![]() |
|
|
|
#3
|
||||
|
||||
Hey Shern,
I start by reconstructing the broken part with super glue. The intent is to restore as much of the original geometry as possible. Next I use a pair of calipers and measure off what I believe are the critical dimensions. Some dimensions can be just picked out arbitrarily. Others not. On this clip the holes where the clip slips over the switch were critical. Also the thickness of the same hoops had to be spot on because they slip into a slot on the switch. The little ears on the sides are actually what locks the switch retainer in place so their relative position to the holes is critical. Also the location of the slots in the middle need to match some tabs on the switch. That was about it. I just measure and start drawing in Tinkercad…a free online SW made for small kids. alternatively I use Freecad. They have a photo import which is kind of a poor man’s scanner. You take a picture and import it. Then you measure some feature in the photo with your calipers and tell the software how big it is in real life. The CAD then scales the entire photo. Now you just trace over the image to get your locations. The contours are relatively simple. These parts were designed in the 1980s with old CAD tools. Most curves are just circular primitives because people drew curves with compasses and circles back then. I simply guesstimate a radius if I cannot measure it by holding a coin or other round object against the surface and just measure the coin with the caliper. Then I iterate with the printer to get closer to reality. Alternatively you can print your own radius templates with the 3D printer. On the switch retainer you can see I just cheated - it had an organic shaped thumb tab but I just made a simple round thumb tab to save time. Once you get 3D models you print them. There will be issues with shrinkage, fits and such but you basically edit and print interactively. I printed this switch cover four times to get it just right. My prints for this design were $0.11 each and they just take a few minutes to edit and send to the printer. You can refine a fit really well in a few tries. That’s about it. For the most part it’s using a bunch of kids’ toys. Tinkercad, Cura Slicer, a harbor freight digital caliper, and a $200 printer. Money is no longer the excuse not to do this. It mostly costs bench space. I’m afraid we are going to be doing more and more of this. The good news is the materials are getting better. JLC PCB is offering metal printing now. Just send in your 3D models (after testing them in plastic for fit) and you get back metal parts. I just spent the last day working on HVAC diaphragms on my new w202 c280. Behr has simply discontinued manufacture of the flap actuators for 5/6 of the vacuum pods. Gone are the days when you could just get what you needed from pelican. It is getting bad very fast. I was thinking my 1997 C280 would have better availability than a w123. Nope. I think the market for diy on that car is so small that manufacturers simply won’t support it. I cannot even get an antenna grommet. I’ll be modelling one for printing in TPU (rubber) soon. 3D scanning cameras have recently dropped in price. Creality has just released one. I think these will democratize the reverse engineering business.
__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD) 82 240D stick shift 335k miles (SOLD) 82 300SD 300k miles 85 300D Turbodiesel 170k miles 97 C280 147k miles |
Bookmarks |
|
|