Engine was running at all times.
On the first file the current clamp was between ballast and bulb, on this system the ignitor is part of the ballast, then the two HT wires come out to the bulb. The current clamp was on one of those wires - so the current clamp would've seen the start-up phase and then the actual current required to sustain the lamp - which has a duty cycle of 50% at a frequency of approx 400Hz.
Voltage dipping down is due to voltage drop, to my uneducated mind it appears that initially as though the ballast gathers itself, accounting for the sloping nature of the 'switch on' then after 15ms the ballast must be charging a large capacitor to account for the voltage drop, which is then charged for 7ms, before being used to ignite the lamp.
After which the lamp is warming up, then it switches over to half power as the pulsing comes in...
I would still be careful about measuring directly with probes, I have one of the later scopes myself, but if for any reason the arc should be lost during the measurement, your scope could be exposed to 92Kv of juice if the ignitor attempted to restart the arc... That could be an expensive test.
I would expect the inductive clamp is picking up alternator ripple, I'm back home now, but will do a ripple test in near future to be certain.
Incidentally, the first time I set the trigger up, it triggered from the low beam hids turning on, rather than the high beam as it was set-up to, fair to say there is a fair bit of noise generated around the systems.