...
In the above picture, L refers long cable. One key point is that the connection to ground should be at one point, and this point should be at the electrometer input. However, there is the option of connecting the Ammeter Lo to V-source LO. We probably don't want to do that (because we need a return cable from the solar cell anyway) but we should think about it (and ask Jim). Jim agrees that we should not use the internal connection and we should have the external ground connection in part because it will be easier to test things.
Generally, we want to be able to cancel about 2-5 nA of leakage current, which with a few kOhm solar cell resistance is ~ uV of applied voltage. If we use 300k/100 ohm resistors for the voltage divider, 1 uV corresponds to ~ 3 mV on the voltage source. That's lower than we'd like. If we do 500k/50, then we're at 10 mV at V-source, which still isn't really good enough resolution. But if we did 500k/10 for the divider, I think we'd be fine.
...