After noticing some of my lights have been randomly flickering, I was concerned that we'd lost our 7.2kV neutral again.

To better characterise what was going on, I hooked my scope up to the AC line. Yeah, that felt a bit scary, but the pre-worry was out of proportion to just actually doing it. It’s not like I’ve never scoped the line before…geez (I didn't connect probe ground, instead relying on the scope's AC ground). Sure enough: glitches galore. I then realised I’d get a lot more out of the experience by scoping both phases of the line. Good choice:

MD_neutral-woes_01 (cr) Figure 1: Short glitches

MD_neutral-woes_02 (cr) Figure 2: Longer glitches

MD_neutral-woes_03 (cr) Figure 3: Freezer startup (I think)

The complementary waveforms scream “bad neutral!” (the neutral centre-tap from my transformer secondary that is, unlike the missing HV neutral of my past woe).

I walked the long 240V line (75m-ish) from the house to the transformer, and… Tada. A small tree was resting against the line just steps away from the transformer. Though no damage was visible, the coincidence seemed too great.

The electric utility arrived a few days later (the day after I called in a panic because of extreme flickering, and reading peaks up to 150V—on a slow DMM!), dealt with the tree and after inspecting the transformer and line to the house, said they found no damage. My still-flickering lights said otherwise.

MD_neutral-woes_04 (cr) Figure 4: One of the two support poles between the house and transformer. At first, I thought those "clamp" blocks were splicing the two neutral segments with a jumper, but no, the neutral wire goes straight through. That rules out the chance of a bad connection there.

They insisted on poking around my AC panel (can’t blame them, but of course I knew it was fine), and after finding nothing, I convinced them to take another look at the transformer and the wire connections there.

My power went off once or twice as they worked (though I’d turned off most of the breakers anyway to guard against possible spikes). Then it came back on for the final time, and…and…??? Lighting bliss! Not only had the flickering stopped, but I’m pretty sure my line was more stable than ever. I surmise the neutral was poor to begin with, and the shock of the tree landing on the line just made it worse. I may never get an explanation of exactly what they fixed, as they didn’t return to tell me and the utility has not been able to give me an answer yet. But...

MD_neutral-woes_05 (cr) Figure 5: The transformer, before the utility arrived.

MD_neutral-woes_06 (cr) Figure 6: The transformer, after repairs.

If you compare the transformer images, you'll notice the difference. There are two shiny new splice connectors in the after picture. They don't make any new connections per se, but obviously the existing connections had gone bad. I guess that's what they fixed.

During this investigation, I also noticed that my AC power looked pretty distorted:

MD_neutral-woes_07 (cr) Figure 7: I've seen prettier sine waves.

MD_neutral-woes_08 (cr) Figure 8: This is the spectrum I used for my distortion calculation.

As my scope doesn’t compute THD (total harmonic distortion), I threw together a quick & dirty spreadsheet into which I entered the worst of the line harmonics. The result: about 4% THD. Is that reasonable?

Michael Dunn is editor-in-chief at EDN with several decades of electronic design experience in various areas.

First published by EDN.