PDA

View Full Version : Electrical Q


christyf5
03-20-2010, 04:31 AM
Earlier today I set up a mag 7 on a reactor. Everything was fine. The mag 7 is plugged into my aquatronica power bar. During a water change I turned off everything in the sump (I do my waterchanges via the sump) did my waterchange and restarted everything. Upon restart of the mag 7, the GFCI tripped. I thought it odd but turned off power to the halides (they suck the most power 2x250W) and tried again, tripped the GFCI again.

I unplugged the mag 7 and tried plugging it directly into the GFCI, tripped again. I then plugged the mag 7 into a non GFCI circuit and it works fine.

However, now I'm wondering wtf is up with the mag 7. Is it leaking power? I have no grounding probe.

Its my understanding that it doesn't take much to trip a GFCI but it takes more to trip a breaker? The plug is completely dry. Why did it work fine earlier and now its causing problems?

So my question is, if something trips a GFCI and not a breaker, should I be removing it from my tank or is it just "one of those things" and its probably fine. The fish look fine however I don't want to wake up in the morning to a bunch of dead or wonky looking fish.

H22_TURBO
03-20-2010, 04:40 AM
maybe try plugging it into another gfci and test it? could be the gfci is no good anymore???

christyf5
03-20-2010, 04:52 AM
but everything else on the GFCI is fine when I run it without the mag 7 so the GFCI appears to be fine (?)

Delphinus
03-20-2010, 05:02 AM
I'm a little bit fuzzy on how a GFCI works in that it either implies power on the ground wire and thus a "leak", or whether there is an imbalance between hot and neutral and thus a "leak". Either way though, I'd say it's not good. I would look at that mag 7 and see if there's anything wonky with the wire where it attaches to the pump itself.

In an electrical leakage situation, the fish themselves aren't ground so they're fine (although it's still not good for them - think it can cause HLLE ?), but as a person with feet on the ground it's bad.

At least that's how I sort of understand it. :neutral:

Starry
03-20-2010, 05:32 AM
could always put the mag in a tub of sw and meter it for stray current. (not with your fingers standing barefoot on concreat)

ottoman
03-20-2010, 05:33 AM
Is it possible that you have a bit of splash or moisture on the plug or GFCI during the water change? Just my thought.

DiverDude
03-20-2010, 05:42 AM
GFCI's work like this: Normally, current flows between HOT (black/larger blade of the plug) and NEUTRAL (white/smaller blade). Under normal circumstances, there is no current flow in the ground wire (green wire).

When someone somehow makes contact with either HOT or NEUTRAL, they can potential conduct (via their body) to ground. This poses a safety risk -particularly if one hand is in contact with the live wire and the other is at ground potential since the path is across the chest and there is a chance of fibrilation (this is BAD, very bad).

Enter the GFCI. It senses if there is current flow in ground. If there is, it surmises that something is not koscher and cuts power to protect the human it supposes is doing this. It's worth pointing out that GFCI's are to protect people, not the equipment. Equipment is protected by the breaker.

So.....this would imply that your pump is leaking to ground somehow -or the GFCI is faulty. I'd try plugging the pump into another GFCI and see if it pops.

Does the Mag7 have a 2-prong plug or a 3-pin (grounded) plug ? If it has a grounded plug and pops another GFCI, I'd replace the pump.

outacontrol
03-20-2010, 05:44 AM
Okay, here is how a GFCI works, it measures the current flow on the white and black wire and compares the 2 values, if the value is different by more than 6mA (0.06 amps) it will shut off the Circuit, because that current must be faulting to ground. Old GFCI receptacles had no self check circuitry in them and if they failed internally there was no indication until an accident or external test occured. New GFCI receptacles (sometimes labeled "smart lock") have an internal self check circuit and if they have an internal fault they will shut off.
So how old is the receptacle if it is newer than 2 or 3 years it might be a smart lock, and just need to be changed.
I would plug the pump into another GFCI (after checking all the wires for any damage or salt creep or moisture) and try that to see if it works.
Feel free to ask me any other questions.
I am a Master Electrician.

Delphinus
03-20-2010, 05:54 AM
Everytime the question of how a GFCI works I see exactly this (which is why I said I wasn't sure). One post that says "current on the ground = trip" and one post that says "hot/neutral amp imbalance = trip". I know a few years ago I caused a GFCI to trip on an otherwise empty circuit (ie., nothing was on that was plugged in) because I touched my light reflectors and got a static shock. That suggests "current on the ground", but the "hot/neutral imbalance" makes more sense to me or at least would be a safer option because if electricity can escape out of the wires into the tank, it's not necessarily going to work it's way back to the same ground wire that the GFCI is on.

Is is possible it is a combination of both options? Sorry, I'm not trying deliberately to be dense, but I am legitimately confused since I always see the same two different explanations when this topic comes up.

Either way though - if the GFCI itself isn't faulty then it's bad that the mag pump trips the GFCI. It may be the end of the road for the pump.

outacontrol
03-20-2010, 06:00 AM
the current on the ground wire would not in it self be dangerous, the danger occurs when the current leaves the wire or device and travels through you to ground.

the GFCI absolutely without doubt measures and compares the current in the hot/neutral (if you want to call it that) and respondes to an imbalance between the 2 wires of 6 mAmps.

here is a link: http://www.google.ca/imgres?imgurl=http://www.codecheck.com/cc/imagetoo/gfci_circuit_NS.gif&imgrefurl=http://www.codecheck.com/cc/gfci_principal.htm&h=517&w=763&sz=13&tbnid=0S55_ENy3ko06M:&tbnh=96&tbnw=142&prev=/images%3Fq%3DGFCI&usg=__lhyt2PMPb0FVUriwNPLcXmchGb8=&ei=12SkS5b1O4K6NbmqtdYI&sa=X&oi=image_result&resnum=9&ct=image&ved=0CCYQ9QEwCA

Delphinus
03-20-2010, 06:25 AM
Cool, thanks. So, question though, why would it have tripped on the static discharge on the ground line? Does that make any sense to you? Sorry, just trying to understand why that happened. (BTW, sorry for the hijack, Christy..)

outacontrol
03-20-2010, 06:50 AM
Hard to say, was the lights plugged into the GFCI? was the cord for the lights a 3 prong or 2 prong and was id a DIY or brand name bought fixture?

Chase31
03-20-2010, 07:17 AM
well if you work up the static and it sends it down the "Neutral" wire it could set off the gfi, a static shock is around 5 Amps, so it could set it off

Doug
03-20-2010, 01:37 PM
Christy if you remember the time I had the leak in the stream pump that caused the arc from the halide fixture to my head, :lol: , it never tripped out a standard plug but tripped the gfi plug in. I could also measure that leak with a meter from my tank to ground.

muck
03-20-2010, 01:43 PM
Had a Mag 3 pump leak current on me.
Found out when i went to adjust it while it was running in my saltwater mix bucket with my hand. BAZAPP!!

christyf5
03-20-2010, 04:04 PM
I didn't notice any zaps when I put my hand in the water but I wear crocs around the house so I don't know if those would make a difference. The pump runs fine and there are no odd wiring where it attaches to the pump. It was used as a calcium reactor pump for a number of years and has sat around for about 3 years or so doing nothing. This morning I think I might try it in a bathroom GFCI and see if it trips all of them or just the one I want to plug it in on.

As for the GFCI's I think they're all brand new or just a couple months old.

No splashes or moisture on the GFCI its well above and to the side of the tank (plus I inspected it, no sweats or salt creep).

Lance
03-20-2010, 05:08 PM
I'm betting that the GFCI is wonky. I've seen several of them trip for no apparent reason when they get older. They seem to wear out much faster than a regular breaker, especially when they are switched on and off again repeatedly.

Delphinus
03-20-2010, 06:21 PM
Hard to say, was the lights plugged into the GFCI? was the cord for the lights a 3 prong or 2 prong and was id a DIY or brand name bought fixture?

Does it really seem hard to say? The static charge went down the ground wire because the reflector was grounded. Nothing else was on at the time. No current leaked to the hot or neutral wires. Just the ground. Zap. Ground wire. GFCI tripped despite nothing plugged in was on. So it just seems fair to me it could be a combination of current imbalance between the hot and neutral ("black and white") wires and checking for any potential capacitance or whatever it's called, on ground. Ah well. It doesn't really matter to me. Point is it shouldn't be tripping.

Christy - my suggestion for the easiest thing to try is find another GFCI outlet (eg bathroom plug), turn it on and see what happens. If it trips you know it's not the plug. If it is the plug though then at least that's cheaper than a new Mag7.

But I've had Mag drives leak current after a few years of use. I think they get little microfractures in the housing over time. Not sure the explanation though.

PoonTang
03-20-2010, 07:04 PM
may also be that since you had just done a WC than there were splashes and dribbles all over the place around your sump. If you created a path to ground through a simple dribble down the outside of your sump that may have been enough to let the GFI trip. Give it a try today now that everything is dryed out.

outacontrol
03-20-2010, 07:36 PM
[QUOTE=Delphinus;503726]Does it really seem hard to say? The static charge went down the ground wire because the reflector was grounded. Nothing else was on at the time. No current leaked to the hot or neutral wires. Just the ground. Zap. Ground wire. GFCI tripped despite nothing plugged in was on. So it just seems fair to me it could be a combination of current imbalance between the hot and neutral ("black and white") wires and checking for any potential capacitance or whatever it's called, on ground. Ah well. It doesn't really matter to me. Point is it shouldn't be tripping.

WTF!

IS THERE A REASON YOU ARE MOCKING WHAT I AM SAYING? DO YOU HAVE ANY ELECTRICAL TRAINING?
I helped you when you could not get your 3-way switching working. Not impressed at all.
Did you read any of that link that I posted? A GFCI does not function any other way than I described it measures the current on the 2 current carrying wires (the hot and the neutral that you like to call them). It does not measure or check for current on the bare bonding wire (or ground as you like to call it) at all. Period no questions about it.

mike31154
03-20-2010, 08:48 PM
I'm pretty sure there's no mocking intended here, he's just posting what happened to him with regard to a static charge. I see that the schematic at the bottom of the page (Neutral Transformer Approach) for the GFI device (from your link) contains a couple of coils which I assume sense the electromagnetic fields of the hot and neutral wires. This signal is fed to an IC which appears to have the function of controlling the relay which trips the device. The circuit is tuned to be quite sensitive so that it cuts the power with only a fraction of an amp imbalance. Any solid state device and inductance coil/transformer is susceptible to a static charge or EMI (electromagnetic interference). I'd say it's entirely possible that a static charge or EMI pulse could trip the device even if nothing is plugged into it. As long as line voltage is present at the input of the GFI device, it is sensing, whether it's actually powering something or not. If you read down farther on the page you linked to, there's a case of lightning tripping a GFI device. So IMO, there are more ways to trip these devices other than the 'designed' imbalance between line/neutral.

christyf5
03-20-2010, 08:59 PM
I plugged the pump into a bathroom GFCI and it tripped, guess the pump is done. Anyone need a mag7 impeller? :wink:

mike31154
03-20-2010, 09:06 PM
Oh no, not yet!! You need to liberate the smoke from deep inside of that bad boy first, ha ha.

christyf5
03-20-2010, 09:07 PM
Oh no, not yet!! You need to liberate the smoke from deep inside of that bad boy first, ha ha.

:rofl:

Gary
03-20-2010, 10:24 PM
GFCI's work like this: Normally, current flows between HOT (black/larger blade of the plug) and NEUTRAL (white/smaller blade). Under normal circumstances, there is no current flow in the ground wire (green wire).

When someone somehow makes contact with either HOT or NEUTRAL, they can potential conduct (via their body) to ground. This poses a safety risk -particularly if one hand is in contact with the live wire and the other is at ground potential since the path is across the chest and there is a chance of fibrilation (this is BAD, very bad).

Enter the GFCI. It senses if there is current flow in ground. If there is, it surmises that something is not koscher and cuts power to protect the human it supposes is doing this. It's worth pointing out that GFCI's are to protect people, not the equipment. Equipment is protected by the breaker.

So.....this would imply that your pump is leaking to ground somehow -or the GFCI is faulty. I'd try plugging the pump into another GFCI and see if it pops.

Does the Mag7 have a 2-prong plug or a 3-pin (grounded) plug ? If it has a grounded plug and pops another GFCI, I'd replace the pump.


Not sure if you have a typo, but the neutral (white) should be the larger blade of the plug not the hot (black).

Lance
03-20-2010, 10:39 PM
Oh no, not yet!! You need to liberate the smoke from deep inside of that bad boy first, ha ha.


Good one Mike! I love it! :pound:

Delphinus
03-20-2010, 10:52 PM
WTF!

IS THERE A REASON YOU ARE MOCKING WHAT I AM SAYING? DO YOU HAVE ANY ELECTRICAL TRAINING?
I helped you when you could not get your 3-way switching working. Not impressed at all.
Did you read any of that link that I posted? A GFCI does not function any other way than I described it measures the current on the 2 current carrying wires (the hot and the neutral that you like to call them). It does not measure or check for current on the bare bonding wire (or ground as you like to call it) at all. Period no questions about it.

Whoa there dude. Please be assured I am not mocking you. Not in the least. In fact rather the opposite, I am trying to use your electrical knowledge in an effort to better understand something I saw. I think your issue is the fact I asked if if it's hard to say. I didn't mean that as mockery, I really meant that at face value - is it hard to say if the static discharge was responsible for a GFCI trip? The reflector was most assuredly grounded, so the zap of the static absolutely did travel down the ground wire. I was merely asking if it was a possibility that GFCI's could be tripped by more than input condition.

Breathe easy my friend. I will buy you a coffee or whatever your favourite equivalent beverage may be one day should we ever meet in person. :) How a GFCI works is less important to me than whether they work at all. It was only curiosity on my part.

At any rate however ... regardless of how a GFCI trips, a pump shouldn't trip the GFCI. Now that it's confirmed that the pump itself seems faulty (ie., confirmed trip on a different GFCI) I'm afraid the end result for Christy is that she needs to replace her pump. :( Sorry Christy. FWIW, I had to garbage can a couple mag drives too over the years for the same reason. It's too bad, they are otherwise really reliable workhorses.

DiverDude
03-21-2010, 12:01 AM
*sigh* TWO errors in one post....what WILL I do with me ?

First, I stand corrected by outtacontrol. How I described the function of the GFCI was what had been taught to me by someone who should know better (an electrical engineering professor). However, schematics trump credentials and I now know better -my apologies for the mis-information !

Second, someone give a free frag to Gary for paying attention. He is correct, Neutral is the larger blade of course and I need to type slower or think faster. More mis-information; more apologies -only this time I can't blame anyone but myself.