Testing LED lights for flicker or strobe effect?

[ericmark]

I fitted some electronic switches which caused problems, one lights flashing when off, cured using a load capacitor, and two flicker, the latter was first cured by fitting one tungsten bulb in a set of 5, and then my wife ordered a new set of 5 from internet so old bulb looked like this

G9-small.jpg (21.24 KiB) Viewed 1132 times

and the new one more like this

G9.jpg (20.08 KiB) Viewed 1132 times

however within a day one lamp stopped working, and putting other type in colour was wrong, did not really do anything for months, but out of interest thought I would see what was inside the failed bulb.

Could only open part of it, and sorry no photos, but there were two capacitors, one ceramic and one electrolytic, and the ceramic one was disconnected, looked as if some one missed with soldering iron, so soldered on again and tested, and it worked.

However the electrolytic capacitor was larger than could have possibly been fitting inside a package like this

G9-small.jpg (21.24 KiB) Viewed 1132 times

so that seems to answer the question why the smaller package flickered there is no smoothing capacitor.

So moving on, my daughter says our lights give her a head ache, and it seems reasonable to assume they may be flickering even if I can't see the flicker, but other than opening up a bulb, how can one work out if bulbs flicker? Camera, solar cell and meter set to frequency, although the light I repaired was a G9, the lights in living room are E14, and we use loads of E14 so maybe just a case of swapping them around. But if I can't detect flicker, I can't eliminate it. Can't buy pearl E14 tungsten bulbs any more, so no option have to be LED, I have stock of BA22d tungsten, but not E14 (SES).

So how can I test the light from a bulb? Not noticed banding on DSLR camera, but may be rooms used in lights not pulsing.

[kellys_eye]

Yes, cameras could discern the flicker. But I believe if you fit dimmable LEDs they have the power supply filtered to prevent flicker (when run at 'full volume' - i.e. don't attempt to dim them)

[Dave54]

Not sure how scientific this is, but what about taking a series of still shots with the DSLR in a room with just the lights in question as a source. Using anything over 1/50th shutter speed should give some dark (or darker) frames I'd have thought, just by the law of averages.

[ericmark]

The DSLR fastest speed shutter starting to open to fully closed is 1/180 of a second. Although it has a shutter speed of 1/1000 second, it basic scans like a document scanner but quicker, so we should see images with banding as light switching on/off during the scan, but not seen this happen, have when taking pictures of moving vehicles had some odd effects where it has shortened or lengthened the image of a vehicle, but not the banding one would expect even with fluorescent lights.

SLR cameras have used focal plane shutters for as long as I can remember, no change when they went digital. Phone camera different, no shutter as such. But I did not pay that much attention when being taught about light on how it all worked.

[Dave54]

I'd simply forgotten, I'm embarrassed to say, although I did know about it, particularly as when used with flash sync.
So if you're not seeing "banding", then likely that there's no flicker?
If you use 180th, being faster than the flicker, then you should get some whole "black frames" in a series of shots if the light is flickering?