fusing pv panels

icarus
icarus Solar Expert Posts: 5,436 ✭✭✭✭
I apologize if this is posted elswhere, but my search has revealed nothing.

I am rebuilding my system, and an interesting question has com into my mind. In the current system, the wire from the panels to the controler are fused at the controler. My thought is this protects the wire from the amperage from the battery bank just fine. The question is, since the panels put out enough amperage to cause a fire, should the wire be fused at the panel end as well?

Since there are two sources of current, one on each end of the wire there should be two fuses to protect the wire. Am I right?

Once again, thank your for all the expertise out there,

Icarus

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    Well, there is what Code requires, vs what should be needed for safety. Sometimes they are the same, sometimes they are not.

    From a design point of view (not NEC), the solar panels are a limited power source by physics. Even if you short the wires at either end, the current supplied by the panels will never exceed the capacity of the wire to carry current.

    And, if you look at the MX 60 manual, you will see that they have 80 amp breakers at both the PV panel side and the battery bank side. In my humble opinion, the only current source that needs to be protected is from the battery side (thousands to tens of thousands of amps are available from even a smallish battery bank). Once you have a breaker there, the PV breaker is redundant. However, if you need a switch to turn of the PV to MX-60 for servicing, you can use either a breaker or a properly rated switch (again IMHO) to achieve the desired function (for example, my Xantrex GT 3.0 grid tied inverter uses switches instead of breakers for the PV and 240 VAC connections--the only fuse/breaker is the standard double pole breaker in the panel for protecting the branch circuit from the utility mains--excluding the 1 amp fuse used for PV ground fault detection).

    But, there is one situation where you do have breakers/fuses near the solar panels--those are at the point where multiple parallel strings of solar panels are joined together. In this case, you are protecting the wiring/solar panels themselves from all of the rest of the pooled current/energy available from multiple parallel strings of solar panels (panel maximum series fuse requirement).

    Again, I am not saying to ignore the code (and I am not even sure what code requires as I have not researched it), just my personal opinion from a solar system design point of view.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: fusing pv panels

    assuming you are to nec rules already, but you are wondering of further protection, you can do it. if you do it at the pvs keep in mind that if the fuse pops that you will have to access it day or night summer or winter.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    One of my desires when designing was to try and make decisions that helped keep cost and size down, and increase reliability...

    In almost every case, adding a fuse or breaker to a system increases costs (both in more parts and labor to install), increases size, and decreases reliability (fuses are among the least reliable items in a computer today, they are subject to thermal stresses from power cycling and run "hot" compared to normal wiring--let alone the issues of installation errors and other points of failure when converted from a straight run of wire to a box/fuse-breaker/terminals/etc.).

    Many people believe that fuses/breakers are there to product devices--and the answer is no--they are there to limit the chances of fire (and other safety issues).

    A device with a blown power supply rectifier or a soldered in blown fuse has still failed--and it will cost just about as much to repair the device/power supply in either case (from a corporate point of view).

    Fuses and breakers when used as appropriate are good things, but when added carelessly, can cause issues (I have run into several old homes that had fuses/breakers in the neutral legs, salvaged breaker panels that were removed from homes by recall only to be installed in another home later, etc.).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: fusing pv panels

    I care not one whit about NEC or CSA, as I am on an island 200kms from the nearest town, and 35kms from the nearest road! Since I am my own fire department, (as well as public works, water sewer dept, and any other dept!) I care greatly about safety. I understand what both of you are saying about reliability. What I'm not clear on though, is,,,is there enough potential energy from the panels (~150 watts, ~15 amps 12vdc) that if a mouse were to chew through the wire and it were to find the other wire in the cable, or some other source of ground for example, could it cause a fire? I understand the potential from the battery side, but how 'bout the panels. My intuition is that I am creating a solution for a problem that doesn't exist. If convention calls for fusing, I'll fuse, if not, then not.

    Icarus
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    If you want to trust some guy you never met, 4,000 miles away, on the Internet--No, I would not put any fuses between the charge controller and the PV feed (assuming you have appropriate sized fuse/circuit breaker on the battery side of the charge controller). You may wish to install an appropriate PV disconnect switch if you need to de-power the unit for maintenance. (or just throw a tarp across the array in the rare case you need to kill the PV power for service).

    The only thing is I don't remember is the brand/model of solar panels you have, their configuration (series/parallel), and their series fuse requirement...

    If the 15 amps is from two parallel connected panels (or strings), that means about 7.5 amps from each leg (really should be Ishort-circuit--call it 8 amps) and if the series fuse requirement is (for example) 10 amps--then you would not even need the series fuses...

    I would not install any fuses in the PV combiner box unless it is required for series protection...

    Using the fuses for series protection is probably not a bad idea--for example; 150 watts of power (or whatever is in excess of the fuse rating is) going into a failed panel could cause flaming/dripping backing from the shorted panel (if all conditions of the failure were ideal)--dripping on the roof or vegetation--While a rare (or possibly never has happened before in real life failure)... It is something that your local Fire Chief/Inspector would ask about. ;)

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: fusing pv panels

    Bill makes some very interesting interesting points. In some battery-less utility-interactive systems, there is an NEC exemption [690.9(A)] from the requirement for fuses/CB's between the array and the inverter.

    The requirement for fuses/circuit breakers between the PV array and the charge controller is apparently out of concern for the possibility of the battery back-feeding through the controller to the array. The fuse/circuit breaker between the battery bank and the controller may not be small enough to protect the PV array and/or its wiring.

    See Wiles' "...Suggested Practices, Appendix J", for further discussion.

    Interesting issue!
    Jim / crewzer
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: fusing pv panels

    "is there enough potential energy from the panels (~150 watts, ~15 amps 12vdc) that if a mouse were to chew through the wire and it were to find the other wire in the cable, or some other source of ground for example, could it cause a fire?"

    anytime you deal with electricity there is that possibility of fire. 12v can spark a fire, but is less likely to than 120v as the spark or arc will be less intense. in the case of shorts allowed to occur when no fuse is used then the wire itself will heat, but most of the time in solar pv we have the wiring sufficiently big enough(to keep voltage drops low) that it takes that excess heat well. in short, you will have to be the judge on whether to fuse at the pvs or not. if you do and the fuse blows you still have to gain access to it no matter the time of day or year so its placement and replacement is up to you if you elect to do it.
    now knowing that i will say that if it were me i would not fuse at the pvs.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    From reviewing "Appendix J", I can see one of the issues... The rating of the breaker/fuse between the battery and the controller should not exceed the series rating of any of the individual solar panel strings. If it does, either the protective device between the battery and the controller is too big, or individual series fuses will be required for each PV panel string to protect against a charge controller short (between battery terminals and PV terminals on the controller).

    Basically, if you have a "simple controller" (PWM or other non-MPPT type), it is pretty easy to assume that the control element between PV panels and the battery (the "on/off" switch for charge control) could fail "on"... If there were any shorts in the panels, then "dangerous" currents could flow from the battery, through the controller, to the panel(s).

    If the controller is an isolated MPPT type--normally power can only flow when the circuit is in operation (transformer coupled). Once the circuit fails (transistor opens or shorts), no power can be transferred and the panels are safe...

    However, there is no NEC/NRTL standard for typical DC to DC charge controllers that discuss isolation or if they have "fault isolation"... For Grid Tied systems, usually an Isolation transformer would be required to prevent Utility AC power from "lighting up" the DC side (PV panels)--which would provide the required "back current" isolation (my Xantrex unit, ~2 years old, came with a supplemental letter from UL discussing this exact point and why back feed protection would not be required for the solar panels as back feed, for NEC purposes, was defined to be ZERO amps).

    One could use design rules for AC systems (double isolated transformers, xx mm of spacing, ground screens, NTRL controlled materials and construction, etc.) in a DC system--but that probably adds cost and the "market" may not see the justification for adding this level of complexity (and it may not make any sense in a >200-400 watt system where MPPT controllers would
    be deployed).

    -Bill

    I should add that "dead" shorts are, many times, less of a problem. They blow (clear) fuses and breakers quickly (milliseconds to seconds). Also the voltage drop is low at the site of the short--very little heat can be generated.

    The bigger problems can be when you get a resistive short (say the copper fingers in a solar panel). They don't have low enough resistance to clear a fuse/breaker and can start generating a local hot spot (~150 watts of heat) and causing local melting/ignition source. Hence why NRTL's may also want to control the material used in construction to ensure that they are not "too flammable".
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • System2
    System2 Posts: 6,290 admin
    Re: fusing pv panels
    BB. wrote: »
    The bigger problems can be when you get a resistive short (say the copper fingers in a solar panel). They don't have low enough resistance to clear a fuse/breaker and can start generating a local hot spot (~150 watts of heat) and causing local melting/ignition source. .

    =============================================

    Just wondering if there is any history of such things happening with PV panels. I know that if anything can happen, it will happen, but does anyone know it it has, or is happening?
    Thanks
    Wayne
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    Kind of "funny" reading through a bit of the download that Jim recommended and ran across one of the problems I had with NEC 20 years ago (Annex K)...

    I remember the UL/NEC requirement of no special tools required to terminate wiring--and because I was designing a DC powered Voice Mail system that required several kWatts of DC power--I needed to terminate the flexible fine strand cabling with special terminal ends and crimp tools (required by the telephone companies)... Because fine strand cable (like welding cable) was not safe to use in normal UL/NEC approved camping terminals (see examples in Jim's download of nice/messy failures). Was a real pain.

    Oh--the "good old days"...

    -Bill

    PS: I intended to mention this from the download, from Page J-2:
    Many installers of 12-, 24-, and 48-volt PV systems ignore the module OCPD requirement and connect modules/strings in parallel. Can it be done and how? Dave King at Sandia National Laboratories and I have smoked a few modules and determined that the module OCPD requirement is valid.
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • GreenerPower
    GreenerPower Solar Expert Posts: 264 ✭✭✭✭✭✭
    Re: fusing pv panels
    icarus wrote:
    ... What I'm not clear on though, is,,,is there enough potential energy from the panels (~150 watts, ~15 amps 12vdc) that if a mouse were to chew through the wire and it were to find the other wire in the cable, or some other source of ground for example, could it cause a fire?
    The nature of PV cell would limit the current in case of a short. The PV wires are sized for the Isc adjusted for temp. and other factors for worst case conditions. This wire size should be able to sustain the Isc in case of a short. And in case of a short, V~=0, so the wattage ~=0 on the wire. With DC current, a "touching short" would tend to fuse the short contact point compared to AC where you could have a low current "simmering bad connection" to cause a fire. So, IMO PV wires with proper size and "lug-tight" connections at both ends should be sufficient.

    GP
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    Since I have not tried to smoke a solar panel--this is all guess work...

    One issue is that solar panels are only designed/"listed" to be "safe" with limited current available. Isc of one panel is not the issue, it is Imp/Isc of multiple panels in parallel. 5 amps is not expected to be a problem--but (for example) 48-60 amps (MX-60 maximum listed input current--3,200 watts at 48 VDC battery bank voltage) plus up to 140-150 VDC maximum working voltage is enough to case problems with the leads/internal grid/etc. is more than enough to cause "arc welding" type currents/arcs/fires. These currents easily exceed lug/distribution wiring capacities in large systems if series protection fuses (for each panel string or string sets) are not used.

    I don't know about solar panels, but, for example, there are other possible sources of failures--On PC boards, there are copper dendrite growths (copper crystal fingers that grow between opposite voltages).

    Other things I can think of happening: panels that are broken by hail/vandals/accidents. Internal cracks/opens that can arc at higher voltages used in Grid Tied applications (up to 600 volts); other failures (charge controller shorts between PV and Battery Bank; accidental miss-wiring...

    I have seen enough smoked boards/fires on PC Boards from shorts in PCBs, components, and wiring exposed to just 5 volts (at 10's-100's of amps). In those cases, the materials used were flame rated and inside of properly constructed metal cabinets and frames to reduce the ability of fire to spread beyond the point of failure. One famous call to our tech support department was when a customer called asking what to do when one of our systems was outputting tremendous amounts of smoke (answer back was--unplug/turn off the system's power).

    Solar Panels are not installed inside of metal cabinets--but are instead mounted on potentially flammable roofs and/or over vegetation. Limiting the power available (via series protection devices) makes very good sense.

    One of my systems had to go through Telephone company certification--and in one set of tests they actually placed small trays of lit alcohol in three or four places inside of our running equipment to see what happens, how much smoke is generated, and if the resultant fire can be putout with 1 or 2 standard fire extinguishers. Obviously, stuff would catch fire--but it did not spread outside the cabinet and our material selections/construction ensured that the resulting fires were small and was easy to extinguish (problem was that this was our first fully populated prototype system ($1,000,000 retail worth of equipment) and our engineer forgot to exchange the working boards/modules with a set of old prototypes for the system burn test :cry:).

    The NEC/Listing/Testing/Design process is not perfect--But it is built around the idea that failures will happen and that using (hopefully) well thought out rules and practices, that any failures will be limited in their scope.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: fusing pv panels

    Bill wrote:
    Isc of one panel is not the issue, it is Imp/Isc of multiple panels in parallel.
    Right you are. I believe it's OK to parallel two strings of PV modules, but three or more require separate fuses/CB's just for the reason you suggest.

    Regards,
    Jim / crewzer
  • icarus
    icarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: fusing pv panels

    Hey guys,,,
    I've been away for a few days, and look and the replies. Thanks for all the information, although in my case it is verging on the overload. I think my answer lies in fusing only the battery end. I think that the potential for fire from the panels is so remote as to be adabsurdum (sp?). My woodstoves and thier questionabe installations pose a much greater risk.

    Once again, thanks for the help.

    Icarus
  • bfitzgerald
    bfitzgerald Registered Users Posts: 26 ✭✭
    Re: fusing pv panels

    This is a pretty old thread but a good one. I hope someone can answer these questions....

    I have 12 - 250w panels that I intend to install in 4 strings of 3 panels. Isc is 8.87a. What is the best way to fuse the parallel strings, and what size/type of fuse would be best. I was thinking maybe a Midnite Solar baby box with 4 15 amp dc breakers? Could this work as a combiner and array ocp at the same time?

    Thanks
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    Yep, that is what those combiner boxes are for... You really do not need another Over Current Protection device. Normally the wire run from the combiner to the solar charge controller input is heavy enough to carry Isc-array--Or frequently we use heavier than minimum gauge wire to keep overall wiring losses low.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bfitzgerald
    bfitzgerald Registered Users Posts: 26 ✭✭
    Re: fusing pv panels

    Thanks for your response. Would 15 amp breakers suffice? I believe that I am protecting the panels from a panel short circuit, not the wire.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: fusing pv panels

    15 amps should be fine. You are protecting both the panels and the wiring to the panels from over current.

    Modern solar panel specifications/data sheets should have a series protection rating called out in the spec.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Re: fusing pv panels

    Hi Brian,

    Just to make certain that you know that the Baby Box, is very very small. Especially with four breakers installed. Am certain that you know that it is not rated for use outside, in the elements.

    Using the BabyBox as a combiner for four strings on a 3 Kw STC system may disappoint. Very little room for a BusBar, and dressing the probable largish output conductor may also be a bit difficult. Believe that there is only a single Ground terminal -- no insulated BusBar for the PV Neg leads. You will probably need/want to use a MN Positive Busbar which is included with the Combiner, but not with the Baby Box, and so on.

    Mn Combiner:
    http://www.solar-electric.com/mnpv6.html

    Not to pick at your idea too much, just FYI
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
  • bfitzgerald
    bfitzgerald Registered Users Posts: 26 ✭✭
    Re: fusing pv panels

    Thanks for the advice on the breaker sizes and pointing out the limitations of the baby box. I'll definitely rethink that one.

    Thanks so much