Capacitor behavior in the circuit - Bypass Capacitor

I was trying to measure the supply current of a CAN transceiver during fault conditions, one thing which I noticed was that during CANH to GND(when the current consumption is supposed to be the highest), the current was kind of like a sine wave and going negative during '1' bit transmission(recessive state). Below is the circuit,
[​IMG]

R410 was unmounted and current was measured at Vcc. When I unmounted the bypass capacitor C405 (470nF), I noticed that the current waveform as expected, as in was not going negative during '1' bit trasmission.

Waveform with capacitor :The yellow and pink signals are CANH and CANL signals, the green is the TXD signal from the function generator, the blue signal is the current signal and the last signal is the 'CANH-CANL' difference signal. [​IMG]


Waveform without capacitor:The green is the current signal, blue is the CANH signal, pink and blue are RXD and TXD signals respectively. [​IMG]


I understand that the capacitor was trying to filter and hence the waveform was not a square waveform, but could someone please elaborate , I'm trying to understand why the current is going negative ? (the CAN transceiver is being powered by a 5V regulator, CAN supply current going negative means that the current is being sinked to the 5V regulator ?Is that possible? or is the current flowing through the output capacitor of the 5V regulator be a possibility?)

Apparently a transient oscillation between two bypass capacitors, either due to circuit inductance or low phase margin of the voltage regulator. Without seeing the complete schematic it's just guessing, but not so unusual.

It is an oscillation that involves the inductance in the wire loop you have through the current probe.
I think you should have some value for R410 mounted and calculate the current by measuring the voltage drop over it.
You can probably reduce the value of C405 a lot to get a better value for current spikes, but it is probably not a good idea to remove it completely.

Hi, thanks for your reply. The voltage regulator used in my application circuit is TLS850b0tbv50 , and the output capacitor is 10uF.

Oscillation frequency of about 500 kHz refers to 200 nH circuit inductance, more than I would expect for a LeCroy current probe, but may be the loop is wider than necessary. At least the right order of magnitude.

Could you please explain how you calculated the inductance ? Also the loop is the same with the capacitor removed.