CFL current profile.

I have the B version of that scope. What are you using for the current probe ? I may want to look into getting one.

Tektronix P6019 with the attenuator accessory

How much of a techie/geek can you be...with an AOL account? ?

What is the time division on the display? relay reaction time is in the msecs. It is safe to say that loads have inductive reactance value causing the spike like that.

Better than anyone with a Gmail account...Damn Man!

I suspect they are roughly coincident with the peak (slightly leading) of the voltage waveform. So, apparent power is much higher (2A @ ~80+V?)

The problem with relay drives is you can't readily predict how long it will take for the contacts to open. So, you can't realistically try to open it near a zero crossing (or, ANYWHERE that is "away from" the peak of the cycle and the peak current draw/power consumption).

Likewise, unless it's a Hg-wetted relay, you can't tell how long the contacts will *bounce* when making.

And, of course, if you're making consumer kit, you're usually quick to trade off *any* sorts of real "specifications" for a few micropennies!

Do you have access to a DSO? If so, you might try catching the turn-on transient to see what a discharged primary cap looks like to the line!

That was a great 'scope! I lugged one around the country at my first job. Wouldn't enjoy doing that nowadays, though! :<

Presently using a rescued TDS754D -- for the few occasions when I actually need a 'scope.

Should be pretty obvious -- current repeats with a periodicity that tracks (rectified) line frequency. So, 8.3ms between repetitions...

You're not concerned with the actuation/release time (except if you try to synchronize it with a certain point on the waveform (e.g., to deliberately avoid these spikes which are roughly coincident with voltage peak). You're more concerned with how long the contacts bounce as they make/break -- especially when opening the connection (cuz current is flowing and will want to maintain an arc across the widening gap --> contacts melt)

I have had it since AOL started, long before anyone even knew what the internet was. (Prodigy before that) Right now I like it for my public Email address because I run AOL7 and those nasty script based attacks simply pop up a gray error box.

That is just looking at the input capacitor(s) charging. The signal is 60hz and I knocked it out of calibration to get the best photo.

I looked on ebay and they are around $ 200 used. I doubt that I will get one, Just don't think I would use one enough to justify the price. I will keep an eye out for one at fleamarkets.

"field engineering" would say use a big honking relay rated for DC.

I might be able to catch it with the mask and an open shutter. I am just not sure I care that much

I suspect you can turn up generic waveforms someplace on-line and extrapolate for your particular "load".

I have a clamp on accessory for my Fluke meter that works as well but you can't really trust the calibration as much, although the scope has not been calibrated since the Clinton administration. It used to be on a regular schedule when we were ISO 2000/1. I got it as a present when I retired, along with my 453 Tek and a bunch of other test equipment that nobody who stayed, knew how to use. One of my favorites is the Dranitz 626 line monitor. You don't need it much but when you do, there is not much else that will work.