How Will We Adapt To Foreign Power In The Southern Hemisphere?

Preparing Duet for her South Pacific journey included deciding how to deal with foreign electrical power. Duet is configured and outfitted for US-style split-phase power: 120/240 VAC, 60 Hz.  This is the familiar 3 conductor system comprised of 2 hot wires (often labeled L1 and L2), and 1 neutral wire (plus a grounding wire).  This is shown in figure 1.  120 VAC is available between either L and neutral, and 240 VAC is produced between L1 and L2.  Most of our AC equipment requires 120 VAC, so each item is connected to either L1 or L2, plus neutral.  Only our watermaker and laundry machines use 240 VAC, and they are connected to both L1 and L2, without need of a neutral.  

In contrast, the South Pacific offers single phase 230 VAC, 50 Hz (see figure 1).  This is a 2 conductor system, comprised of only 1 hot (L) and 1 neutral wire (plus grounding wire).  Between L and neutral 230 VAC is available; without some intervention on our part, no other voltage can be drawn from this system.  The power available in the South Pacific therefore presents two potential problems for us:  voltage and frequency.  

Initially, I considered installing an isolation transformer, which is a common solution employed by long-legged cruising boats.  Such a device takes single phase 230 VAC power and creates split-phase 115/230 (close enough to 120/240).   The one weakness of this approach is its inability to convert 50 Hz to 60 Hz, except with a very expensive and bulky frequency converter(e.g. Atlas Marine) that is not suited for our size boat (or budget).

Frequency is not a problem for some kinds of AC equipment, like light bulbs and heating elements.  Motors vary in their susceptibility to damage from operation at the wrong frequency.  The motors I was most concerned about were our reverse-cycle air conditioning compressors.  I expected to be in some very warm locations and wanted to be able to run air conditioning without having to run the generator continuously at the dock. Unfortunately, the manufacturer of our reverse-cycle equipment told us that prolonged use on 50 Hz power would burn out the compressors.

Because of this frequency problem, I decided that an isolation transformer would not solve our foreign power needs.  Instead, I chose to  install an inverter stack that could handle some of our reverse-cycle loads.  Because this amount of inverting requires a substantial amount of incoming DC power to keep the batteries charged, I also would need to install a bank of universal DC battery chargers that could run on any marina voltage or frequency we were likely to encounter.  

Duet was already equipped with a single 3 kVA Victron Multiplus inverter-charger that handled our smaller AC loads (refrigeration plus miscellaneous domestic appliances –microwave, coffee maker, dishwasher etc.).  Together, all of our reefers and freezers require 0.8 kVA.  Miscellaneous appliances episodically add another 1.2 to 1.5 kVA.  This did not leave enough reserve to power any reverse cycle, especially when taking into account the predictable reduction in capacity as an inverter gets warm.  Like many manufacturers, Victron states the power of its inverters at a temperature of 77F.  At 104F, capacity declines by20%.      

Duet has a total reverse-cycle capacity of 56,000 BTU.  Running all of this on an inverter bank was not practical for us, for reasons that will become clear.  Based on previous experience in the tropics, I knew that running 20,000 BTU would keep the boat comfortable in the mornings, evenings and overnight.  Our plan was to use inverted power to provide cooling during those time periods, and the generator as necessary for maximum cooling during the middle of the day.  

Measurements told us that 20,000 BTU would require an additional 1.4 kVA.  This did not include the large reverse-cycle seawater pump which, thankfully, is capable of running directly on dock power.  

I estimated that a second 3 kVA inverter would address our reverse-cycle needs.  Even accounting for derating as the inverters warm up, there appeared to be more than enough power to handle all of our intended loads (1.4 kVA reverse-cycle + 0.8 kVA refrigeration + 1.5 kVA miscellaneous = 3.7 kVA).  

Apart from considering the sustained load inverters need to handle,  one must also consider transient peak loads.  Motor start-up loads can be 4 to 6 times their sustained running load.  Our largest reverse-cycle unit requires 1.1 kVA to run.  At worst, I could expect a 6.6 kVA transient start-up load. Our next largest reverse-cycle unit requires 0.8 kVA to run with a possible start-up load of 5 kVA.  The combined peak load handling ability of two Multiplus inverters is 12 kVA, theoretically enough to handle the simultaneous start up of our largest and second largest compressors.


DC SIDE OF EQUATION

Our inverters would be useless without adequate DC charging capacity.  To a first approximation, one kVA of AC power requires one kVA of DC power (I am not considering inverter efficiency, which for the Multiplus is around 93% when optimally loaded).   I wanted to install sufficient DC charging capacity to keep up with average power utilization over a 24 hour cycle, with periods of higher than average load handled by the battery bank.  The most significant loads would be reefers, freezers, and 20,000 BTU of reverse cycle, all adding up to 2.2 kVA.  At 12 volts, that requires183 amps DC.  However all those devices don’t run continuously. Assuming 75% duty cycles for each of those devices, over a 24 hour period we would need a continuous DC charge of 138 amps. Two 100-amp Victron universal chargers appeared to be more than sufficient for the task, even allowing for their output reduction with warming up (20% reduction at 120 F).  They are capable of running on 90 to 265VAC, and on frequencies between 45 and 65 Hz, so plugging them in to foreign dock power is no problem.

The two 100 amp chargers require a combined 16 amps at 230 VAC, the typical voltage at south pacific marinas.  The seawater pump for our reverse cycle, which I intend to run directly on dock power, requires another 4 amps, bringing our dockside power requirement up to 20 amps.  Typical single phase power pedestals in the south pacific are either 16 or 32 amps.  I have the necessary adapters (IEC 60309 pin and sleeve wiring devices – see figure 2) to plug in to either, but obviously a 16 amp pedestal means we can run only one charger and this reduces the amount of reverse-cycle we can power.   An isolation transformer would not have altered this limitation– there is only so much juice available from the dock.  Hopefully, we will encounter more 32 than 16 amp pedestals.




To simplify power management, I installed a dedicated foreign power inlet on the transom of our boat, rated for 230 VAC, 32 amps (I actually used an ordinary 120/240 50amp inlet since it was less expensive than the ‘european’ model). The inlet cabling (10 AWG) runs to a double-pole 32 amp breaker. Blue Sea Systems makes a nice surface mount enclosure for their AC breakers, rated to IP66 (protected from high pressure water jets), perfect for installing in the lazarette (see figure 3).  



ABYC standards require connection of the DC charger chassis to the boat’s central bonding system using a cable at most one size smaller than the battery bank connection (yes, really, a 3/0 grounding cable!).   Since the grounding wire on the incoming AC shore power is also connected to the chassis, there is no galvanic isolation from other boats at the marina. Therefore, I installed a galvanic isolator, which is where that green bonding cable in the picture is headed.


WIRING UP THE INVERTER STACK

Before installing the second Victron inverter-charger, I needed to decide how it would be connected to the original one.  Victron inverter-chargers can be stacked in parallel or in series (see figure 4 below).  The parallel stacking produces a single phase 2 conductor (one hot, one neutral) 120 VAC system.  The series stacking produces a split phase 3 conductor (two hot, one neutral) 120/240 VAC system.  It is possible to construct a three-phase power system as well (with a third inverter), but this was not relevant for Duet and I will not discuss it further.  




These two options have obvious differences in their external wiring.  What is not obvious from the schematic is that the devices have to be programmed to work in these two distinct ways.  One of the devices is designated ‘master’ and controls how the ‘slave’ is generating power; a cat5 cable maintains communication between the devices .  The key is proper synchronization of the AC wave form (see figure 5).  Our AC power appears on an oscilloscope as a sine wave, with a frequency of 50 or 60 cycles per second.  The sine waves of the paralleled inverters have exactly the same timing.  In contrast, the sine waves of the split-phase inverters are displaced from each other by a half cycle. When one inverter is at peak positive voltage, the other is at peak negative voltage (see figure 5).  




Because the sine waves of paralleled inverters are in perfect synchrony, at any instant in time the voltages in their hot wires are identical.  So, connecting them together will not create a short circuit.  With the split-phase arrangement, when L1 is at plus 120 volts, L2 is at minus 120 volts. There are 240 volts between L1 and L2, and 120 volts between either L and neutral.

As mentioned at the beginning, Duet is set up for 120/240 VAC power.  Our various 120 VAC devices are wired to either L1 and N, or L2 and N, with roughly an equal number on each circuit.  Our only 240 VAC devices are the laundry machines and watermaker.   I was not interested in generating 240 VAC from the inverter bank since running the generator to do laundry is an acceptable option.  In marinas we have no reason to power up the watermaker except to flush, which does not require 240 VAC. Therefore, I decided to wire up the inverter stack in parallel, which provides only 120 VAC power.

When inverters are stacked in parallel, it is important to insure that the DC cables to each inverter are of equal length and cross-section.  If this is not done properly, the inverters will be operating at different DC voltages and this will affect their ability to properly share the AC load.  Equal load sharing can also be adversely affected if the AC output cables of each inverter are not of similar length and cross-section.  Victron’s website contains a variety of helpful documents that discuss the subject of inverter stacking including an excellent Powerpoint presentation entitled “Theory On Wiring Large Systems” and a whitepaper entitled “Parallel and Three-phase VE.Bus Systems.”
When putting together a system like this, it is helpful to have a meter equipped with an AC current clamp (see figure 6) for confirming equality of load sharing between the inverters.



INVERTER TRANSFER SWITCHING

As mentioned above, roughly half of our 120 VAC loads are connected to L1 and the other half are connected to L2.  The same is true of our reverse-cycle machinery.  Without some additional switching, a single parallel inverter stack can be connected to the loads on only one of those hot legs, L1 or L2, but never both.  When external power is available (e.g. generator), L1 powers the inverter and turns it into a charger.  L1 also passes straight through the device and powers all the loads that were previously on inverted power. If external L1 power is connected to loads that are also receiving L2 power, the result is a destructive short circuit.  

Therefore, there needs to be a way of connecting those L2 loads to the inverter-charger when external power is not available, but safely disconnect and isolate them when external power is on line.  To provide this transfer function, I installed a 3-pole 2-position Blue Sea Systems rotary switch in the pilot house near the AC panel (figure 7). The switching logic is shown in figure 8.  




The inverter loads transfer switch would usually be left in the ‘normal’ position which means that the inverter is connected only to certain equipment items that always derive their power through this circuit (mostly small domestic items like microwave, coffee pot, dishwasher, garbage disposal, electrical outlets, etc).  The remaining 120 VAC equipment (bigger items like reverse cycle, hot water heater) get their power through L1 or L2 which are only energized when external power is available (generator or a 120/240 VAC shore power pedestal).  

When we are in a marina with only 50 Hz power available, the switch can be moved to the ‘inverter’ position, which moves all those larger L1 and L2 loads over to the inverter and safely isolates them from any external power that could be accidentally turned on (e.g. the generator).  


POWERING THE REVERSE CYCLE SEAWATER PUMP

All of our reverse cycle units are cooled by a single large seawater pump.  As mentioned above, it is rated for dual frequencies (50 and 60 Hz) and low or high voltage (120 or 230-240 VAC).  Powering the pump directly off of dock power allowed us to take 0.4 kVA off the loads that the inverter bank would otherwise have needed to power.  There are two sets of windings in the motor, and depending on how they are wired, the pump can accept low or high voltage.  The nameplate contains all the necessary wiring information (see right side of label shown in figure 9).  I wired in a 4-pole rotary selector switch so that the pump could be powered either from the new transom foreign shore power inlet (high voltage) or from the usual 120 VAC sources (generator or US-style shore power).  The wire colors in the pump’s junction box (figure 10) conform to the colors noted on the nameplate.  





PERFORMANCE THUS FAR

We have used the system for about 3 weeks while docked at a marina in Papeete, Tahiti.  It is the tropical summer and the weather has been very hot and humid.  During the mornings, evenings, and nights, we air condition the boat and run all our necessary AC loads on the inverter bank.  Our average AC loads are as expected and the universal DC battery chargers do a good job of keeping the battery bank fully charged over a 24 hour period. As a safety check I have run an IR heat probe over all the high amperage DC cabling and termination lugs, and all are appropriately cool.

I have occasionally pushed the system beyond its original design parameters, by running up to 38,000 BTU of reverse-cycle.  The inverters handle this load without any difficulty (around 3.5 kVA total with refrigeration and freezers) however our charging capacity is exceeded and the battery bank slowly gets drawn down. This seems acceptable for limited time periods and with appropriate monitoring of the batteries.  I have considered adding a third DC charger, but space is tight in the lazarette.  It’s always good to have new projects in mind…

Speaking of  new projects, our lazarette is not well ventilated.  Because of heat generated by the stacked inverters and DC chargers, I find myself leaving the lazarette hatch open.  We have not experienced any heat-related equipment shut-downs, but as I have pointed out, excessive heat causes a drop in capacity of inverters and chargers.  Not to mention wear and tear.  Ideally, any forced air ventilation would run off dock power rather than imposing any further loads on the inverters or DC chargers.  It looks like I will have things to keep me occupied as we continue our cruise through paradise.