The power supply is extremely important and is quite complicated in a Phyllo since there are so many PCB as well as several voltages needed.
We have 79 LEDs working under 600mA which sums to 48A. As we flash 30 times every second and each flash lasts around 100us, the LEDs are on at most 0,3% of the time. Thus the average power supply needed is 0,16A.
We have decided to rely on capacitors to fully provide the LED power supply. There are two main reasons for this choice. First, the LED are switched off most of the time. Thus, it seems overkill to plan every component as if they had to withstand the whole 50A theoretically possible when on average they will not even have 0,5A. Second, we wish to provide the power through a battery 3S or 4S whose voltage output is far too high for our needs. Thus we absolutely need a switching regulator. However, such a regulator has a very bad response time to pic consumption and we precisely need such a responsive power supply as the flash of the LED are extremely short. When the flash will stop, the regulator will probably not even have started to respond.
Then, according to the following formula : C = 1/Voltage_drop * Imax * Duration, the total capacity needed is less than 14mF which is not that much since it is spread over all petals and can also be stored on main_pcb. Moreover, as the voltage drop will be less than 1V, the intensity required to replenish the capacitors will be small in front of 1A which should not overload the regulator.
The 5V regulator will be placed on the bottom_pcb instead of the main one. Indeed, we have a lot of space there and the required power will not exceed 6A. It was not possible before choosing to rely on the capacitors as we would have needed too much intensity to go through the wires between the fixed and rotating parts.
We also need some 3.3V power supply. We will place a dedicated regulator on each PCB which needs this voltage. This decision is mainly to reduce the number of wires needed. We will use switching regulator on main_pcb and LDO on the others (even the bottom_pcb). The reason for this difference is that we need more than 2A at 3.3V on main_pcb whereas less than 0,5A are enough for the others.
Another reason to place the 5V regulator on the bottom_pcb is to ensure a simple backup solution. In case we have problems with anything, we will directly use sector alimentation to supply the main_pcb with 5V and thus bypass the 5V regulator. This way, we will have as much intensity as we need.
We plan two sets of pads for 5V supply on main_pcb which are next to the motor axis. The supply for the other PCB will come from main_pcb. The ground will stick to the SPI wires. We plan 3 buses for now.
The first motor we tried did not function with a 3S battery. However, we are soon going to test another one with an integrated ESC which is supposed to support 3S batteries. Thus we have not completely decided on the battery yet. However, the battery pack protector we chose is compatible with either 3S or 4S ones. New motor vs Old motor.
The simplest way would be to directly buy an already existing solution. However we did not find anything satisfying yet.
Please do tell us if you have any ideas.
The alternative solution is to design a transmission ourselves.
We have thought of many solutions (see this article for example) and do not have anything completely functional yet either.
In the current transmission system, VCC (5V) is transmitted through the motor axis. This axis has a diameter of 8mm. It is inside a hollow axis carrying the ground. The rotating union is made for both axis with ball bearings.