After the first prototype, I redesign some parts and I design and included new ones: I changed the design of the upper part to include a second rail to hold de paper; in the base, I included the footprint of the components (servos, SHARP sensors, relay module and bateries) so all the components are organized; I did a "rail" in the vertical structure and the cross to hold the paper; also I designed a three more of 3d printed pieces to hold the sharp sensor, bulb cap, and phototransistor PCB; and finally, I did the laser cutting pieces a bit "bigger", in such a way that all the structure is stable by the press-fit without needing to use glue or tape. So I fabricated the new pieces and I started assembling the lamp.

But everything was a complete disaster: I don't know how or why but I messed-up everything, maybe I was too tired when I prepared the files or I set something wrong while laser cutting... I don't know, but I even couldn't assemble the pieces together, the pieces didn't fit, sometimes they had weird geometries, the press-fit was too tight... so it was extremely hard to fit them... The rail wasn't high enough to hold the paper properly... I don't know what happened, what I couldn't work with these pieces at all.

So I did all the files again... And I cut them again... And this time everything fit quite nicely. I did the rail higher (6mm.) and thicker (7mm.) and it seemed to work very good, it is able to hold the paper strongly but with enough space to move, so there is no way that the paper moves out of the circular shape. And the holes that I did to hold the paper in the vertical structure and the cross worked very well too.

So with the lamp finally assembled, I prepared everything to run the first test and finally to see how the servo, helped by the cross, folds and extends the layer of paper. I used the same code that I used for the first running in the first prototype and the output devices week. And... It was a huge fail.
On the one hand, all the MG90s servos that I bought they don't work, none of them, very shameful and disappointing, even the one how was working last time, it didn't work... Luckily I had one SG90s around. And, on the other hand, the motor wasn't powerful enough to fold completely the paper... And it wasn't able either to come back to position 0... Here you can see what happened:

So at that moment I literally got frozen... It wasn't working at all... I worked so much in this, you know, I fixed the errors, improving the design... And it wasn't working... After 20 minutes looking at the lamp with my mind blanked, I decided to move forward and think in other design in where I would use the same components but with other mechanical movements that reach the same effect that I want but without needing to fold the paper.
I started drawing on my notebook, figuring out a new design... I ideaed a system in where the servo moves a gear which moves other gear which does that the center of rotation changed to the center of the lamp. Then this second gear moves 120º a whole section of the lamp (well, just the paper part), which will be hidden behind other section, so that means that I need to have the gears and motors at different levels if I want to have more than one servo, more than one section of the lamp working at the same time... And with that in mind I went to sleep...

However, when I woke up I decided to give the lamp a second opportunity. I wanted to try with another type of paper, a thinner paper, as the one that I was using was a bit thick, like watercolor paper, because my idea was to engrave something with the laser cutter in the paper so the lamp could have that customizable part (but you cannot do that if the paper is very thin, it will break very easily). But anyway, I took a piece of normal paper (bond 90gr.), I made and cut the folding geometry and I replaced it in the lamp. And it worked!!!!

So I know now that the mechanism can work, but I have to be careful with the kind of material that I use. My instructor suggested me to use linen and 3d printing. The idea is to print a couple of layers of the "structure" of the folding layer (sticks, for instance) then pause the machine, place the linen, and then continue printing, and then both parts of plastic join and you have the linen in the middle. I really want to try this technique, above all because, I think that the paper is very boring and this will give to the lamp and a cool extra point.

During this step of the process, I also tested the relay that needed to control the LED bulb. You can find more information about this in output devices week.
I also did the networking between the phototransistor and the relay. You can find more information about this in networking and communications week.

Finally, durign the wildcard week I made the packaging for the final project: