Taking an oscilloscope to my flickering lights

[lukash]

So, lights in my kitchen flicker when my neighbor's AC system comes on. But they do not dim cause of the load, they are LED lights and they flicker with a fairly fast frequency.
I discovered this by accident but since then I have many times waited for the AC system to come on and off to be 100% sure there is an exact correlation and I'm 100% sure it is related.

This is how they flicker (the camera kind of attenuates the affect by adjusting to it but if you look hard enough you can see it):

After messing around looking for a typical voltage drop issues I ended up pulling out my differential probe and hooking one fixture up to an oscilloscope. After a bit of messing around I came up with the following, it happens on each peak, the max and min and on all of them.



Here is the same thing as before as a YouTube video for better quality:

Here is a bit of a zoomed out version: , at 0:06 in the video the AC system turns off and you can clearly see the behavior stops.


I could just go talk to them to have the system looked at but then it would be no fun :-)

Can anybody explain the behavior?

I am honestly just curious. The AC system also sounds like the compressor is having a hard time and you can hear it ramp up/down/up/down/up/down. I would guess the capacitor on the compressor is not good anymore and the compressor is struggling on the single phase without a run capacitor (or a degraded value capacitor?).

I am in the US and this is a 120V / 60Hz power.

[moffy]

Have you taken a measure of your mains voltage without the flickering? Then we could see a comparison of before and after. Nothing obviously wrong with the mains waveform, except for a little jitter on the peak. Are your LED lights dimmable, or is there a dimmer control on them?

[CaptDon]

Any light on a dimmer is very sensitive to voltage fluctuations when set to less than full brightness. Some dimmers are really bad in this regard. Some of the older LED lights which used ONLY a series capacitor and low ohm resistor to control LED current are real turds and EXTREMELY sensitive to high frequency harmonic content on the mains. When placed near a heavy inductive load they would often have a very short life.

[TheMG]

If these are dimmable LED fixtures, I find some of them can be quite sensitive to any change in not only voltage, but also the waveform shape, as they are designed for compatibility with standard TRIAC dimmers. That's one of the pitfalls of such dimmable LED.

[MarkL]

A relative of mine had this exact issue of a neighbor's A/C and heat pump causing her recessed LED lights to flicker.  A contractor had installed the cheapest Home Depot retrofit LED lamps in the fixtures.

Replacing them with a better known brand solved the issue.  Philips claims to have a design that eliminates flicker, and it seems to be true.  These worked perfectly, and were qty 6 for US$50:

  https://www.amazon.com/Philips-myLiving-Dimmable-Downlight-Recessed/dp/B07KLPFWNP

Philips Lighting #801266, if the link stops working.

[Conrad Hoffman]

IMO, that waveform is pretty typical of a loaded line and stuff has to be able to contend with it. BTW, I usually look at my line using a (quality) transformer to isolate it, so I don't have to set up differential probes.

[james_s]

I have some early Philips LED bulbs that flicker slightly when running off my generator. It's a Honda inverter generator and the waveform *looks* even nicer than utility power on a scope, but those bulbs still flicker slightly. Some of them are just fussy.

[lukash]

Thank you for opinions.

I researched a bit into the bulbs and yes, they indeed are dimmable. Specifically they are these bulbs if anybody is interested: https://www.amazon.com/gp/product/B07MLSP12C/

What really puzzles me is that I myself have two air conditioners and living in Nevada makes them pretty big to survive the summer. When they turn on there is a noticeable voltage drop for like 0.5 sec as one would expect I suppose. After that no flickering is visible. So I still think something must be wrong with the neighbors unit or I am not really sure why his would make the waveform jitter but mine would not (otherwise comparable).

Anyhow, I took one bulb apart just to see and I am attaching the PCB photo. It seems similar to how many bulbs are designed from what I saw in bigclivedotcom's video:

I will try different bulbs just to see if it makes a difference and I'll talk to the neighbor to see if the unit might have some issues.

Here is the PCB picture if anybody is interested:

[TheMG]

You mentioned the neighbors A/C unit ramping up and down... could it possibly be one of the newer type with a variable speed compressor? That type would have an inverter (a type of VFD but some such compressors are actually BLDC motors). Anyways, because those are electronically driven they can add some distortion to the power waveform, if the current drawn is not perfectly sinusoidal.

[moffy]

Fantastic photo, great detail, were there any other components on the board or in the light, specifically any energy storage e.g. inductor or capacitor?
P.S. Watched the video which answered my question.

[StillTrying]

A loose bulb.

[lukash]

You mentioned the neighbors A/C unit ramping up and down... could it possibly be one of the newer type with a variable speed compressor? That type would have an inverter (a type of VFD but some such compressors are actually BLDC motors). Anyways, because those are electronically driven they can add some distortion to the power waveform, if the current drawn is not perfectly sinusoidal.

I do not think so but I haven't had the chance to look at it yet. I am only judging it by being a really loud older unit and it always sounding the same, so I do not even think it is a two-stage. I'll go talk to the guy and see

[lukash]

Fantastic photo, great detail, were there any other components on the board or in the light, specifically any energy storage e.g. inductor or capacitor?
P.S. Watched the video which answered my question.

The EC1 component is an electrolytic capacitor that is on the bottom, I think I heard they mount it like that to get it out of the way and minimize heat transfer to the capacitor.