fll-freak wrote:It would appear that the power supplies I use at work are fully isolated from earth ground. This is why I have been able to pick an arbitrary location for the probe's ground clip. These are expensive Agilent type power supplies. At home I am using computer power supplies modified to run as lab supplies. The grounds of these are most definitely tied to earth.
And with this a small light bulb has formed over my head. This explains numerous small strange things that have happened to me over the years.
And the light buld has formed a small dark cloud. How am I going to measure a small signal with a large DC offset if I can't reference an arbitrary ground? I would normaly just AC couple, but my signal is at 2 Hz, well bellow the AC filter cutoff. Sounds like finding a scope I can run of batteries might be a good strategy, or running my circuit of batteries during test. Or, is there such thing as an isolation 'thing' I can use to isolate by power supply or scope from each other?
Almost all scopes have their ground leads earthed, though there are one or two that don't. The Tek TPS2000 series, for example, has isolated channels and is intended for people working on power supplies.
Do bear in mind that even if you have dc isolation between the scope's earth and the device you're measuring across, there may still be significant capacitance which will affect your measurement. For example, I always have the scope itself earthed but sometimes use a lab supply which is floating so I can earth an arbitrary point without dc current flowing. But that's not so say there isn't capacitance between the PSU output and earth even if there's no dc connection, so the circuit under test still sees the scope ground connection as a capacitor to earth. At high frequencies there's no substitute for a true differential measurement using two channels of the scope, or better still, a true differential probe.
To measure your 2Hz signal, just leave the scope in dc coupled mode and fit a capacitor between your signal point and the probe tip. If you have a x10 probe then you have a time constant of 10MOhms*C, so a 10uF cap should allow you to measure it without significant degradation (100 sec time constant), and 1uF might even be more suitable. Bear in mind that you'll need to connect the signal to the scope for some time before the cap charges to the correct dc level and the trace appears on screen.