I would say there seems to be a problem somewhere because even your open circuit voltage is too low.

The Voc is 37.85 volts, and assuming its warm (45°C) on your roof, the temperature coefficient of -0.35%/°C means your panel Voc should still be about 35.2 volts. So at that temperature rise, a string of 10 should have an open circuit voltage of 352 volts. You are getting 320 volts. The difference of 32 volts is very close to what I would expect if one of your panels was shorted.

For each of your panels to be at 32 volts due to high temperature, thus creating a string voltage of 320 volts, the roof top panels would have to be at 69°C (156°F), which is hot even for a roof top installation.

My first suggestion would be to test each of your panels to see if they each create about 35 volts when they are disconnected. It might save some time to isolate each string of panels and measure each string voltage separately to see one string is generating a lower voltage than the other, and if you find that result, then check the panels in the low string.

The large voltage drop when you are trying to draw current from the system could also be caused by a bad panel or a shaded panel. The maximum power point voltage is 30.5 volts, so you should be seeing about 300 volts since you are using a MPPT charge controller. If none of these problems is the cause, then I would check for a loose connection at your wire termination points.

The 8 awg wire is more than large enough to avoid excessive voltage drop given that the total current from your rooftop is less than 18 amps. Even with a 100 foot run of 8awg wire, at 18 amps the voltage drop would only be 2.26 volts, so you would see 298 volts. After correcting for the temperature related power loss of -0.41%/°C, with the roof at 45°C (113°F) and the maximum power point current of 8.27 amps, you should still see around 275 volts at your charge controller.

These calculation are based on the

datasheet I found for your panels here.