I think I recall seeing a thread somewhere on here discussing using a Battery Negative to Ground bond (in the epanel or elsewhere) rather than enabling the Classic's built in GFI. I can't find the thread, but I was wondering what the current thinking is regarding this setup.
I'm not bound by the NEC, but I would like my setup to be as safe as possible. So, what are the group's thoughts?
Thanks
Bill (at the NAWS forum) has written extensively about this problem... some of his best writing is on this midnite forum.
The NEC requires GFP on all rooftop arrays. The purpose of GFP is to disconnect the array and prevent fires in the event of a parallel ground fault in the array wiring. The problem is that when the GFP trips, there is no longer any bond between ground and DC negative. This is very dangerous and can lead to very high currents in the unfused neutral wiring and/or high voltages between neutral and ground.
The best way to prevent arc faults (series and parallel) in the pv array is to use an arc fault combiner. The arc fault protection in the classic will respond to series arc faults, but not parallel arc faults.
--vtMaps
VTmaps said it well. I agree Arc Fault in the combiner is the safest way period. Ground fault does nothing for fire suppression as proven by multiple high visibility fires.
Now without my MidNite hat on I will tell you what I do on systems that have to be inspected. I use an appropriate sized jumper in the Epanel between Earth ground and Battery Negative and then I install the GFP jumper on the Classic.
Realize Midnite can not promote the non use of the GFP systems required by code but we can express our concern over some inherent safety issues we see presented by them.
Here is a great article done by Bill Baldwin that really points out some fatal flaws http://www.midnitesolar.com/pdfs/DC-GFP-Draft3-5.pdf (http://www.midnitesolar.com/pdfs/DC-GFP-Draft3-5.pdf)
Gents,
Thanks for your thoughts. I do recall reading the thread and Bill's pdf, but I couldn't find them again for rereading.
QuoteI use an appropriate sized jumper in the Epanel between Earth ground and Battery Negative and then I install the GFP jumper on the Classic.
I'm a tad confused on this point. If you install the jumper, then isn't a second bond created? I take it that you don't enable GFP in the Classic.
Sure I enable the GFP but it will never detect a fault as the jumper in the E-Panel over rides the GFP device
Ryan
Thanks Ryan
Hi All,
I can't believe what I read here!
In a previous post, I was talking about a SPD issue (PV SPD led very dim) after a storm (three panels lost). After reading this post, I've made a neg/ground bond and I've seen that my PV SPD was working again. If I understand correctly, with my array on roof and GFP enable in the classic (no bond), there was a risk of chock/injuries or fire after this strike because of bad Nec rules. Obviously, it was a GFP problem (no fault) and we've had several strikes after the closest one which had caused troubles to my system. I'm feeling very lucky and thanks to Midnite, SPD had saved my system.
Erik
A few more questions about SPD and Classic GFP:
From what I saw this morning, a PV SPD has dim led because of defective neg/ground bond, am I right?
If a SPD is defective without neg/ground bond (GFP) a Classic is not protected against strike overvoltage?
A system with defective GFP and no neg/ground bond could also be dangerous for end users?
I understand that you can't go against the Nec, but why don't you warn about these problems before? Forum or manual?
I've made several installations following Nec, thinking that my clients were safe and now I realize that they could be in trouble...
Oops! Where could I send a bill to... Nec ????
Erik
Quote from: SolarMusher on August 28, 2013, 04:08:47 PM
If a SPD is defective without neg/ground bond (GFP) a Classic is not protected against strike overvoltage?
A system with defective GFP and no neg/ground bond could also be dangerous for end users?
I'm not sure I understand what you mean by "defective GFP". A GFP is a bond between ground and negative. When the GFP is tripped it breaks the bond. There's nothing wrong or dangerous about a GFP ... until it trips.
--vtMaps
Hi Vt,
I mean by defective that after this strike, SPD led was very dim, I think that it was dim because there was no more bond (blown fuse ?) between neg/ground inside the classic but the classic GFP didn't show a fault. After this post, this morning I've wired a jumper across neg/ground in the E-panel and the SPD was working fine again.
Erik