Randy, 8pA is quite good a leakage / bias current for such a differential VM, and also for the LT1050. What makes me wonder, why three different OpAmps should all have exactly -8pA. The variation from one OpAmp to the others should vary greatly, even if these are from the same batch. I don't remember your circuit, but maybe you have some sort of input protection on the non-inverting input, and that may cause this constant bias. Have you assembled the OpAmp completely on the PCB, or is the + input air-wired to the polarity switch S4E ? Ordinary PCB material may also cause leakage currents. I would also check the switches, if they are contaminated. Frank

Frank, The 8 pA bias current is well within the spec for the LTC1050 (+-10pA typical, worst case +-50pA at 25C) but I was surprised about the lack of variation. I also tried an "air bridge" wiring from the 887 filter to the LTC1050 to avoid any leakage of the circuit board. It made no difference, essentially eliminating board leakage. It could still be switch leakage. Eventually I plan on just [putting a 1 megohm resistor from the LTC1050 positive input to ground and see if I still see the 8pA leakage. If I do, then it's definitely the IC and not the switches. Randy On Sat, Apr 29, 2017 at 3:59 AM, Frank Stellmach <frank.stellmach@freenet.de > wrote: > Randy, > > 8pA is quite good a leakage / bias current for such a differential VM, and > also for the LT1050. > > What makes me wonder, why three different OpAmps should all have exactly > -8pA. > > The variation from one OpAmp to the others should vary greatly, even if > these are from the same batch. > > I don't remember your circuit, but maybe you have some sort of input > protection on the non-inverting input, and that may cause this constant > bias. Have you assembled the OpAmp completely on the PCB, or is the + input > air-wired to the polarity switch S4E ? Ordinary PCB material may also cause > leakage currents. > > I would also check the switches, if they are contaminated. > > Frank > > _______________________________________________ > volt-nuts mailing list -- volt-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/m > ailman/listinfo/volt-nuts > and follow the instructions there. >