Ground Fault Protection

Started by keyturbocars, March 17, 2011, 05:51:04 PM

Previous topic - Next topic

keyturbocars

Hello Guys,

Here's part of a discussion we were having on the NAWS forum related to grounding.  Below are some comments that I cut and paste from NAWS member BB (Bill).

<start of Bill's comments>

Ok, I am going into deep water here... And take what I say with a grain of salt. boB or Ryan would probably a much better source of information regarding this than I.

As I understand people's solutions to the NEC requirement for ground fault protection for DC solar arrays, it really does violate the very definition of safety ground.

Pretty much all DC GF systems put a 1-5 amp fuse/breaker/ or in the case of the Classic a 1 amp PTC Resistor (basically a resistor that when it passes too much current gets hot and dramatically increases its resistance) between DC ground and safety ground (add missing text).

At this point, the popping of a fuse/opening a breaker will open the connection between the + solar array and the DC charge controller (or GT inverter) (the 5 amp breaker trips, then this trips the 80 amp breaker too with the handle bar/rod).

In the case of the Classic and a few others, the "high voltage" across the open fuse or hot PTC signals the charge controller (or GT inverter) to stop conversion.

Sort of sounds OK--but, from my humble point of view, all they have done is stopped the power transfer into/through the DC input of the device.

The have not limited to less than 5 milliamps (~limit that can cause heart failure), and they have now potentially energized the DC grounded section of the circuitry by "floating" the ground (open fuse, open breaker, hot PTC)...

In the "olden days", when something was ground referenced, it was done with at least a 6 gauge piece of wire that could handle >600 amps of fusing current and would guarantee to trip a circuit breaker/fuse in the + or hot leads from the energy source (DC or AC systems).

Now, your DC "ground" is a 1-5 amp fuse, that when popped, it has now un-referenced your ground system. And, depending on what happens, it has de-referenced your otherwise operating battery bank--while just stopping the solar panels from charging the batteries (hope you notice in time the warning LED, beeper, etc.).

This is typically considered very bad form (again from my humble opinion) to have the ground automatically lift... Heck, when I designed/installed AC equipment we had to use a separate ground stud so that when a person was servicing the device they could not unhook the green wire by accident (or double nut the safety ground/other ground connections).

<end of Bill's comments>

We were just wondering what you Midnite guys thought about the safest way to ground DC equipment (traditional method of grounding negative battery connection versus newer GFP method).

Thanks,

Edward

boB


I may have to read BBs post again, but the intent for PV GFP is not the same as for bathroom AC GFCI protection.

5 milliAmp protection for GFCI is to reduce electical shock and PV GFP is to reduce electrical fires.  This is why
the current is different.  Not worried about voltage, just current through a less than optimal connection....

i.e.  a bad connection with high resistance.  A nice solid ground fault will usually be fine because there
would be little or no heat buildup.

boB


K7IQ 🌛  He/She/Me

Halfcrazy

I will just simply say "That is the way the NEC says to do it" do I agree ??? Seriously as a manufacturer we have to follow the code by the letter and could not possibly advocate not using this wonderful device  ::)
Changing the way wind turbines operate one smoke filled box at a time

Westbranch

Hey guys,definitely not an expert or otherwise NEC savvy, so please explain...
From a non understander of the NEC's point, these seem to be in opposition. 

<start of Bill's comments>

-but, from my humble point of view, all they have done is stopped the power transfer into/through the DC input of the device.

The have not limited to less than 5 milliamps (~limit that can cause heart failure), and they have now potentially energized the DC grounded section of the circuitry by "floating" the ground (open fuse, open breaker, hot PTC)...

boB said:

5 milliAmp protection for GFCI is to reduce electrical shock and PV GFP is to reduce electrical fires.  This is why the current is different.  Not worried about voltage, just current through a less than optimal connection....


If I have it right, based on bills comment, there is now another path for the current to follow once the PVGFP is triggered, which would be handling all the PV output.... ??? ???

Eric
KID FW1811 560W >C&D 24V 900Ah AGM
CL150 29032 FW V.2126-NW2097-GP2133 175A E-Panel WBjr, 3Px4s 140W > 24V 900Ah AGM,
2 Cisco WRT54GL i/c DD-WRT Rtr, NetGr DS104Hub
Cotek ST1500 Inv  want a 24V  ROSIE Inverter
OmniCharge3024  Eu1/2/3000iGens
West Chilcotin 1680+W to come

boB

Quote from: Westbranch on March 17, 2011, 08:13:35 PM
Hey guys,definitely not an expert or otherwise NEC savvy, so please explain...
From a non understander of the NEC's point, these seem to be in opposition. 


If I have it right, based on bills comment, there is now another path for the current to follow once the PVGFP is triggered, which would be handling all the PV output.... ??? ???

Eric


Yeah, you probably are right...   A better way to take care of this is probably to short the entire array at the array.

We built a PV GFP in the 1990s at Trace that did that.  It was very expensive.  Opening up most of the path is cheaper.
Not everything the NEC or UL says to do is perfect.  And usually does not take care of a double fault.

However, ground faults are very rare and GFPs usually just show the installer that it was mis-wired in the first place.

Usually, that is.

boB

K7IQ 🌛  He/She/Me

BB.

#5
Hi, it is me, Bill (BB.) from the Wind Sun forum.  ;D

I will give a couple of examples why, I believe, that the NEC GFP type protection is actually making things less safe rather than safer...

First example, typical charge controller setup with "GFP" supplied by 1-5 amp fuse/breaker between safety ground and negative DC ground (does not really matter if Classic or any other brand--details may matter, but I am not the one to answer the specific issue and how a Classic would behave)... Assume charge controller has solid connecting from PV Input (-) lead to Battery (-) lead (no switching, no fuses/breakers).

Setup system per NEC design. GFP 1-5 amp fuse/breaker from (-) controller ground to safety ground.

1. Short Array (+) line to metal frame/solar array mounts.
2. Current flows from + array to safety ground, return path from safety ground to - array, pops fuse/breaker, controller stops + current flow to + battery lead (or trips double pole breaker which opens + array lead to charge controller/battery bank).
3. Now, + array lead is "hard grounded" to safety ground.
- array is now at Vmp/Voc below ground (~17 volts to 150 VDC below safety/earth ground).
4. Ground reference is now carried through negative PV to negative Battery connection to, now floating, negative battery lead at ~17 to 150 VDC (or more) below ground.
5. Your battery bank - bus is now held to 17-150 VDC below ground with XX amps from the solar array worth of current behind it.
6. The + battery bus is now +Vbatt above the 17-150 VDC below ground reference.
7. If you have a car radio (am/fm or CB, Ham, etc.) that uses the case as a ground/safety ground reference... And have an antenna cable with braid to case and the braid tied to earth ground for lightning protection, you have a hard return path through the braid from - Varray to safety ground, through short to + array and can be carrying 10's of amps (or more) of Isc from the array.

So, in the above, your - bus which as always been assumed to be at or very near earth ground / safety ground potential is now at Varray--which is both a voltage hazard and a current hazard.

And, if you have system where you have DC ground loops (not a good idea, but not always obvious that such a ground loop exists), you are now driving Isc current through those incidental ground loops.

There is also are failures which are a fire hazard...

First failure--assume that you have a second fault (or incidental ground loop) with a + Varray to earth ground fault (the only fault that the DC GFP system "protects against). If you trace the entire path, you will find that there is no over current fault protection from Power of the array to the current flow in the negative bus (say you have a small LED circuit with a 2 amp fuse on the + lead, there is no 2 amp fuse in the - lead)... You will have Isc Array pumping full current through the small gauge return lead that has (for this example) a negative ground to safety ground short...

The above failure would only be protected if A) all ground returns are sized for Isc of the array (even ridiculously small LED fixtures, radios, etc.) or B) if all negative return leads are fused/breaker-ed just like positive power leads are (central fuse/breaker panel for negative bus).

And if you follow NEC requirements (for example) 120/240 VAC split phase circuits--Such circuits are protected by two pole breakers--If one breaker overloads, then the second breaker is automatically switched off because it is ganged with the first breaker.

That would mean that, to be compliant with NEC code, all +/- leads would need to be protected against shorts using ganged two pole breakers just like a 120/240 vac split phase system.

Another thought experiment... Imagine I wanted the same GFP system for my grid tied home.

I would put a 1 amp breaker between safety ground and neutral bond. And if that breaker tripped, then it would turn off my Grid Tied Inverter but leave the Utility power connected and operating normally. And now my neutral was floating with respect to safety ground... Would never be allowed in a million years by NEC (as I understand).

I can create more examples--but this is what I was trying to say that NEC DC GFP is actually a huge safety problem and should never have seen the light of day. By its vary action (tripping to open the positive lead to the charge controller/gt inverter) now creates a huge safety concern with the "negative" side of the system.

It takes a minor problem (short from + Varray to safety ground short) which, in itself is not dangerous (remember, all array wiring is either rated to carry Isc*1.25, or in the case of paralleled array strings, each string is protected to Iseries max rating)--And now creates a huge fire hazard with driving current into an unprotected ground/return wire power distribution system.

Again, I have no knowledge of how a Classic (or any other) charge controller/GT inverter operates/connects from Varray to Vbattery--This is my assumption of how, in general, the system operates and the issues that this NEC DC GFP affects the generic DC system.

In decades past (when I designed larger -48 VDC powered computer systems), NEC had some very bad assumptions about DC power in code at that time too.

-Bill

By the way, here is the original thread from the Wind-Sun forum:

http://www.wind-sun.com/ForumVB/showthread.php?t=10650

Kent0

#6
The ground fault protection specified in NEC 690.5 is intended to reduce fire hazard not to protect personal. Evan at that purpose, it's hardly perfect. And it certainly creates other risks.

When GFP is is implemented with a fuse or breaker, a negative to ground fault is very likely to be undetected. That leaves the system unprotected against positive to ground faults. Even if the GFP fuse of breaker operates, the fault current will not be interrupted.

When the GFP is activated, the normally grounded PV wire may be operating at -17 to -600 volts. That's why 690.5(C) requires a warning label at the GFP device that states that when then GFP is activated there is an electrical shock hazard - normally grounded conductors may be ungrounded and energized. A person that isn't familiar with these risks, shouldn't be trying to find a ground fault in a PV array!

Bill, your concern about the the negative battery bus being operating below ground potential is incorrect. When the GFP is activated, the battery is still grounded to earth through a high impedance. Voltage measurements to ground would be completely normal.


BB.

Quote from: Kent0 on March 18, 2011, 02:06:07 PM
The ground fault protection specified in NEC 690.5 is intended to reduce fire hazard not to protect personal. Evan at that purpose, it's hardly perfect. And it certainly creates other risks.
It provides only a very small realm of protection (the + wire from the array combiner box to the charge controller--and possibly from the + battery bus--to safety ground.

It does not protect against + to - faults (main PV breaker will probably never trip as Isc << Ibreaker by design).

And, if you have a + to earth ground short, the only breaker/fuse that trips is the 1-5 amp safety to - bus connection--Rest of system is still energized, and if there is a second - return lead to safety/earth ground short, the battery bank could very easily cause an over current/fire hazard in the -/return wiring paths.

QuoteWhen GFP is is implemented with a fuse or breaker, a negative to ground fault is very likely to be undetected. That leaves the system unprotected against positive to ground faults. Even if the GFP fuse of breaker operates, the fault current will not be interrupted.
Yep... A failure elsewhere (such as a -/return to earth fault) is not detected and can bypass the DC GFP protection. That sort of protective circuit design would never be allowed in any standard AC wiring system.

QuoteWhen the GFP is activated, the normally grounded PV wire may be operating at -17 to -600 volts. That's why 690.5(C) requires a warning label at the GFP device that states that when then GFP is activated there is an electrical shock hazard - normally grounded conductors may be ungrounded and energized. A person that isn't familiar with these risks, shouldn't be trying to find a ground fault in a PV array!
So, with or without DC GFP, we have the same problem and a $1 label can fix it...

QuoteBill, your concern about the the negative battery bus being operating below ground potential is incorrect. When the GFP is activated, the battery is still grounded to earth through a high impedance. Voltage measurements to ground would be completely normal.

No, in general, there is no intentional connection between return and safety ground... In the case of the Classic, there is a high impedance connection through the PTC--But it is by definition unable to "hard ground" the Isc/Imp energy from the Solar panel.

And depending on where the + Battery bus to safety ground fault occurs (on the + bus at the battery bank with no fuses, or the + lead of a 2 amp LED lamp), the -/return bus is now energized, potentially with more current available than the "DC return Branch Wiring" is capable of carrying.

Again, there are some assumptions in my argument about the negative/return connection from the Varray to Vbattery through the charge controller--And NEC is not designed for protection against double faults (i.e., you do not assume both a circuit breaker fails shorted and a charge controller fails shorted at the same time)--However, any single and subsequent faults should fail in a safe manner. Energyizing a current path (- return branch circuitry) with more current than it is capable of handling (when properly installed per code) is (in my humble opinion) a violation of the single and subsequent faults requirement.

In this case, DC GFP requirement, fails the design guidelines (as I understand them) for the NEC.

-Bill

keyturbocars

Thanks for discussing this guys.  I have so little background in this area, that it is very educational for me to be reading this.

Edward

Westbranch

Hi bill is it possible for you to post a quick sketch of example #1. 
I am not conversant with the difference between 'Safety' and 'Earth' Ground as well as several other terms you have used. have to do some reading ...
thanks
Eric
KID FW1811 560W >C&D 24V 900Ah AGM
CL150 29032 FW V.2126-NW2097-GP2133 175A E-Panel WBjr, 3Px4s 140W > 24V 900Ah AGM,
2 Cisco WRT54GL i/c DD-WRT Rtr, NetGr DS104Hub
Cotek ST1500 Inv  want a 24V  ROSIE Inverter
OmniCharge3024  Eu1/2/3000iGens
West Chilcotin 1680+W to come

BB.

For my discussion--Earth=Safety Ground... Basically, I am typing about +Varray, +Vbattery, Safety/Earth Ground, and assuming that -Varray=-Vbattery.

And the functional difference between a "Normal" system ground (-Varray/-Vbattery tied to Earth/Safety Ground in one location) vs the DC GFP which puts a 1-5 amp fuse between -Vbatt/-Varray and Earth/Safety ground.

The second difference is there a switch/breaker OR an Inhibit (DC charge Controller/GT Inverter) that prevents +Varray current flowing to +Vbattery. I submit that this second difference (inhibiting current flow through the "converter") does not increase safety--but instead, is simply the equivalent of a flashing light/buzzer that you cannot ignore if you have a "Ground Fault" between V+ (array or battery?) and safety ground (cannot ignore because the "converter" turns off).

The Switch / Inhibit does not prevent excessive (i.e., hazardous) current flow UNLESS the system is improperly designed / installed in the first place (i.e., no fuses/breakers in battery bus, too small gauge of wiring, etc.). And in many cases, an improperly designed/installed system would have had problems anyways (too small of breaker would trip, too small of wiring would overheat with Imp anyway, etc.).

A Ground Fault will not cause "excess current" to flow anywhere in a well designed/installed system. About the only thing a DC GFP will reduce the hazard of is the point of the short circuit and limiting current to 1-5 amps for the initial fault (i.e., short may not get as hot as it pop ground fuse/breaker).

But we could just as easy have a +/- fault (wiring failure from rats chewing, or a broken panel with internal shorts, shorted diodes, etc.) and the DC GFP will have no chance of detecting/interrupting those types of failures. Those are still addressed by proper design/wire size/combiner box protection.

I will see what I can do about sketching something up (my hand drawing is terrible--an't no artist--And all my professional drawing stuff is back on my old jobs).

-Bill

BB.

#11
One other quick way to look at the DC GFP setup... Imagine you designed a Negative Ground system and put fuses/breakers on all positive wiring. No fuses/breakers on negative wiring, and one positive solid connection between the negative bus and earth ground (including grounding to water pipe, fixtures, wiring boxes, conduit).

If there is a fault from anywhere in the Positive circuit path, positive wire will be protected by an appropriate fuse/breaker which will be tripped by the excessive current flow.

If there is any negative/return wire to earth ground, virtually nothing will happen as both are very near the same voltage. (it is possible with ground loops to overheat return wiring, but for now, NEC does not address over current from ground loops other than recommending/requiring single point grounding).

Now, you replace that one Earth Ground with a 1-5 amp fuse. And now earth ground, randomly, anywhere in the Positive Wiring Path (PV wiring or Battery wiring). Pops the 1-5 amp fuse, and stops power flowing from solar panels to/through charge controller... Now you have a positive earth grounded system with fuses in the earthed positive paths.

No current limits on Negative Grounds. Items that were assumed to be "touch safe" (like lamp sockets) are now powered by battery or PV Array voltage/power. And the current limit available anywhere on the negative/return wiring is dependent on where, specifically, the Positive to Earth Fault was made.

-Bill

boB

Quote from: BB. on March 18, 2011, 10:13:35 PMand one positive connection between the negative bus and earth ground


Bill, I may be having a brain fart here, but I think I'm reading that part wrong ??

boB
K7IQ 🌛  He/She/Me

BB.

boB,

I intended Positive Connection as meaning bolted up / solid connection (on purpose) as opposed to faulted connections later in the discussions...

Probably too many uses of the word "positive" in one paragraph/post on my part.  :-[

-Bill

boB


Thanks Bill  !   I DID try reading it that way once.

I'm not positive which ones' negative.

OK, now  to go read it again...

boB
K7IQ 🌛  He/She/Me