Thoughts On Stray Voltage

[Salty Cracker]

So had a return pump fail in an unfortunate way recently....

Must have been a split in the case or something, and the tank filled with stray voltage. Enough to give your finger a strong zap when you touch the tank. Now I had a bubble magus pump fail in a similar way a few years back, but nothing like this. At that point you had to have a cut on your finger to feel it, this one was an all out ZAP. Not enough of a problem to flip the gfci, but enough to wipe out a whole lot of coral.

I'd say I suffered an 80% loss before I found the culprit. I was checking RO/DI, swapped out all media, was checking parameters, and nothing seemed to be too far out, but the house smelled like sps when you pull it out of the water and sniff it...x100.

So wondering if there is any kind of voltage monitor besides hooking a multimeter up to the tank. Considering this is my 2nd experience with this (I had suspected the pump previously, it had started to get noisy, but after a good cleaning it would quite down for a while).

Anyway, there you go, stray voltage in the sump can pretty much wipe out a tank of its coral inhabitants. Fish seemed completely unaffected, but who knows if it was rough on them or not. I'm in the recovery phase, it's weird what did and didn't make it. Rainbow chalice? Made it. Torch? Dead. Zoas? Closed up but fine. Blue millie? Fine. Orange millie? Obliterated. Like I said it was really hard diagnosing the problem, as there didn't seem to be a culprit. Now I just need a ton of carbon and hi-cap GFO to scour the tank with, oh and when is the next fragfest?

Big thanks to Matt for hooking me up with a "new" return pump. I probably should learn my lesson, the waveline pump that failed had been bought used, I probably really should buy a new pump, but damn they're expensive.

 

[TORX]

Long story short...

Get a grounding probe.
Take it out and test with multimeter periodically with maintenance.

Also GFCI does not protect against these shocks. There is a different outlet for that. I have an article somewhere about it that breaks it all down.

Sent from my SM-G950W using Tapatalk

 

[Salty Cracker]

Long story short...

Get a grounding probe.
Take it out and test with multimeter periodically with maintenance.

Also GFCI does not protect against these shocks. There is a different outlet for that. I have an article somewhere about it that breaks it all down.

Sent from my SM-G950W using Tapatalk

Went through a lot of research when this happened previously, and grounding probes are not going to happen. Short story is I would have had voltage streaming through the water, instead of just lingering in it. Many suspected Hole-in-the-head disease can happen from streaming voltage, and sps growth was very much suspect as well, people having all sorts of issues with their tanks that they couldn't nail down. I had just a grounding wire in the last tank hooked to ground on an outlet, but decided against it on this tank. I may not have known there was a problem with the pump if I just grounded the tank. If there's a different outlet that can be used, I'd look into that, although I wouldn't want something with a hair trigger that shut down the power every time there was a thunderstorm. Basically I don't want to mask the problem, I want a really quick alert if it happens. Just think about how much potential voltage we have in our sumps.

I'm thinking of designing some sort of arduino module that monitors voltage, and sets off an alarm if levels change. Probably wouldn't be compatible with a reef controller, but it would be better than a tank crash (this is the first one that I was not actively negligent with, as I was constantly monitoring the tank, and noticed the problem almost immediately)

 

[Pistol]

The gfci won't work properly if the water is not grounded

 

[Salty Cracker]

The gfci won't work properly if the water is not grounded

Well then I guess I'm doomed, cuz I'm not grounding the tank

I did drop an electrical cord in the sump when I first set up the tank, and that of course tripped it, and I would like to think that a surge grounding through my body enough to do harm would trip it as well, but ideally an alarm on the tank that shows there's stray voltage in the tank would keep me from putting my hand in there in the first place. I may even design something that shows the stray voltage, maybe wire in an inexpensive multimeter. I just thought maybe someone had already designed one.

 

[Josh]

Ok so THANK YOU SO MUCH FOR THIS STORY. I have been battling a WTF situation too and guess what i just did. Measured 30 F*&%ing volts in my water. Lesson learned here. Looks like a bad heater, so thank you for sharing your story you may have just saved all my acans.

 

[Matt1997]

Volts or millivolts?

 

[Josh]

Definitely volts

 

[Salty Cracker]

I have to question the "won't work without a ground" thing...since I have sold houses without grounded wiring, where gfci plugs have been used (or gfi if there is a difference) to simulate a path to ground in case of failure... I mean, the neutral line is still just connected to ground at the panel, so I can't see a gfci not working if there was no ground, it would just be a sudden short across the terminals wouldn't it? Dang I wish I had taken electrical wiring in school
I didn't measure my stray voltage, as I just yanked the pump as soon as I got the nasty jolt. I was assuming I was feeling ~5-10 volts judging by what I felt, but I also could have been touching something else in the sump (the led fixture in there has given me a jolt too when it got wet from the skimmer). I'd have to check the power output on the brick for the pump to see what it could have been putting out at max.

Heaters would be a whole other issue entirely. Heat usually equals a fair bit of current, and that can be dangerous. I would like to assume that the full current of a heater into the water would definitely snap either gfci or the breaker on the panel, as 15amps very likely could be set loose, but a motor can only put out whatever the power supply allows (I think). I dunno, I only took basic electronics. I'm swapping out the fish-street pump tonight for the M1 (I really thought the discount pump would have been the one to fail, it has a 3" spark that comes out of it when you hook/unhook the power connector, and that doesn't seem good). 100% it was the pump though, the carnage in the tank stopped immediately.

I'm chatting with my buddy from my stereo/amp forums that is some sort of electronic prodigy, seeing if I can part together an alarm, he likes challenges so he'll likely help me with the schematic, and I may build a prototype for what I want for the tank. I doubt I'll be offering any up for sale, as I think getting safety approval for something like that would be virtually impossible (and cost prohibitive) but if I do get a useful module made up I may release the schematics, I think this is something every reefer should have (and maybe everyone with an aquarium). If I've had this twice in 20 years, then it's just a matter of time for anyone else, as josh may well have just discovered.

 

[Pistol]

The gfci or gfi monitors the current on the hot wire and the neutral wire and if there is difference of about 4 ma it assumes there is a ground fault and trips. Your water is not grounded so it becomes electrified but there is no current flow therefore no ground fault, a ground probe would give a ground path that bypasses the neutral wire thus causing a ground fault and tripping the gfci.
In the case of ungrounded wiring if current from a shorted appliance ( say hot wire shorted to the metal case of a fridge that can't go to ground and trip the circuit breaker) tries to take another path through your body to ground say through a conductive floor or another ground source the gfi will see the current is lower on the neutral wire than the hot wire and assume a ground fault and trip thus preventing your electrocution.
If nothing is grounded then there is never a ground fault.

 

[Josh]

You might have a hard time because any magnetic based equipment produces an induced voltage into the water, it's completely fine. Most tanks have 10 to 15 volts if you measure them. Its when you are up higher that you have an issue. In case of my heater shouldnt have been adding to that number and it was.

Measure your tank if you dont have a grounding probe I bet you read 5 to 25 volts.

Creating an auto alarm would be difficult. The best thing you could do is set an alarm on a gfci and plug it in the outlet below your tank. Gfcis trip if any outlet in the circuit grounds so it should react to your tank but not shut your tank down. Past that it's about how to use the gfci to notify you something happened.

It's something I've done at work. We called it tattle tails

 

[Salty Cracker]

The gfci or gfi monitors the current on the hot wire and the neutral wire and if there is difference of about 4 ma it assumes there is a ground fault and trips. Your water is not grounded so it becomes electrified but there is no current flow therefore no ground fault, a ground probe would give a ground path that bypasses the neutral wire thus causing a ground fault and tripping the gfci.
In the case of ungrounded wiring if current from a shorted appliance ( say hot wire shorted to the metal case of a fridge that can't go to ground and trip the circuit breaker) tries to take another path through your body to ground say through a conductive floor or another ground source the gfi will see the current is lower on the neutral wire than the hot wire and assume a ground fault and trip thus preventing your electrocution.
If nothing is grounded then there is never a ground fault.

Ah okay, that makes sense. But it does come back to the whole bit about stray voltage turning into actual current if there is a path to ground, and if that current was low enough, you might never know there was leaky voltage, no? Also, this being a DC motor, would it make a difference to this equation? (I honestly don't get the nuances between amps, volts, current, AC/DC etc). I'm taking an online course oddly enough trying to get a grasp on it, but honestly I am doing that to deal with electronics voltages, not mains voltage...

I also regret not taking welding in high school. No idea what good geography has ever done for me, but welding would have been useful soooo many times (I almost suck worse at welding than I do at electricity theory)

 

[Salty Cracker]

You might have a hard time because any magnetic based equipment produces an induced voltage into the water, it's completely fine. Most tanks have 10 to 15 volts if you measure them. Its when you are up higher that you have an issue. In case of my heater shouldnt have been adding to that number and it was.

Measure your tank if you dont have a grounding probe I bet you read 5 to 25 volts.

Creating an auto alarm would be difficult. The best thing you could do is set an alarm on a gfci and plug it in the outlet below your tank. Gfcis trip if any outlet in the circuit grounds so it should react to your tank but not shut your tank down. Past that it's about how to use the gfci to notify you something happened.

It's something I've done at work. We called it tattle tails

Hmmm, but let's say you get a base-line of ~5 volts in the sump, and you could set the meter to alarm if there is >150V in the tank, couldn't that work?

I just tested the sump.... I'm reading .5 volts, almost identical on 2 different meters. Now I know with my old setup that had an AC return pump, I had more than 5V in the sump all the time, so -possibly- the DC motor that's there makes a difference (I read from the water in the center of the sump to the ground on the outlet).

 

[Josh]

Correct DC pumps should give you no induced voltage. Best way of thinking about electricity is to think about water. Voltage is the pressure in the lines. Aka the potential to flow water. Current is the actual flow rate. You can have pressure in a pipe without the water moving. Similar to a voltage in a tank. If it has nowhere to go can it really flow? How it would work with a gfi is using a ground probe but having it trigger the gfi rather than the aquarium gfi. Careful how with the wiring because 1 gfi can cover up to I think 8 outlets on the same circuit.

 

[Josh]

The complicated part if you are getting your electronics buddy to help you is the fact that most electronic circuits use technology like scr's which are like the ones in apex eb8's. They trigger off DC voltage which is basically battery/car voltage. The stray voltage is likely going to be AC. Any type of switching methods for detecting it cant use the actual power source to operate the alarm. Stray voltage contains micro amps, it is enough power to kill a person but not enough to operate an alarm or a relay.

I'm rusty as all hell with this but 16 years ago I went to college for this lol.

 

[TheGreatWave]

.

 

[Salty Cracker]

The complicated part if you are getting your electronics buddy to help you is the fact that most electronic circuits use technology like scr's which are like the ones in apex eb8's. They trigger off DC voltage which is basically battery/car voltage. The stray voltage is likely going to be AC. Any type of switching methods for detecting it cant use the actual power source to operate the alarm. Stray voltage contains micro amps, it is enough power to kill a person but not enough to operate an alarm or a relay.

I'm rusty as all hell with this but 16 years ago I went to college for this lol.

I -really- wish I had taken electrical engineering in uni, but I really had no direction, so I ended up with a very useful social science degree.

So I suppose that simply hooking a multi meter to contant power and leaving it running in the sump would be equivalent to putting a ground probe in there, but you could manually check the voltage and see if there is more current than normal? Not really an alarm, but one more "test" that you could check during regular maintenance or if there is trouble?

Electronics whiz is some sort of super-guru, he used to teach electrical engineering but now has his quantum physics degree. He talks on a level I can't even comprehend, but I think he enjoys our interactions because our conversations go off in weird directions and I think it helps him think in different directions (he should just try weed). He's got me messing with modding the power supply on my signal generator so that my equipment is 'up to snuff' so that I can find a gremlin in my pre-amp (another hobby) that wasn't related to the 128 caps I replaced. I love electronics but don't understand almost any of it (little men have 3 ways to get from the top floor to the bottom floor, some paths are easy, some are hard, when they reach the bottom they take an elevator back up to the top and start again, this is resistors in parallel!)

 

[Salty Cracker]

I wonder what the effect would be of shielding pumps and motors in metal. I'm assuming this stray voltage is induced magnetically.

There is also an older heater that I used during my issues lately getting rid of aiptasia. Anyhow I lost a bunch of SPS to what I assumed was either Alk or too much light. Back when I started I had what I thought were voltage issues and come to think of it this was the heater I used, one of those titanium style with the separate probe. The issue never fully came to light because I have changed equipment and got a little better at reefing then I was at the start. I know I replaced one little pump that was responsible for some of the voltage as the replacement read lower. Perhaps what was remaining was induced voltage, I don't know.

This thread is a good reminder to maybe somehow check our tanks every once in a while. The tank I am setting up has mostly new equipment, maybe I should take some measurements to have a bit of a baseline to compare future readings to.

If I hadn't gone through this before I would not have thought of it this time. Last time it wasn't as major or pronounced, and resulted in me upgrading a lot of things, so that was good. At this point I didn't have much to upgrade, so it had to be something basic. I knew it wasn't what the LFS guy used to call it: "old tank syndrome".

 

[Pistol]

Most multimeters have 10megaohms of impedance so no ground probe effect. an old school analog meter would give some current flow

 

[jroovers]

Here's an older thread that I started that covers some of the same ground (no pun intended):

http://thefragtank.ca/community/thr...tage-etc-for-the-electrically-ungifted.10623/

Definitely a difficult topic to get one's head around for those who are not electrically inclined. In my old 120 I definitely had stray current I think from my DIY T5s and reflectors, and I had HLLE issues in a couple of tangs and angels... not definitive by any stretch but I think there is a possible correlation there for sure, as all my fish had clean water and a great diet.

Interesting about your thoughts on current and SPS dying Salty, in some parts of the ocean they actually use current to help stimulate coral growth... but maybe you had way to much and its not comparable in such a small enclosed system vs. the vastness of the ocean!

https://www.globalcoral.org/to-rebuild-coral-reefs-quickly-just-add-electricity/