What about connectors you say.

Well for SMD ones I do not see an issue.

For THP connectors and or components I know from experience that you can make the drill holes tight enough that you don't need to solder.

Push the component lead into the tight hole in the PCB and the contact is good to go.

On Sparkfun and Adafruit you can find electronics that is open source.

Typically these sites provide the EAGLE schematic and PCB files for a project.

I grabbed a few of those and applied the cavity process to them.

What I found is that the original design did not have 3D models assigned so going 3D to Fusion 360 produced no 3D PCB.

It means I'd have to make realistic 3D models for those and also their cavity variants.

Over time one would build a library for those and the process thereafter is fairly simple.

I simply ran out of time to do this to showcase the process before the competition cut off date.

But that is something that I am interested in exploring because I think hobbyists may be keen adopters of the approach.

Evidently as with every new idea comes unexpected twits and challenges.

Cavity plate material

The first time I tried this process I picket the wrong material for the cavity plate.

So far I have presented the idea in the next phase of this project, the implementation phase I anticipate spending a lot of time finding the right material to use to make cavity molds.

3D printing

If I go with 3D printing I'll be limited with the materials usable with this technology. 

The  good news is that I know someone who has access to a 3D printer at work and he is keen to try this technique out too.

In the country I live the government encourages 3D printing and provides facilities free of charge. I'll be checking this service for sure.

I don't have direct experience with 3D printing yet and I am skeptical that I will be able to create rigid boundaries between cavities.

I have seen some 3D printed objects and was not impressed with the overall quality.

Electronic components dimensions have tolerances. Some parts may fit snugly in cavities others of the same package type might be too loose. In designing the cavity the right tolerance and material stiffness has to be found so that I don't end up with parts rattling inside.

Someone brought up the issue of heat and deformation or even melting of 3D printing materials. This remains to be seen. 

CNC milling

I had a Sherline 2000 + many mods before so I know what it is capable of. I'll fall back to CNC milling if 3D printing does not work out.

Contact pressure

There is a possibility that the contact pressure of the cavity plate is not sufficient to make all contacts reliable.

The mitigation plan for this is glue and or fasteners. Small screws that will squeeze it against the PCB.

I am not so keen on glue because of defeats the purpose of having easy access to the parts but it has value. See next.

Corrosion

The PCB traces may suffer from oxidation from being exposed to the elements especially if I omit the solder mask process in the PCB manufacture. That's why I am thinking it may be a good idea to glue down the cavity plate for it will seal the components in place.

Another idea is to put the whole thing in a recyclable vacuum bag and seal the whole thing. Like shrink wrapping.

Adoption

The next hurdle is getting folks to adopt this packaging method. Does the benefits outweigh the added process and cost? That remains to be seen. Imagine if all the circuits you can buy from hobbyist online stores like Sparkfun or Adafruit provided a cavity plate alternative purchasing option would you go for it? Anyhow here I am speaking about the electronics hobbyist market, there may be a reuse interest there however to maximize the impact on the planet it would be best that the technique is adopted in consumer electronics.

Imagine your run of the mill TV remote control unit being made this way. When it fails maybe you could sell it back to a shop that can open it up and resell the components. 

The PCB industry has advanced and streamlines its process and here I come along with a disruption.

What no solder mask?

What no reflow?

And now you also want our pick and place machines to pick up parts upside down and drop them inside little cavities in a plate.

I can see a lot of resistance there. Can existing tooling do this? Can glue dot placing machines drop glue dots inside small cavities? 

Scaling

Yet another hurdle is the scaling aspect. It is not far fetched to think that one a cavity plate is designed for a product it can be stepped and repeated to make a mold that will mass produce them. Its like making a mold for a mold.

These are a few of the issues that I'll be faced with next.   

If you are impatient: tldr here is the end result:

Cavity plate

The parts of an electronic product that serve to keep the components in place and force their terminals to make contact against a printed circuit board.

Cavity plates can be 3D printed or CNC milled. 

A mold of a cavity plate stepped and repeated can be made to produce them in large quantities.

They should be made of recyclable materials.

Cavities

The holes in the cavity plate that adapt to the shape of components to keep them in place. 

They may have glue dots placed at their bottom before inserting the components upside down so they don't fall all over the place. 

They can be 3D printed as part of the plate or CNC milled.

Cavity plate assembly

Components are placed in the cavities in a cavity plate. This assembly is then mounted and pressed against a printed circuit board. The cavities are designed to adapt to the variations of the component dimension variations and just deep enough to create a force that presses the component pins against the printed circuit board pads. 

Cavity retainers

These are conical shaped pins that protrude from a cavity plate. 

The pins fit in holes in the printed circuit board tightly as they are pushed in.

The pins can be augmented with the use of small screws to provide the contact force when assembled.  

What if I were to say, the hell with soldering at all. Lets not solder!

Make the product packaging itself responsible for holding the parts in place and making them make contact.

When the product is broken chances are it's just a few parts gone bad. 

In that situation just open it up and get to the individual parts easily. 

Test them until the faulty ones are found. 

Replace them and get the product working again.

If the product has gone obsolete. 

Open it up. Shake it and all the parts fall in a tray. 

Reuse the parts.

Oh boy now I really opened a can of worms now!

The folks in the EU will like to hear this with their ongoing “planned obsolescence” battles.

The best they managed to do so far is to ask manufacturers to label their products with a repairability rating.

Let the customer decide which to buy.

In my lunatic approach to electronics packaging everything is recoverable and everything can be replaced.

So does that mean it gets a 10/10 rating?

Come on what is your revolutionary idea luni Alain? Spill it out already!

Here it goes in brief:

The key idea

What if electronics could be built in a way where the housing or sub-housings inside holds the components in place forcing their leads to make contact with a traditional printed circuit board. The housing can consist of one or more modular subunits. Open these subunits and loosen the parts for reuse or repair. This opens up new markets for the easier repair and resale of used electronic parts. 

We don't want just the metal powder left after the fumes have dispersed.

We want to reuse the bits and pieces! The more the merrier.

From now on I will elaborate on the general idea.