D2 Meeting 3/6
Array layout and integration with topshell design:
[Shared the Flat Topshell Design, the curved Topshell Design, and the new array Layout]
It would be difficult to do a bend on an individual module if the module lays halfway on the curve. The module might want to pucker up / not sit flat. Might work better with a 2 by 4.
Lower angles of the sun would be more penalizing on a larger curve.
It would be good to combine modules that are at a similar angle together, so that they having matching current.
Bypass diodes may not be worth it unless there is really low sun.
Bypass diode every 8 cells (would prevent the entire string from being taken down)
Drop half a volt when current flows through the diode (lose 3 watts through the diode)
Shottky
$1 a piece max
As long as all of the panels are identical then it is pretty okay
would work for both 2 by 3 or 2 by 4s. ( must be even number of cells)
Should be fine if we mix and match between 2 by 3 and 2 by 4s in the same submodule
PV Cyst : Commercial tool for modeling array at different angles relative to the sun and how the array performs at each of those angles
SAM Model by nrel: [another tool similar to above]
Array Attachment / Backing:
700 watts / kg - 200 micron encapsulin cells
Fiberglass back sheet 200 watts / kg
If we wanna create 600 watts, then the array would weight 3 kg
No Backing: 6 square meters, 1.2 kw , 6 kg for the array
300 watts/kg
If we wanna create 600 watts, then the array would weigh 2 kg
Creates same amount of power with less weight
Flexibility of the backing should be fine - 5 mms thick for the backing
Cooling: you probbably get conduction through the back to the body of the module
Thermocouples:
They use really thin thermocouples from Omega
Would need to look at
Dont need a joint at the cold read (goes through a connector and plug into a data logger).
Where are the connectors on the module? Since the connectors are 12 feet they route them through the module whenever they laminate
6 modules approximate that would have the thermocouples
each module will have 2 electical connections (positive and negative) and the thermocouple wire. The wire will stick out and we can route it wherever we need it to go.
Wire that comes out of the sides or solder pad on the front of the pannel (connect the pannels through the car vs route on the surface)
One ribbon to the next moduler under the modules
put pad on the front and then can connect them after we glue them all down
Pseudo-Square:
The irradiance sensors will be put onto a square that is located between modules. Then can wire the connections of the sensor through the topshell.
Only if we decide to use thermisters would we need to route the connections through the pseudo-square
Ordering Panels:
1 week: decide what configuration of panels to print for testing (how many have what characteristics, which have thermisters, diodes, etc)
Summary (Initial Thoughts):
Thermocouples: We can probbably use their thermocouples. They are really thin and the cold joint measure shouldn’t be an issue otherwise they would know about it.
Talk to Tanya about how we would connect the thermocouples / placement of the thermocouple wire in addition to positive negative leads (placement of leads)
Backing: We can try getting some modules with various types of backing to determine how flexible the actual modules are. We will probbably end up going with the thinner backing out of the options just so that we don’t affect cooling too heavily, but also have some additional stability so that the cells are not as likely to break.
Testing Backing: (Primarily mechanical testing to determine flexibility of modules with / without backing)
How large are the holes in the topshell that will be sitting under the modules?
Weighing the modules with backing to calculate the additional weight overall
Cooling testing: Possible conduct a test for this?
Irradiance Square: We will probbably go with having pseduo squared between the modules, which will then be wired through the topshell. We would have likely around 3 per subarray (9 total). Then each sensor will be routed to a blackbodyC (9 boards).
Blackbody C gets both 1 thermocouple and 1 irradience sensor. We need to make sure that when we manufacture the panels, we have 9 modules with thermocouples embedded in them.
Talk to Tanya about the number of modules with thermocouples, since D2 recommended having about 6 modules with thermocouples for the entire array. So, it might be overkill to have that 9 modules with thermocouples.
Could also have less irradience sensors (i.e. 2 per array)
Aeroshell: Likely going to go with the medium curve design
Bypass Diodes: Need to make sure to have bypass diode for each module. They made it sound like it wasn’t as much of a priority, but for our application we need to optimize for real-time shading as much as possible.
Asking for 1 module without bypass diode and one with a bypass diode and try out to see how much power we get in partial shading
Types of Connections: On the front with a Solder Pad vs ribbon leads through under the module (back of module)
Pros for Connecting On the Front with a Solder Pad: Easier to assemble on the array first, then connect all of the modules after they are all placed and secured
Cons for Connecting On the Front with a Solder Pad: exposed leads might be dangerous (need to ensure that leads are covered in some way, i.e. electrical tape)
Back of Module: Would likely need to connect prior to placing the modules on the the topshell. Would also need to connect them very closely to eachother which might be difficult after placing.
Sample Module: Could maybe get one of each type so that we can see which types of connections we prefer
2 weeks: Finalisize quantities of panels, verify how we want the connections between the panels to happen → they can send us a quote after this
Around this time we can also send the updated topshell layout
3-4 weeks: they want to know exactly what they are building; let them know what would be most helpful for us (what type of interconnects are needed);
6 - 7 weeks until we receive the panels
8 - 10 sample modules; attached to the backing
G10 fiberglass is part of laminate