicle ... but as opposed to none because a part couldn't be found ... it would work.

[Half-Inch Stud]

Promising print; lighting? Flutes give style, but get dirty and tough to clean. Seems once i clean em, they stay rather clean.

[HoovDaddy]

Impressive! Nice work.

[dataway]

Messing with some TPU filament now. This material is commonly used in modern car bushing, bumpers, gaskets, "grippy" things etc.

Drew up this fuel pump gasket (from memory so not accurate) to test it. Basically a wonderful, flexible gasket. If you can draw it up in any CAD program that will export an STL file you can create any gasket you want, in any thickness. Add a sealing lip/ridge, change the bolt hole sizes etc. If you have a flatbed document scanner you can scan a gasket, bring it into a CAD program, change it the way you want and print them out by the dozen if you want.

This material is oil, grease and solvent resistant, Shore 95A hardness. In my opinion the max temp would be about 200 F, some say 210 F. So that would preclude it from being used in most engine applications where it comes into contact with the engine block. But, suitable for valve cover grommets, air cleaner fittings, HVAC gaskets, body gaskets (door handles etc.), various small bumpers, bushings such as sway bar, sway bar struts, shocks, vibration dampening, sealing washers, small weather strip pieces, elastic cable holders etc.

Tired of flimsy, ill fitting door handle, side mirror, key hole gaskets? Make your own. Need a slightly larger pull through bumper (console door, glove box door etc) draw one up and print it. Want robust HVAC gaskets instead of that totally crappy foam that is sold ... print your own. From about .030" to inches thick. Want to embed a washer into a bumper? (hood/cowl bumper) design the part with a cavity for the washer ... stop the print a the right place, insert the washer ... continue the print ... and produce a proper bumper with the encased washer.

Yes I'm kind of excited about this stuff

[242177P]

For a while, some repro pedal pads were like hard plastic. I'm thinking grippy might not last 100,000 miles, but a wear item that you replace regularly?

Speaking of wear...

[Shiny]

That has to be a lot of fun! You are obviously keeping yourself occupied!

The CAD model is impressive. You've done this before.

The gasket ideas are a great use for a 3D printer. Good thinking!

I'm surprised you can print something that will hold up to 200F. That's quite impressive.

Have you ever messed with electroless plating? I know you have electroplated some parts. Are any of the 3D print materials conductive enough to electroplate?

I have also thought about ways to smooth out a rough 3D printed part. Maybe solvent vapors? We used to reflow solder with freon vapor before they learned we were destroying the ozone layer. Maybe tumbling in a fine abrasive slurry? Are there "commercial" solutions for this? I know the printer resolution is getting better but my zero knowledge suggested it was still difficult to print a mirror.

Fun stuff!

[Shiny]

Do you use any 3D scanning software? You mentioned starting from files used to develop model cars. How hard would it be to scan a real part, generate a model, then clean it up in your CAD software? If it was relatively easy, your dream of building a CAD model for each of the 10,000 parts in your car could be closer to reality.

I'm ignorant but know there are many software tools that combine 2D scans/images of a part to generate a 3D model. For example, I have a Microsoft Kinect and PC interface. I know they have software to use it for 3D model generation. Might be fun to play with. I suspect there is way too much geometric distortion in a cell-phone camera but maybe not if you can "adjust" the imported model in your CAD world. To me, part designs back in the sixties and seventies were based a lot more on simple and "true" geometry (lines and arcs) than today where NC tools can follow geometries a lathe and mill never could. This might make it easier to make an accurate model as you did by direct measurement of your knob.

I always stop going down this mental path when I see the cost of CAD software like your Inventor....

Mike

[dataway]

I've been using CAD software for 20 plus years ... still a newbie really though, as I don't do it for a living.

You can at the moment print nylon, carbon fiber nylon, ABS, polycarbonate etc. for high temp products. Nylon prints at around 480 degree F and can easily live in an environment north of 300 F, ABS will handle 250+ F . Most of the filaments I use print at min 360 degrees F so their actual melting points are pretty high .. but degradation and creep will take place at lower temps.

Many filaments are available with carbon fill, copper fill, iron fill which does make them conductive .... so now you have cursed me with another project ... trying to electroplate something made of conductive filament ... which of course leads to chrome plated plastic.

Some filaments can be smoothed with an acetone vapor process ... basically suspend the item above a bath of acetone ... timing can be critical as the part surface can run if left too long. And most of the plastics will sand reasonably well. Most of the parts I'm posting photos of are printed in "draft" quality, not in the slower, higher definition modes ... which still isn't "smooth" but it's better than what I'm showing ... downside is it takes quite a bit longer.

I've looked into laser scanners ... problem is that a decent one is still about $5k. The Amazon variety for a few hundred appear to be very limited in usefulness other than for scanning the general shape. There are also high end probe type scanners that physically probe a part to generate the file ... but they have obvious limitations for cavities and hard to reach areas. Basically, if you have the money, it can be done and done very accurately .... a LOT of money. And yes this is often used for scanning in entire autobodies etc.

A lot of the CAD software I use is .... well, "obtained" and "cracked" by processes I'm not proud of. Spent a LOT of years on the internet, been around computers a long time, the stuff can be had, but considerable precautions must be taken if you don't want to end up with your PC under the control of someone in the former Soviet bloc or the software manufacturer disabling the software remotely. Back in the day the major CAD software companies kind of left open back doors to their software because they realized that in order to sell software that cost thousands they needed a user base, young people that could go into school already being familiar with a product ... who would encourage universities to buy hundreds of licenses for said software ... kind of like cigarette companies getting kids started early

But .... the good news is, there is very good stuff out there for free now, both the open source CAD/CAM software and things like Autodesk Fusion 360, probably the most widely used CAD/CAM software in the world .... and it's free. Autodesk gives it away, it does have some limitations but is perfectly usable for the hobbyist. And .. SolidWorks, the dominate CAD/CAM/Engineering software in the world ... you can get a student version (fully capable) for a subscription of $100 a year ... which I will be getting before long. Right now I use .. we'll call it "bootleg" versions of AutoCad, Autodesk Inventor, 3DS Max etc. for design work ... then import into legit Fusion 360 for CAM work. Which is why I want to change to the student version of SolidWorks ... can do all the design and CAM work in the same program.

[Shiny]

Well, I'm definitely aware of the commercial world where cost is no object but hard to justify a lot of that stuff for hobbyists. CAD/CAM software is one of those. I'll look into the lower-cost options.. thanks for the info.

As to scanners, the cost of accurate laser-based and/or stylus-based digitizers is a barrier. That's why I was wondering if you had played with optical image-based scanners? They obviously would have lots of distortion but if they generated a model you could "adjust" in CAD, I'd think it would be affordable.

Being a met E I like metal parts and my professional life was made difficult many many (and many more) times by unexpected polymer properties.... but that's not the point. ABS is clearly a workhorse. Nylon absorbs water and changes properties so was never popular in my previous life when other "lubricious" options existed. Can you, for example, buy PTFE-filled ABS filaments for your printer (bushings, sliding surfaces)? I'll bet the carbon-fiber ABS options are pretty strong.

Cool beans on the conductive options. That knob needs to be plated. Just saying.

I've seen some fairly high-res "home-printer" parts that had good surface finishes. I'm sure the more you spend, the better the outcome. No surprise here but it's really great the world allows hobbyists to even do this stuff.

Keep having fun and sharing it!

[dataway]

The carbon fiber nylon is pretty good, helps a lot stabilizing it dimensionally and prevents creep under constant load. Nylon has to be handled carefully when printing it because it absorbs so much moisture, which then flashes to steam in the print head which creates a nylon foam So they have inexpensive "dryers" to store the spools and you can feed the printer right from the dryer.

Don't think teflon is possible ... in fact some printer "hot ends" actually have teflon in them, just way too high a melting point. Although .... as a particle composite like you suggest ... yeah, I'd say that's possible .... as it's done with tiny carbon fiber strands.

Basic list of what you can print with a sub $1000 machine.

ABS (and various filled variants)
PLA (quick easy, not suitable for most engineering requirements)
ASA (similar to ABS but with UV resistance)
PET
PETG (what I use most, prints easy, more impact and temp resistant than PLA)
PC (very clear, good to 150C )
PEEK, PEKK, ULTEM (require a high end printer)
PP
Nylon (numerous composite formulas)
TPU (which I use a lot, basically like rubber but better)

Tons of formulas out there for specific purposes and more coming out monthly. Now that 3D printing is being done commercially in the transportation, medical industry and such, the big chemical companies are getting involved and bringing their resources, I expect great advancements to be made in the available filaments.

When you think about it ... way easier to print something than create the molds and do injection molding. I'm absolutely amazed by what I can now do sitting right at my desk with the printer behind me. Don't even like to think about the thousands of hours I've spent designing some stupid bracket or spacer or something and then machining it. Now I can just draw it up (which I'd have to do anyway) and print it to within about .003-.005" accuracy.

[Shiny]

Thank you

If you can print PEEK, you have a great option for underhood, electrical connector housings, wetted parts, maybe even solvent-wetted. Nice! But must be expensive!

PTFE would be amazing if you could print it but like you say, the low surface energy might not play well with the printer. I'll bet you can find filaments of PTFE-filled polymers with enough PTFE content to give you some good slippery stuff that won't suck water from the sky.

Interesting stuff, thanks again for sharing

[darbikrash]

For 2D geometries like the gasket in the first post, if you have a $200 copy/fax machine/printer you already have a scanner. Just lay the part on the copy glass and scan it to a thumb drive. Once on your PC, without too much effort you can convert the scanned image to a format (.jpg, .obj, etc.) that will load right into almost any CAD software in the 2D "sketcher" utility. If you know one dimension of the scanned part, you can scale the image inside the CAD to correct any errors and you are ready to 3D print any 2D object.

[dataway]

Yep, pretty much the way I've always done it. I typically scan at 600 dpi and then import the image into AutoCad and trace the lines in a drawing. Have no idea how AutoCad knows to do it, but the image is exactly 1:1 at 600 dpi when it's in the drawing window. So basically the accuracy is 1/600th of an inch. Use the various drawing tools to trace the image, extrude it to the thickness you want ... save as an STL file, slice it with the printer software and print.

[dataway]

Printed a 68 GTO rear hood bumper with an encapsulated washer. The repops don't have the metal washer in them so are basically useless, you but a screw in them and snug it down, it just pushes the rubber out of the way and you can pull it off the car quite easily by hand. The OEM pieces had the metal washer in them.

Here is the print paused so I can put the washer in place:




Dropped the washer in place:




Here the print is finished and you can see the washer encapsulated in the bottom of the bumper.




Here is the printed part next to a repop .. the repop had to be trimmed down on the top because it was taller than the original OEM piece.




Here is a cross section view of the CAD file show the cavity I put the washer in.




Made some errors in the G-code for the print pause, next example should print out better. Also this has the wrong size washer in it, the washer hole should be smaller than the hole through the bumper ... didn't have the right washer in stock.

[dataway]

A persnickety person could print out bumpers just the right height to hold the back of the hood right where they want it

[Shiny]

Sorry if I'm driving you crazy but I'm obviously intrigued by your projects.

That's a clever way to get an "overmolded" part. I'm kind of surprised the material will fuse to itself after pausing. How does this work? Does the new layer carry enough heat to "melt" the previous layer? Does the whole part need to be heated during printing?

Can you actually get some adhesion to the washer? Not needed in this application but I'm curious if you can get enough of a bond to stress the metal/elastomer interface without it coming apart?

How do the TPE material properties seem to compare with original rubber(?) parts? Can you get a range of durometers? Any info available to you about durability (ie cyclic stress, UV, chemical stability, etc.)?

OK, OK.... yes I should go look it up!!

[dataway]

Strangely enough, yes the adhesion after a pause is good, bear in mind the heated printing bed keeps the part around 120 F (or more if you want), the newly extruded plastic seems to melt right into the previous layers.

I'd guess you'd never get the bond to the metal that you would find with rubber parts, but it does bond to some degree .. depending on the filament used, as some of the filaments bond very well even to glass. I will put the parts I want to insert onto the print bed to warm up to the same temp as the print before I insert them.

As a print pause gets farther from the heated bed (Z axis) it can cause problems, but you can also specify the amount of cooling used on each layer to help compensate (on the print head there is a small cooling fan to help solidify the plastic after it extrudes to firm up the layer)

TPU printed solid is harder than typical rubber, Shore 95A, the resilience is controlled by the amount of "infill" used. In a cube for instance you can print anywhere from 100% (solid) to 1 % infill. The infill is a pattern of internal walls that support the exterior of the cube, many different pattern types are available. Also you can control the top, bottom and wall thickness to create a stiffer or more durable exterior surface. For instance on the hood bumper I had it print multiple top layers for abrasion resistance, multiple bottom layers for mounting point strength, and less layers in the walls to increase it's compressibility.

Attached is a photo of my selection of test cubes printed with varying levels of infill so I can get a tactile reminder of how infill effects flexibility before I design a part.

Info on UV and chemical stability is fairly easy to find, when it comes to things like abrasion resistance and tested durability you'll typically only find ratings like good, fair, bad etc. Much like the guidelines you might find on raw materials on McMaster-Carr ... however I'm sure the information is available from the companies producing the raw stock used to make the filaments ... but a lot of the filaments are custom formulations created by whomever is doing the filament extruding to enhance various properties ... usually with the goal of easier printing. You can for just a few thousand buy your own filament extruder and use raw pellet stock to produce your own filament

[242177P]

For bonding to metal, could you fake it with perforation? Print tiny little dowel pins, and laser cutter the washer. And with a laser cutter, you can make your own washers - think of how much $$ you can save.