While boatbuilders have been slow to embrace distributed power, its growing versatility suggests a promising future.
It has been five years since I took a detailed look at distributed power systems (see "Networking," Parcs 1 and 2, Professional BoatBuilder No. 97 and No. 98; and "Shakeout," PBB No. 99). It seemed to me then that because of its benefits-reduced wiring harnesses, simpler installations. and improved diagnostics-production boatbuilders would rapidly adopt distributed power. With the exception of certain bass boat and ski boat builders (more on this later), I certainly got that wrong! Although a handful of other boatbuilders have implemented distributed power systems on one or two model lines, as far as I know, the only production boatbuilders that today install this technology on all their models are Sweden's Hallberg-Rassy and Australia's Riviera Australia.
Given this slow acceptance, and the consequent low volume of sales, it's been a rough few years for the various manufacturers of distributed power systems. Nevertheless, and somewhat surprisingly, most of the companies I wrote about, including E-Plex (Kent, England), DNA's Digital Switching Systems, or DSS (Raleigh, North Carolina), Moritz (Dublin, Pennsylvania), EmpirBus (Uddevala, Sweden), and Capi2 (Nuenen, The Netherlands), are stili in contention. They have been joined by newcomers such as BEP (CZone) of New Zealand, Mastervolt (Amsterdam, The Netherlands), and E-T-A (Nuremberg, Germany).
Cost, Complexity, and confidence
A number of issues contribuite to the slow growth of distributed power systems. At the top of the list are the limited resources we have in the marine field, and the systems' complexity.
It has taken an enormous investment by the automotive industry, including expensive early failures and recalls, to develop the highly sophisticated distributed power systems now found on al1 modern cars, combined with a slick user interface at the dashboard. There are no equivalent resources in most segments of the boatbuilding industry. Attempts to match the sophistication of the automotive world without the resources to fully develop the user interface have led to a leve1 of complexity that's difficult for boat builders and owners to master. A number of builders have entered the distributed power field and then withdrawn in frustration, spreading similar frustration among their boats' owners. An engineer from SeaRay, which experimented with a couple of E-Plex systems early on, spoke of the difficulty of "introducing high-level technology into an industry with relatively low training levels." He mentioned other problems, such as circuit boards being bent and damaged when display units were screwed onto uneven fiberglass surfaces, illustrating the difference between a production boatbuilder and the highly controlled manufacturing processes in a car factory. Paul Holland, the general manager of E-Plex, says: "The degree of complexity which makes absolute sense to a group of engineers in a meeting becomes too much for boat owners, who only use their boats on an occasional basis. Just because we can do something doesn't mean we should do it." He adds: "Even though complicated things are going on in the background, we have to maintain the illusion of simplicity." Peter Udden at Capi2 expresses similar thoughts: "As a developer we are always conflicted inasmuch as we know we can add a lot of technical sophistication, but we need to limit ourselves in order to keep it simple."
Then there's cost. In theory, the savings in the length and weight of copper cables in a boat, combined with reduced installation times, are supposed to more than offset the increased hardware costs in a distributed power system. In practice, it usually doesn't seem to work that way. My own experience is instructive (see "A Multiplexed Malo," PBB No. 119). We installed a couple of heavyduty power buses in my 46' (14m) boat, with the runs optimized to reduce cable length, and then tapped into those runs with numerous Capi2 distributed power "nodes," or digital switches, located close to the load that each node is feeding. Many of my cable runs from the nodes to the loads are really short: for example, less than 3' (1m) to the fridge and freezer units.
Because this was the first boat the builder had wired with this system, and because I was not present when the load cabling went in, the electricians sized the cables from the nodes to the loads based on custom and practice, which is to say, considerably larger than they need to be. There's less footage of cable in my boat than in sisterships, but I doubt there's any less copper. The installation also took longer than normal as the electricians worked their way through the new approach. The underlying problem here is the extremely limited engineering resources available to most boatbuilders. Getting from theoretical savings in a distributed power installation to actual savings takes not only a substantial engineering commitment on the part of the boatbuilder, but also a commitment to educate the work force. Initially, costs will rise significantly. John deRijk, the technical director at Mastervolt, notes that to realize the full benefits of digital switching, a system must be fully engineered, including substantially modified cable layouts as compared to a conventional boat. "This is scary for some boatbuilders," he says, "because then there is no way back; so we still see some designing layouts in the traditional way and then adding digital switching, giving only half the benefits. With growing experience and acceptance of the technology, more benefits and greater cost reductions will be realized." Jarrod Sagar at CZone echoes those comments: "The most critical part of this technology is the implementation. If people don't plan for the installation and adjust their thinking for electrical designs, then they miss the benefits." Aside from cost and complexity, there are other challenges that the distributed power players as a whole seem to have met. "Seven or eight years ago," reports Henrik Nikiasson, the managing director at Trigentic, which owns EmpirBus, "everyone was asking if the electronics wouid survive in a boat. Now no one is asking this. It is ali about function: how much current you can handle, and how the system is integrateci with other systems on the boat." To increase conlidence, most manufacturers provide some mechanism for bypassing the control network if it should fail, enabling the electronic switching devices in the nodes to be operated at the local leve1 (for example, E-Plex and EmpirBus), Others, sudi as Capii, Mastervolt, and E-T-A, go a step further by providing a mechanism for bypassing the electronics altogether (including the electronics in the nodes), in effect hardwiring the load circuits to the power bus. In both bypass methods, the cucuits on the boat have to be turned on and off at the nodes themselves. In the former case, it's done by manually activating the electronics in the nodes; in the latter case, it's done via a mini circuit breaker, or by inserting and removing a fuse.
Production-Line Implementation
Hallberg-Rassy and Riviera Australia are the only builders of larger boats I have come across to cite cost savings as a reason for installing distributed power systems. Most still see the technology as substantially more expensive. Hallberg-Rassy has 10 models ranging in size from 31' to 64' (9.4m to 19.5m). These are true production boats. The wiring harnesses are assembled elsewhere and then dropped in the boat. The harnesses come with plugs that clip into the blocks of distributed power nodes supplied by EmpirBus, another Swedish company, based in Uddevala. The nodes, which are the expensive bit in the system, are instalied in the boat late in the production process, which reduces the capital that's tied up. A laptop downloads a preprogrammed configuration for that particular model, and the boat is ready to go. This is exactiy how things are supposed to work, and it's clearly working well for Hallberg-Rassy: during my recent visit the factory was humming, jam-packed with boats under construction. Riviera Australia chose the CZone system from BEP. Jarrod Sagar reports that Riviera has achieved "time and cost savings even on the smaller craft (30'/9.1m), and the savings get better the bigger you go. One of our New Zealand manufacturers of a luxury 65' (19.8m) custom flybridge sport fisher/ cruiser has seen the installation labor time reduced from 1,300 hours to 1,000 hours, and has also saved around 20% on cable and other components, such as light dimmers." The significant exceptions to the low overall adoption rates are bass boats and ski boats. DSS has made notable inroads in this marketplace.
The company's Tim Sweet points out that to remata competitive, manufacturers of these highiy specialized boats are driven to conscantly innovate. One key consideration with bass boats, for example, is the survival of the fish. "This drives many of the distributed power issues on the boat. Multiple pumps tied to sensors and switches, and the fact that the boat remains unpowered for Iong periods of time while fishing using a trolling motor, require sophisticated and integrated controls to manage power consumption, monitor battery charge, and measure oxygen and temperature levels." Bass boats have two different power distribution stations-the helm while under way in fasi transit, and the bow when fishing-both of which have limited dash space with little room for power control functions. Distributed power systems are particularly well suited to dual controls and a small footprint. Sweet reports: "All of this makes bass boats ideai adopters of this type of technology. At this point, most OEMs have some form of digital power distribution, at least on their high-end boats, but none that I know of has fully integrated systems with displays, GPS, engine data, and power distribution all sharing the same bus strutture." Ski boats are another specialized application requiring sophisticated technology to meet a single primary function: "Towing a skier, wakeboarder, or surfer down a very regulated course at a precise speed with a uniform wake pattern tailored to the weight and preferences of the individuai athlete," Sweet said. Another requirement is music for the "performance." Between runs, the boats are entertainment platforms, necessitating additional accessories. Distributed power systems integrate GPS for speed control, ballast pumps for weight control, and volume on the sound system. Sweet estimates around 20,000 of these boats bave distributed power systems, all with proprietary messaging systems.
Perhaps the most interesting development over the past five years has been the steady shift of distributed power systems converging with the NMEA 2000 CANbus network. NMEA 2000 is a plug-and-play system in the sense that any compatible device added to the network will immediately, without additional programming, be able to receive and transmit data and messages for which the system has a predefined Parameter Group Number (PGN). NMEA 2000 started as a network for sharing navigational data, with the necessary PGNs to do that. The PGN dictionary" has been steadily expanding with PGNs for a range of engine and steering functions, for tank levels, and for monitoring AC and DC power such as volts, amps, and frequency. A number of distributed power manufacturers have recognized that the growing PGN base provides a mechanism for sharing status information and exercising limited switching functions with NMEA 2000 display devices. Early moves into NMEA 2000 took the form of "gateways" that translated data and messages carried on a proprietary distributed nower bus into NMEA PGNs. Over time, distributed power manufacturers began to adopt the NMEA 2000 messaging protocols and the NMEA 2000 network as their default network. The first to make this move was Moritz. Subsequently, DSS and BEP (CZone) followed. E-T-A did the same. This past summer, EmpireBus switched its new NXT product line to NMEA 2000. Sagar notes that as boats have become more complex and intimidating to operate, the users have become less technical. "Customers want integration and simplilication of their onboard systems: everything on one screen with a common logical menu flow. We developed CZone to bring the control and monitoring to one point and have now expanded this to include integration with Simrad/B&G/Lowrance multifunction navigation devices." A builder or marine electrician cannot simply build a plug-and-play distributed power system. The system needs proprietary pr gramming to set current limits and time-delay characteristics in digital circuit breakers, to determine switch functions (on/off or momentary), and so on. As such, even those employing the NMEA 2000 network must put proprietary messages on the bus (the NMEA 2000 porotocol provides a mechanism to do this). Nevertheless, there is considerable scope for expanding the message base within NMEA 2000 to include a much broader range of standardized PGNs that will be recognized by all compatible devices. Tim Sweet has been chairing an active Power Group within the National Marine Electronics Association to add to the PGN base. Other distributed power manufacturers are participating. The initial focus has been on expanding the communications architecture with messages related to data transfer; to control, such as turning switches on and off; and to status, includine various faults and alarms. A longer-term goal is to create messages and an architecture suitable for configuring distributed power systems. NMEA 2000-compatible touch-screens can already be programmed as vimal dis tribution panels, with all the typical functions of a conventional distribution panel. The shift into NMEA 2000 is not without drawbacks. The protocol requires a very specific cable and fittings (DeviceNet) for its backbone, or physical layer. These are expensive. Because of NMEA 2000's plug-and-play nature, there is not the same contro1 over equipment plugged into the bus as there is with proprietary buses. It's easy for a boat owner to add equipment to the bus. There have been instances of noncertified equipment overwhelming the bus with spurious messages. To prevent those problems, NMEA 2000 certification is rigorous, which can be challenging for manufacturers. Henrik Nikiasson notes that it has taken "a lot of work developing their NXT range," but that ultimately the "certification process has been very good for EmpirBus." Manufacturers such as E-Plex, Mastervolt, and Capi2 stili see a significant benefit to using their lower-cost and independent networks, with an interface for transferring information to and from NMEA 2000 devices. Mastervolt has a parallel CANbus running proprietary messages with a gateway that allows certain messages to be shared with the NMEA 2000 CANbus. "We come at distributed power from a different angle to other manufacturers," reports Mastervolt's deRijk. "We design power products as well and see our MasterBus as a movement towards onboard systems integration... with features needed both for power and load management."
In addition to adopting NMEA 2000 or providing a gateway, most distributed power manufacturers have shared a number of other technical advances in recent years. For lighting and other switches, there was a time when we typically saw wiring ninning from the switches to an "input card housed in a power distribution module. That required small-gauge wiring from each individuai switch to the distribution module (meaning a lot of wiring, albeit small gauge), at which point the switch signal was translated to a digital signal carried on the common network cable. Today, most translation takes piace at the switch, with the single network cable running from switch to switch. It considerably reduces installation time and complexity. EmpirBus recently released a clever halfway technology. Up to four conventional switches are wired in parallel to a single cable, with a resistor between each switch. A distribution module supplies a constant current to the feed wire. Because of the resistors, when each switch is operated, it feeds back a d'iferent current to the distribution module. These differences enable the switch to be identified, and the signal is translated and put on the bus. The result is a mechanism for putting multiple switches on a single cable without having to embed the relatively expensive translation electronics in the switches themselves. On the output side, MOSFETS, the pieces of silicon that do the load switching in a distributed power system, have improved in efficiency and come down in price over the years.
Better efficiency results in less heat generated, which enables higher total currents to be handled by relatively small node blocks. The quality of the MOSFETS enables multiple output channels on a node block to be paralleled for high outputs, without the chamels getting out of balance.
Increasing Sophistication
The current, or amperage, measurement devices that trigger the tripping of electronic circuit breakers now have a resolution low enough (down to 10 mA) to detect even the current levels drawn by an LED lighting circuit. This means a "lamp out" warning can be provided on those circuits. At the other end of the amperage scale, distributed power manufacturers have learned how to differentiate highcurrent "transients" from shon circuits; this eliminates the nuisance tripping that sometimes occurred with earlier generation devices. The problem is that MOSFETS respond much faster to transients than traditional fuses and circuit breakers can. And, of course, the software just keeps getting better. All manufacturers now offer light dimming as a standard function. Load and energy management is becoming increasingly common, including soft start for loads with high inrush currents. There are active interfaces with battery chargers and inverters. Measurement of fuel and water levels, even in oddshaped tanks, is more-or-less standard.
Displays constantly improve in quality, versatility, and ease of use. Distributed power systems have the potential to save data and information about events outside the norm, to improve diagnostics and reporting. As more and more information runs on a single bus, the opportunity for diagnostics rapidly increases, along with depth of analysis and troubleshooting assistance. When an Internet connection is available, that interface enables the present crop of systems to be remotely operated (via an iPhone, for example) or monitored. My boat's designer in western Sweden, its manufacturer in Holland, and I together have operated, diagnosed, and troubleshot my Capi2 system in real time.
The Tide Might Be Turning
The past five years have provided a considerable depth of experience for distributed power manufacturers and boatbuilders. The technology is becoming relatively mature. Peter Udden of Capi2 reports that he now has around 10,000 nodes instalied on boats operating in waters from the Antarctic to the Caribbean. He recently heard complaints from his service engineer, who handles ali worldwide issues, that there isn't enough work for one person to do. We are seeing some intermediate steps toward a broad adoption of distributed power in such boats as Bénéteaus, Jeanneaus, and Lagoons (all members of Groupe Bénéteau), which use a l'ited number of switches in the navigation station panel, each activating a cluster of circuits behind the panel. And I come across more and more boatbuilders testing full-fledged distributed power systems. Capi2 has recendy signed up the major Italian boatbuilder Fiart Mare and its latest 4tFOUR genius model. Michele Illiano, production manager at Fiart Mare, notes: We chose Capi2 in order to have an electrical wiring system that is easy to install and maintain..., and to integrate it with a new generation of boat conml systems. This enables us io offer our customeis a very easy-to-manage boat featuring the most advanced technology available." Capi2 has also signed up African Cats, where the focus is on weight savings in their ultra-lightweight catarnarans. Other distributed power manufacturers are seeing similar successes. DeRijk at Mastervoit notes, "There are several keys to success with distributed power. The product needs to be user-friendly and provide tangible benefits. We need to be able to offer education for the technical staff at the dealerships, and provide strong technicd support globally to work with the engineers at the builder level and to support the boat owners after the sale." Although we have had some interesting and sometimes amusing experiences with our first-generation Capi2 system, I do not regret installing it and would do so again. My wife, Terrie, is perhaps the best barometer of the marketplace. For the first couple of years, every time we had any kind of electrical problem on the boat, she would exclaim: "It’s that f...ing Capi2 system." In fact, it never was-it's just that she doesn't trus new technology. This past summer the Capi2 system alerted us to an impending battery problem, and she remarked: "You know, that Capi2 system is coming into its own." I think she's right. The last few years have been tough ones for distributed power manufacturers, even without the recession. It's a tribute to their commitment and stubbornness that ali the key players I identified a few years ago are still with us. I don’t believe it will be another five years before we finally begin to see fairly widespread adoption of the technology.