If you read the Tek and Agilent ap noes on probing, for diff measurements they all
discuss need for diff probe rather than using A - B setup on older scopes.
Seems a shame they peddled this capability for years even though functionally
it was a broken technique, except at very low CMR levels.
http://www.tek.com/dl/Probing%2520Techniques%2520for%2520Accurate%2520Voltage%2520Measurements%252051W_60161_1_HR_Letter.pdfhttps://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&cad=rja&uact=8&ved=0ahUKEwiM38bW-OLYAhWuk-AKHZ7XA4oQFghEMAQ&url=http%3A%2F%2Fweb.mit.edu%2F6.101%2Fwww%2Freference%2FABCprobes_s.pdf&usg=AOvVaw3jqTqjxTe5xhU8dQaIF70UNewer scopes, DSOs, have a signal path compensation utility, but the scopes I have with
it do not spec or reveal how the algorithm actually works. I have tested, at a crude level,
that capability, all << 10 Mhz, seems OK (this on 500 Mhz and 1 Ghz DSOs). But at 200
Mhz, 500....how good is it ? Don't know.
Diff probe is where its at. Search eevblog, some DIY approaches to save $$$$.
Last thought, if waveform periodic, take a record, store it, then another, same probe,
and do the subtraction.....maybe that's one way of beating the problem......? I have
never tried that. Clearly sources must not drift, change, loading over the time period
one is gathering the data. Note layout of probe cable could impact this approach, from
one measurement point to another. so just one more reason leading to safe approach,
diff probe.
Regards, Dana.