Version #1, Version #3 measure solar cell's open-circuit voltage. Only a tiny current flows. The resistors that have been added do almost nothing. The solar cells provide next to NO output power.
1.Is there a reason I need to place this resistor in Parallel?
I'm assuming the purpose of the added resistor is to provide a means of pulling power from the solar cell. A series resistor doesn't do that. A parallel resistor does provide a load on the cell where significant current can flow (depending on the resistor value you choose).
Version #2 does pull current from each solar cell, through the added resistor(s). It is possible to calculate power delivered to each resistor if you know its resistance. Since it likely heats up, its resistance should be independent of temperature...a power resistor of high accuracy likely meets this requirement. The ADS1115 measures voltage, but power and/or current can be found by measuring voltage across the resistor: \$ Power={voltage^2 \over resistance}\$ , \$current = {voltage\over resistance}\$
For OP's solar cells having open-circuit voltage of 5.5V, ADS1115 current flowing into the ADS1115 can be assumed so small as to be insignificant. However, there's a problem choosing the DC voltage applied to ADS1115 Vdd pin:
- voltage across measuring resistor MUST BE LESS THAN ADS1115 Vdd!
- serial interface to Raspberry Pi runs at 3.3V (possibly 5V-tolerant).
It is likely safer to supply +3.3V to ADS1115 Vdd for its power supply. But that means your solar cell voltage (possibly as high as +5.5V) is above Vdd. So you'll have to add a resistor voltage divider to the shunt resistor to reduce it below +3.3V.
Let's say you choose to load each cell with a 20 ohm shunt resistor. Doing so will reduce cell voltage from 5.5V to something less, but let's ensure that 5.5V will not overload the ADS1115 input range of 3.3V. Each cell is loaded with 20 ohms, yet the maximum voltage seen be ADS1115 is 3.3V across the 12 ohm resistor, when the cell's voltage is 5.5V:
simulate this circuit – Schematic created using CircuitLab
Both 8 ohm and 12 ohm resistors should be precision types that can dissipate about one watt at least. If single-ended voltage measurement is set up for the ADS1115, four loaded cells can be measured with one chip.
2.Can I place the resistor on the Common Ground?
As mentioned before, Circuit #3 only measures each cell's open-circuit voltage - next to no current flows.
3.Do I need to worry about high Amps on my ground-wire if I had many panels being measured?
Yes, you do, if your intent is to measure individual cells output power. Especially if you consider that ADS1115 resolution is quite good...any ground resistance could cause one cell's condition to influence measurements made on other cells. You want to establish a SINGLE-POINT GND connection, where all those 12-ohm resistors meet (called a STAR gnd).
Also, be aware that your R_PI must also connect to this GND point (required for the I2C serial port connection).
Since each cell's current only flows in a loop through its 20-ohm load (8 + 12 ohms) very very little GND current flows into ADS1115.
If you do this out-of-doors (where the sun shines) then be aware that any ground loops can cause a lightning strike to destroy ADS1115, and R_PI. Wiring lengths should be short, because wire resistance adds to that 20 ohms. All those 12-ohm resistors should meet at the single-point GND with ALMOST ZERO wiring length. If you cannot arrange this, then ADS1115 single-ended measurements shouldn't be attempted. Differential measurements should be done, which reduces the number of cells that each ADS1115 can measure from four (as shown in the schematic) to two.
