Manufacturers’ requirements for equipment locations

A statement found early in the NEC^® — Article 110.3(B), to be exact — mandates that all equipment must be installed and used in accordance with any instructions for that equipment. Consequently, to be compliant with the Code, you must install all electrical equipment according to the various manufacturers’ instructions. So if the installation manual says you can install the equipment only on even-numbered days and you install it on an odd-numbered day, you’re in violation of the Code. Granted, that’s a silly example, but it illustrates the importance of knowing a manufacturer’s requirements. I’ve seen installations that failed inspection because the inverters weren’t spaced properly according to manufacturer instructions; in each case, the inspector demanded that the misplaced inverters be moved.

Avoid placing any electrical equipment in direct sunlight (with the exception of the PV modules, of course; I explain how to install the modules in Chapter 16). For the inverters, this is good practice because an inverter’s ability to produce power is reduced as it grows hotter. Equipment such as overcurrent protection devices (OCPDs) are also affected by increased temperatures (you may need to derate them per the manufacturer’s instructions if the temperatures get too high). Other equipment, such as disconnects and combiner boxes (boxes that place all the PV strings in parallel), may not normally be negatively affected by high temperatures, but when in direct sunlight for multiple hours each day, the labels you use on them may degrade with time and be difficult to read. For conduit runs, you may not be able to avoid locations in direct sunlight. If this is the case, make sure the conduit is properly rated for outdoor installation by referring to the manufacturer’s spec sheet.

Any electrical component you install with electrical terminations inside of it can be called an enclosure (some examples include disconnects, combiner boxes, and inverters). Make sure you know the National Electrical Manufacturers Association (NEMA) enclosure ratings before installing any electrical equipment in a PV system because the NEMA rating of the enclosures you install affects where the equipment can be installed. The most commonly seen NEMA enclosure ratings for PV equipment are

• NEMA 1, indoor installations only: This enclosure offers no protection from dust or water entry. (Nearly every battery-based inverter is rated NEMA 1 — in fact, I can’t think of any exceptions.)

• NEMA 3R, outdoor installation, typically on a vertical surface only: This enclosure type offers protection from falling rain. (Most grid-direct inverters — but not all — have at least a NEMA 3R rating.)

  The reason I specify vertical surfaces for NEMA 3R enclosures is due to the listing process. When manufacturers have their equipment tested, the standard for NEMA 3R is that the equipment is mounted on a vertical wall. Therefore, unless specifically stated otherwise, you can install enclosures rated at NEMA 3R only in the vertical position. It can be tempting to install a NEMA 3R enclosure on a roof at the same angle as the roof (I’ve even seen this done), but the problem is that the enclosure wasn’t designed for that application. If installed on the same angle as the roof, it may not be able to properly protect the electrical connections inside from rain and ice.

• NEMA 4, outdoor installations on horizontal or vertical surfaces: These enclosures can withstand direct spray from a hose and splashing water.

Locations for disconnecting means

When I refer to disconnecting means, I mean that, in the most generic sense, you need to install disconnects such that when you go to work on an inverter, you can safely disconnect both the AC and DC sides of the inverter from all sources of power with approved methods at the inverter location.

The NEC^® requires the PV disconnecting means be installed in a readily accessible location, either on the outside of the building or after it enters the building, at the nearest point of entry. But placing the disconnecting means properly can be a difficult task within most PV systems because many PV output circuits (which run from the combiner box to the DC disconnect) enter a building up at the roof location, and no “readily accessible” location exists up there for placing a disconnect either inside or outside of the building. Thankfully, an exception applies to this Code section that refers you to Article 690.31(E). It allows you to install the PV source circuit or PV output circuit inside a building without first installing a disconnect so long as the conductors (wires) are inside metallic conduit until they reach the disconnect.

Most grid-direct inverter manufacturers now include a disconnect switch that can be integrated into their inverters, allowing you to disconnect the DC power source right at the inverter (and sometimes even the AC power source with the same switch). These disconnects are a separate piece that mates to the inverter during installation. The inverter can then be removed safely, leaving the disconnect on the wall and allowing for the installation of a new inverter. (See Chapter 10 for more on disconnects integrated into inverters.)

Many inverters that include an integrated disconnect make provisions for the installer to also use the disconnect as a combiner box. This means you can bring down multiple PV source circuits from the roof to the combiner/disconnect and place them all in parallel at the inverter location. In fact, the 2011 version of the NEC^® will require a disconnecting means at the point of PV source circuits placed in parallel. Using the integrated combiner/disconnects from the manufacturer can help you easily meet this requirement.

For those inverters that don’t have an integrated disconnect, you need to install AC and DC disconnects at the inverter location. This process involves mounting at least one disconnect next to the inverter to disconnect the PV source circuits. If the inverter is installed at the main distribution panel (MDP), the circuit breaker used to make the connection to the utility can serve as the AC disconnecting means. If, however, the inverter is mounted at a location other than the MDP, you must install an additional AC disconnect at the inverter location. If the inverter and AC disconnect happen to be located at the utility meter location, you may satisfy both the NEC^®’s and the utility’s requirements for disconnects with the same disconnect (you find out more details about utility disconnects later in this chapter).

The NEC^® also says that a disconnect “shall not be required at the PV array location,” which means that for ground- and pole-mounted arrays (see Chapter 16), you can run PV source circuits without installing a disconnect at the array. However, most PV installers consider it a best practice to install a disconnect there because doing so provides you with an easy place to disconnect the PV circuits. (Besides, when the 2011 version of the NEC^® is available, you’ll need a disconnect if you combine the circuits at the array.)

Combiner boxes and junction boxes and wiring, oh my!

Making the transition from module conductors in free air to the conductors you run through the client’s building (or underground) to the disconnecting means has to happen in some sort of enclosure, either a combiner box or a junction box. (A junction box is an enclosure that looks very similar to a combiner box on the outside. Inside, it uses terminals to transition the PV source circuit wires [USE-2 or PV wire; see Chapter 10] to the building wiring [THWN-2]. These terminals don’t place any conductors in parallel though, so you have the same number of conductors entering and leaving the junction box.) This enclosure typically lives at the array location, thereby reducing the length the exposed conductors have to run.

Until your client’s jurisdiction is using the 2011 version of the NEC^®, you don’t have to have a disconnecting means adjacent to a combiner box. So if you want, you can install a combiner box at the array location and bring a single PV output circuit down to the inverter. Generally, however, running the individual strings to the inverter and using a combiner box at the inverter location is considered a safer method. If that’s the method you choose to use, then you must still incorporate a junction box at the array location to transition the outdoor wiring to indoor wiring that you can run in conduit.

Make sure the junction or combiner box you choose has the proper NEMA ratings for the environment you install it in (see the earlier section titled “Manufacturers’ requirements for equipment locations” for details). Also, you may need to consider the heat effects on the fuses if you decide to mount a combiner box on a rooftop or in a location where the temperatures will be greater than the manufacturers’ ratings (see Chapter 13 for more on this).

Seeing red (and green and white): Color-coding

When it comes to wiring PV systems, one of the most common mistakes I’ve seen is someone mixing up the color of the conductors used. The NEC^® has a very specific set of requirements for color-coding based on the function of the conductor in the circuit.

The NEC^® requires the following color-coding for wiring in PV systems:

• Equipment-grounding conductors (EGC) must be green, green with yellow stripes, or bare wire.
• Grounded (or neutral) current-carrying conductors must be white or gray.
• Ungrounded (or hot) current-carrying conductors can be any color other than those mentioned for equipment-grounding and grounded conductors. However, the most common colors are black and red.

To help you remember what colors of conductor go where, think of the 120/240 VAC wiring system in your home. When you look at the wires coming into the outlets around your house, you have one black wire, one white wire, and one bare copper wire (which is sometimes green instead). The black wire is the one that’s connected to the circuit breaker in the MDP or subpanel, the white wire is connected to the neutral busbar, and the bare (or green) wire is connected to the ground busbar. (In that same MDP, the neutral busbar is connected directly to the ground busbar.)

If you look at a 240 VAC load, like your electric clothes dryer, you see a red wire in there too. In this case, the red wire and the black wire are connected to a two-pole circuit breaker. In this wiring scheme, the black and red wires are known as the ungrounded (or hot) current-carrying conductors, and the white wires are known as the grounded (or neutral) current-carrying conductors. The bare or green wires are the equipment-grounding conductors; they don’t carry current in properly installed and operating circuits.

So what does all of this mean to your PV system? As you well know, a PV module has two conductors on the back of it. As with all DC circuits, one side is positive (marked with a plus sign), and the other is negative (marked with a minus sign). Because DC electricity flows in a single direction (see Chapter 3 for more on DC electricity), the designation of the positive and negative sides, or polarity, is important to start with. Although all the PV module conductors are marked positive and negative, the best way to define the different parts of the PV circuits really isn’t with the terms positive, negative, and ground. This terminology isn’t incorrect, but you can do better. Most people associate red wires with the term positive and black wires with the term negative. When you install a PV system, you’re going to have a positive and a negative conductor, but one of those conductors will be connected to ground (just like the neutral conductor is in the MDP at your house).

If you can get yourself into the habit now, remove the words positive, negative, and ground from your vocabulary and start referring to them as the grounded current-carrying conductors, ungrounded current-carrying conductors, and equipment-grounding conductors. This way you can easily follow the color-coding required by the NEC^®.

• For a PV system that has the negative connected to ground across the ground fault protection (GFP) device, the negative becomes the grounded current-carrying conductor (just like the neutral in the AC system) and is white or gray. (Chapter 10 has more info on ground fault protection.)
• The positive is the ungrounded current-carrying conductor (just like the hot conductor on the AC side) and can be either red or black. I suggest using red rather than black because even though you may have banished the word positive and its association with red wiring in DC systems from your consciousness, you may be the only person in the jurisdiction who has done so. If someone comes up behind you, opens up a box, and sees a red, white, and bare (or green) wire, that person (assuming he knows the wiring is for a DC system) will probably associate the red with positive. He may still have to think about the white wire, but you’ve done your best to help him out.
• As in all electrical systems, the equipment-grounding conductors are green, green with yellow, or bare copper. This standard is consistent with all the wiring on the AC side.

Managing wires on PV modules

The quick-connect plugs installed on the backs of PV modules are a sort of double-edged sword. On one hand, they greatly reduce the amount of time you have to spend making module interconnections (the electrical connection from module to module). On the other hand, they introduce an issue that wasn’t present before: the need for wire management on the backs of the modules. Before the quick-connects, the wiring was most commonly done by connecting conduit from one module to the next. Although this task was time consuming, it automatically protected and managed the wires. Nowadays, most modules can’t accept conduit because the electrical boxes integrated onto the modules have no way to accept conduit connections. Consequently, you have to find other ways to keep the conductors safe and secure.

The point of wire management is to provide a safe and reliable method of keeping the conductors in place for the life of the PV system. When installed correctly, that system should last more than 25 years, which means you need to protect the conductors from damage from all the elements, especially damage that may be caused if the conductors are blowing around in the wind, in contact with the roof surface, or being pulled down as snow and ice melt along the backs of the modules.

You can use a couple different methods to secure the exposed conductors running along the backs of the modules:

• Specially designed clips: Multiple companies manufacture specially designed clips that you attach to the PV module and then snap the conductors into. You can buy these cable clips in different sizes to accommodate different wire types (such as USE-2 or PV wire; see Chapter 10 for more on these conductors).
• Plastic cable ties: Also called zip ties, plastic cable ties are very easy to use and commonly available, but I don’t like to rely on them. You can buy plastic cable ties that are UV-resistant (usually the black ones), but I’ve seen these so-called UV-resistant ties fail in just a few years of being out in the sun. Even though most plastic cable ties aren’t in direct sunlight, I just don’t trust them to last more than 25 years. I suggest purchasing stainless steel cable ties with nylon coating instead.

A combination of cable ties and cable clips is generally your best bet for long-term reliability.

Protecting wires with conduit

Chapter 3 of the NEC^®, “Wiring Methods and Materials,” specifies the requirements for properly protecting conductors with conduit. In the earlier “Locations for disconnecting means” section, I note the need for metallic conduit for some roof-mounted PV arrays. Another common approach is to use PVC conduit for underground and exposed outdoor locations. (See Chapter 10 for an overview of the common conduit types and acceptable locations; refer to Chapter 13 for details on conduit sizing.)

Regardless of the type of conduit used or the type of circuit (AC or DC), you and a partner can pull conductors through conduit the same way. When pulling the conductors, use fish tape (a stiff, yet flexible, metal “rope”) and push it from one end of the conduit to the opposite end. After it appears on the other end, tape your conductors to it and pull the fish tape and conductors through the conduit.

I also suggest pulling a nylon string through the conduit along with the conductors. If you ever need to pull any new conductors through the conduit, the nylon string serves as the fish tape, saving you the hassle of trying to push more fish tape though a conduit that already has conductors inside it.

If you want extra help getting the conductors into your conduit, you can buy lubricant, affectionately called slime, to help slide the conductors through the conduit. Slime is cheap and helpful anytime you have to pull the conductors.

Equipment grounding

The point of equipment grounding is to make sure all the electrically conductive materials are kept at the same voltage potential as ground. So if a conductor accidentally touches any equipment (a disconnect or PV module, for example), the current has a low resistance path to ground, allowing the safety equipment to activate and reducing (but not eliminating — never forget that) shock hazards to anyone who may touch that piece of equipment. The equipment-grounding conductor (EGC) runs right alongside the other conductors in the PV circuits, as described in Chapter 10.

The NEC^® specifically calls out that “exposed non-current-carrying metal parts of module frames, equipment, and conductor enclosures shall be grounded.” The requirement to ground this equipment is present for all PV systems, regardless of the system voltage, which means you must always install equipment grounding for your PV installations.

In the following sections, I show you how to make these connections as well as how to size them properly.

Making the equipment ground connection

To properly ground all the conductive pieces of equipment, you need to install what’s known as an equipment-grounding conductor (EGC). The NEC^® defines an EGC as “the conductive path installed to connect normally non-current-carrying metal parts of equipment together and to the system-grounded conductor or to the grounding-electrode conductor, or both.” (I cover the system-grounded and grounding-electrode conductors in the later “System grounding” section.) In short, this verbiage means you need to connect all of those metal pieces of equipment together and then connect that mass of metal to the system ground.

For items such as combiner boxes, disconnects, charge controllers, inverters, and any metallic box holding electrical equipment (such as the battery box), the easiest way to install an EGC is with a properly sized conductor that’s connected to a ground lug inside the boxes. (A lug is a device that allows you to terminate a conductor connection.) All equipment manufacturers have a very specific location and type of lug that you can use to connect the ground, so if a ground lug isn’t in the box, buy the lug specified by the manufacturer.

Using a conductor to connect each module together can be a difficult task. PV modules often have a place on their frames that’s intended for an EGC; usually this spot is in the middle of the long edge of the frame. The problem is that manufacturers often don’t supply hardware for you to actually make the connection. And to make the situation even worse, a module’s instructions generally give you little (if any) direction on how to connect an EGC.

This lack of clear guidelines has caused PV installers to seek outside help with connecting modules. Currently, two methods are used (see Figure 17-1):

• Ground lug: By attaching a ground lug that’s rated for outdoor use (and direct burial) to the back of each module and connecting a conductor to each lug in the array (see Figure 17-1a), you can place the lug at the location provided by the manufacturer (although this location may not be in a convenient place based on your chosen mounting method). However, attaching ground lugs is a time-consuming and detailed process.

• Grounding clip: By placing a specific grounding clip between the module and the racking system, you can bond the modules to the mounting rails (in other words, you can establish electrical continuity). A grounding clip pierces both the module and the racking system at the same time, bonding the two together (see Figure 17-1b) and allowing you to use the mounting rails as the EGC, eliminating the need to run wire all along the back of the array. You need to connect all the rails together to maintain a continuous grounding path, but you can do this pretty easily before installing any of the modules. After all the modules have been installed and bonded to the rails, you can connect a single EGC to one of the rails and bring it into the junction box (or combiner box) with the grounded and ungrounded current-carrying conductors from the array.

  Using grounding clips isn’t as labor intensive as using ground lugs, but it isn’t a foolproof method. You absolutely have to make sure that you place the grounding clips in the appropriate locations (as specified by the manufacturer of the grounding clips you’re using) and that the grounding clips are engaged properly (meaning they’re in full contact with the module and rack). If you install grounding clips incorrectly, then the module won’t be properly bonded to the rail. Also, grounding clips are only rated for a single use. So if you tighten them to the rail and then want to move something, you must replace them.

Both the ground lug and grounding clip methods allow you to install a continuous ground path for the PV modules. Make sure that if any module is ever removed from the system, the path to ground for all the other modules isn’t interrupted. If you use the lug method, this generally means you have to use a continuous conductor. If you use grounding clips, then as long as all the modules are properly bonded to the rail (which is the path to ground), the ground path won’t be removed until the racking system is removed.

System grounding

All the EGCs you run will be connected to system grounding. The system ground is composed of a grounding electrode (GE) and a grounding-electrode conductor (GEC).

• The GE is the conductive object in direct contact with the earth. It’s often a ground rod — a copper rod that has been driven 8 feet into the ground (some jurisdictions require two ground rods driven a minimum of 6 feet apart and also that the two be bonded together; if you’re adding new GEs, make sure you know what the jurisdiction requires).

  Another commonly used grounding electrode is called a Ufer ground. This is where the steel reinforcement in the concrete foundation of a building is used as the contact with the earth.

• The GEC is the conductor that connects the grounding electrode (either a ground rod or a Ufer ground) to a point where all the other grounded conductors can be connected. Typically, this point is a 6 AWG or 4 AWG wire that’s connected to the grounding electrode on one end and the ground busbar inside the MDP on the other end. (A ground busbar is where all the EGCs are terminated; it’s in direct contact with the metal MDP enclosure.)

Article 690.47 is the main NEC^® reference for PV system-grounding requirements; a number of changes were made to this article in the 2008 version of the NEC^®. I walk you through the commonly accepted and generally approved methods for connection and sizing in this section.

Making the system ground connection

For utility-interactive PV systems (either grid-direct or battery-based), the PV system grounding is connected to the existing system grounding installed for the utility system. You can make this happen in a couple different ways. Note: The method you choose depends on the location of your inverter and its relation to the existing grounding-electrode system.

Figure 17-2 helps illustrate the two common ways to make the system-grounding connection in a utility-interactive system:

• A ground rod (or two) connected to the inverter system grounding by a 6 AWG GEC: Keep in mind that you must connect the existing GE (from the utility) to this new grounding-electrode system you just installed. The new and existing GEs must be connected together with a GEC that’s the same size as the existing GEC from the utility. (See Figure 17-2a.)
• A GEC that’s run from the inverter to the existing GE: As long as the GE from the inverter is continuous and sized based on the NEC^® requirements (I cover the sizing in the next section), most jurisdictions accept this grounding method (shown in Figure 17-2b). You can even use this GEC to bond the equipment installed on the AC side of the system as it runs through the equipment. You can’t break this GEC, but you can crimp EGCs to the GEC inside the equipment to bond the equipment to the GEC. This method provides the low-resistance bond to the grounding system that the Code wants you to get.

For stand-alone, battery-based systems, you must install a brand new GE (because there won’t be one from the utility). You then have to connect two GECs, one originating from the AC MDP and the other from the DC wiring enclosure (where the inverter connects to the battery bank). These GECs will be either 6 AWG or 4 AWG, depending on the GE.

Determining the utility’s requirements

Anytime you’re installing a utility-interactive system, the utility will want to know that you’re connecting to its grid with a PV system as opposed to having a PV system connected to its grid without its knowledge (known as guerrilla solar). By now, most utilities in the United States are familiar with the process and have some standard paperwork available, typically a net-metering agreement. Here are the standard contents your average net-metering agreement should contain (the agreement in a particular jurisdiction may include additional points):

• The system owner agrees to some fundamental equipment requirements to ensure the safety of the utility’s workers.
• The utility outlines how it’ll calculate the energy sent to the utility and the amount sent to the home or business (this is generally referred to as the true-up period).
• Special utility-mandated safety requirements are noted. (The number-one piece of safety equipment required is a visible and lockable disconnect, which may be the same disconnect that you install to comply with the NEC^®; see the earlier “Locations for disconnecting means” section for more on this).

If you’re installing PV systems in a specific utility’s area, I suggest requesting a copy of the net-metering agreement before you ever talk to customers. That way you know what the utility requires before you even set foot on someone’s roof and can be thinking through all those requirements during the site survey (which I describe in Chapter 5). Having a list of the utility’s requirements ahead of time can help you avoid any surprises (like unplanned disconnects) at the end of the job when they’d be very expensive for you to fix. (That’s right. You’re on the hook for mistakes like this because it’s your responsibility to know the requirements.)

Many utilities want to see a disconnect they can walk up to 24/7 and disconnect the PV system from the utility grid. Figure 17-3 illustrates the common location for a utility-required disconnect (which is before the inverter interconnects to the utility and next to the existing utility meter). The idea is that if a utility worker is working on the lines, the disconnect gives him the ability to disconnect the PV array from the grid and eliminate the possibility of being shocked (or killed) from a PV system. The reality of the situation is that any properly installed inverter will shut down as soon as the utility goes away (see the anti-islanding discussion in Chapter 9). Some of the largest utilities have eliminated or relaxed their requirements, indicating a higher level of comfort with PV technology and the safety equipment already installed. If the utility you’re working with does require the disconnects, you may be able to start a dialogue to eliminate this requirement by doing your research on these other utilities.

Making a load side or line side connection

To meet the NEC^® requirements for connecting to the utility, you have to refer to NEC^® Article 690.64 (if you’re working with the 2011 version of the NEC^®, it’s Article 705.12). The following sections set the requirements PV system installers are held to, depending on the interconnection point: load side or line side.

• In residential applications, you typically achieve an interconnection to the utility by placing a circuit breaker inside the MDP and connecting the inverter output circuit to that breaker. It’s the easiest method of connecting the inverter because adding a breaker to an MDP is a straightforward process. Connecting to the utility this way is known as a load side connection because you’re connecting to the load side of the main OCPD in the MDP.

• If adding a breaker to the MDP isn’t an option (because the MDP is full or the PV array output is too large), you can make a line side connection, which is when you connect the inverter’s output circuit to the utility before it reaches the main OCPD in the MDP. This approach is common in commercial applications.

  If you want (or need) to connect to the line side of the main OCPD, you should discuss the options with the local electrical inspector. This method requires you to make connections that the inspector may not be familiar with, and walking through the options beforehand instead of after you’ve installed the system can save you a lot of trouble.

  Note: If you need to interconnect with the utility at this point, you should be aware of the following main Code requirements:

  • You need to have a disconnect and OCPD immediately adjacent to the point of interconnection. This disconnect should be grouped (located) with the other disconnects for the facility.
  • The disconnect should be rated for at least 60 A, and you must use a fused disconnect to meet the OCPD requirement. However, the fuses can be less than the 60 A specified, based on the inverter output current.
  • You can’t have more than six disconnects to shut off all the power sources. This requirement means that if you’re in a commercial facility that has six disconnects already installed, you can’t add a seventh one for the PV system; instead, you must find a way to disconnect all the power sources with fewer than six switches.

For the more common systems that are installed on the load side of the main OCPD, the NEC^® has seven different requirements you need to meet. Addressing and meeting all seven requirements may seem daunting at first, but in reality, all seven requirements are achievable for the majority of your installations. I note the requirements in the following sections.