Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

RE: measuring volts




I accept what John G. writes about CBL because I had no prior knowledge
here. But what I read about MBL (ULI) conflict with two things:
1) About 3 years ago when I started using ULI I was told that the impute R
is at least 5 giga-ohms ("because it is the Field effect transistor").
2) I know that discharging a low leakage C through ULI, while
measuring the
d.o.p., is much much slower that through most voltmeters. My
observations
were that R is always of the orders of giga-ohms. Too bad that CBL do
work as very high input impedance voltmeters.


I asked an engineer at Vernier about the ULI input characteristics, and he
confirmed that the input is directly into the A/D, and so is very, very high
impedance. However, there is also some leakage current associated with some
overvoltage protection, on the order of a uA. That current will mess up an
experiment where a capacitor is placed directly across the input. I suspect
that the inputs of most devices are not a simple constant high impedance due
to various real-world constraints and protection schemes. As such, I
wouldn't expect a simple exponential decay from a capacitor discharged
across the input.

Again, the input of the ULI will float with nothing connected. It is noisy,
as you can easily confirm by reading the input for a while. The Serial Box
Interface, another Vernier product, is much better in this regard. It has a
10M resistor across the input. The SBI, however, is limited to 50Hz data
collection rates.

The CBL voltage input is definitely not high impedance. It's best to design
an experiment with a moderate 10K or so resistance across the input in order
to hide the influence of the voltage standard. The CBL was optimized to work
with sensors.

J