Low current measurement with Keithley 6485

Good day,

I am using a keithley 6485 as an electrometer in combination with a DUT and a power supply. I am having a problem making measurements. When I make the connections which are as follows:

positive Power supply terminal to terminal of DUT
Other terminal of DUT to keithley BNC Input
I am using the common input at the back of the 6485 as a return path ( I Know one can also use the shielding for LO return path) to negative terminal of power supply.

The porblem i am having is after i have turned on the equipment and given sufficient warm up time, when i apply a voltage across the DUT the readings on the pico start at some high value and slowly trickle down (to what i am assuming is the correct value of current) however this trickling down can taken serveral minutes if not hours. Does anyone know what could be going wrong?

many thanks 
Cameron

If the ohms of the sample are very large, the RC time constants can be surprisingly long.  Minutes or hours is not out of question.

In these cases, it might be sufficient to define a procedure where you take the measurement after XX time of electrification.  Such a consistent method would allow comparison of sample A to sample B, etc.

Unrelated:  you may want to consider some shielding for the setup.  With connections in open air, results are susceptible to EMI and charged particles from air.  Simply walking past the setup can be reflected on the picoammeter.

Low level measurements handbook