Re: Wow! Improper bond...
The comment in the video about the voltage seeking ground in this case was not entirely correct. Electricity, contrary to what we are all taught in school, does not seek ground. It seeks it's source. The earth being a big resistor does not allow for the flow of electricity very well.
The anomaly you saw in the video was two different electrical potentials trying to equalize. To understand electrical potential, you need to watch this video.
https://www.youtube.com/watch?v=SkFH8lLvKZ0 The man sitting on the helicopter and the helicopter itself are in the air at one electrical potential. when they approach the power line, which is at a different potential, there is a clear danger. The probe the man uses is connected to the chassis of the helicopter and when he reaches out with it equalizes the potential differential and the clamp he puts on the wire ensures it stays the same.
In the pool, when a person reached out to grab the railing, they were acting like the probe. They were completing a path between two different potentials. The amount of current flow was not great enough to kill them but, surly it was enough to cause some discomfort and even slight paralysis.
The interesting study here is the metal railing in the pool water and connected to the deck. Several years ago this was thought to be a good water bond to the bonding grid. This did not seem to be the case here. I would have liked to test that pool right after this happened. I am going to surmise that there was no potential difference between the water and the railing in the direct vicinity of the railing but, there was voltage gradients throughout the pool.
Someone mentioned why a GFCI did not trip or if there was even one present. It is entirely possible there was a GFCI present on the circuit (I doubt it though). If you look at the chart I posted below...
1mA Perception Level
4-6 mA UL 943 Class A GFCI Trip Level
10mA Let-Go Threshold
15mA Muscles “Freeze” in 50% of the Population
30mA IEC RCD Standards Breathing Difficult, Fibrillation in children
50 mA - 4 Amps Fibrillation
4 AMPERES AND OVER Heart Paralysis, Serious Tissue and Organ Burning
You will see that the UL listed current threshold to trip a GFCI is 4-6 mA. If you are subject to that much current you will feel it. I was shocked one time while plugging in an extension cord. It was enough to cause me to get a little weak in the knees and let go of the cord. When I realized what had happened I figured the GFCI must have tripped after the event. It did not. The current I felt must have been below the trip threshold for the GFCI but it was still enough to startle me.
Someone also mentioned that the electricity that is sometimes referred to as stray voltage as coming from, say, a construction site or bad underground wire. While these can be sources, one has to realize that voltage potential gradients are around us everywhere. Most gradients are so slight that when you touch something the equalizing difference is so slight that you don't feel it.
About 5 years ago. I had to fix a bonding issue at a pool. it was an above ground pool and when they were standing on the ground in a certain area of the pool there was a voltage difference measurable between the water and the ground. If you touched the water while standing in your bare feet it would hit you pretty hard. A little harder than a static shock when you touch a door knob (two potentials equalizing) I found that the installer did the 4 connection points wrong and that they were all at one end of the pool. The owners thought that the power line that was running along the back of the property was the culprit. I could not convince them otherwise until an event happened that gave me the opportunity. We had a major power outage. The type which the entire town/county is out. Knowing I had to move fast I went over there and disconnected the bonding grid. I then measured the voltage difference between the water and ground and showed them the reading. It was actually higher than it was before. Knowing that the power line at the rear of the property was the one feeding the neighborhood I was sure there was no power flow at that time. I reconnected the bond and the voltage differential went almost to zero.