Best practice for probing capacitance sensitive test point in a switching supply?

Question

I need to probe the hot side of a switching supply between the drive transistor and inductor (Marked TP on diagram). I tie my scope ground to the device ground with short fat wire and when probing with a high quality probe I can hear a perceptible change in the inductor whine presumably due to the stray impedance of the probe being close in value to a 100pF capacitor that is part of the circuit. I am trying to measure the charge/discharge time of the charge pump and estimate the steady state of the inductor on a live circuit and I am afraid I am skewing the result.

A simplified diagram:

^{simulate this circuit – Schematic created using CircuitLab}

Answer 1

There are low capacitance and active probes that you could find (if you are sitting in a work environment with a big closet full of cables, ask someone, there might just be one).

An active probe can go as low as 1 or 2 pF. Though good ones with that low an input capacitance aren't going to be cheap.

You can also hack together your own active probe with the appropriate JFET or low input capacitance (dual-gate?) MOST.

For a nice train of thoughts and comments on getting or building active probes, one could start at the EEVBlog Forum topic exactly about that. (Credit goes to Google for that one)

If you already know the major specifics of your signal and just want to know relative changes, frequencies and/or duty cycles a DIY active probe could be simple enough for the purpose. The biggest hassle in making an active probe is getting it accurate and reliable in the input-to-output voltage domain so that you can use it for small and large signals without too many problems.

And that hassle is big enough for people paid to use them to always buy them from a known brand. Measuring and verifying, before you know it you're 5 days in to get the last offsets and gain adjustments right and you realise you just wasted as much money on your time as the appropriate well built and easy to handle probe would have cost. But, if you can skip that by just not caring too much about 10V or 9V, you'll be done building something quick enough.

Answer 2

Any time you probe a circuit, you "affect" its characteristics. The idea behind measuring any thing, is to keep to a minimum the effect the probing instrument has on the circuit. In your particular case, if you are probing with a 100pf probe, you effectively change the capacitance of the circuit by 100%! If you use a 10pf probe, you would only change it by 10%. If you need greater accuracy than this, you need a "better" probe (1pf?).