Ketchikan to Bainbridge Island – August 6 to September 9

The final leg of our cruise started August 6 when we headed south from Ketchikan and took on 600 gallons of diesel on the way. The first night we made an intermediate anchorage just north of the border which allowed a noon time arrival at Prince Rupert the next day. We had sufficient time that afternoon to enjoy lunch at the Wheelhouse Brewing Company in the old Canadian National Railway Station before provisioning with “freshies” at Safeway for our trip south through British Columbia.

From Prince Rupert we repeated the same long day’s journey as last year through Grenville Channel and anchoring in Hawk Bay on Fin Island near Hartley Bay. The next day, we went fishing in Otter Channel and Marcia caught a silver salmon.

We continued south the next day and fished a late morning bite at Ulric Point at the NE corner of Aristazabal Island but came up empty-handed. After fishing we continued south through Laredo Channel, transited Meyers Narrow finally ending the day at Mary Cove on west shore of Roderick Island.

With benign sea conditions forecast, we exited Findlayson Channel, transited Milbanke Sound into Seaforth Channel. A fishing session at Idol Point yielded nothing so we continued, finally anchoring in Gosse Bay along Gunboat Passage for the night.

An early start had us at Rattenbury Point along Dean Channel for the early morning bite and Marcia landed a coho. We then headed south and fished a mid-day bite at Uganda Point, south of Namu in Fitz Hugh Sound but struck out. With the salmon hardly biting, we bet on a sure thing (at least for fun), and headed to Pruth Bay, and the beautiful beaches accessible from there.

We spent two nights in Pruth Bay, with walks to the beaches in the morning and afternoon. The Hakai Institute research station is kind enough to allow transient boaters dingy access to their sturdy docks. They even have a hose with fresh water at the dock that can be used to wash the sand and salt water from any canine companion that happens to have played in the surf.

On our morning walk to the North Beach, we found a set of paw prints from two wolves that had traveled along it shortly before our arrival. Calvert Island is sufficiently large to support a healthy population of wolves. Tracks, sightings and even encounters of wolf pack members occur regularly, so we kept Drake close by at all-times.

When we left Pruth Bay, we headed across Fitz Hugh Channel to Fish Egg Inlet. We first dropped prawn pots before fishing along the shores of Addenbroke Island, north of the lighthouse. Once again, the salmon were scarce, at least for us. We anchored the night in cove a short distance from where we dropped our prawn pots.

Since fishing had been so poor, we decided the next day to retrieve our prawn pots and head south around Cape Caution. The harvest from the prawn pots did nothing to dissuade us from that decision. Of the three pots we had used, one was not bad, one was poor and the other had a big octopus that had eaten the few prawns that had entered before the octopus joined them (it could have been worse since last year, we had two of the three pots set pull up octopi). The conditions around Cape Caution were pretty good and we didn’t need to put our stabilizing fish in the water. After crossing, we anchored in Port Alexander on Nigel Island about 11 NM northwest of Port Hardy. This was the same anchorage we had used before crossing Cape Caution on the northbound portion of the cruise.

The cruise into Port McNeill the next day was uneventful, and we arrived late morning. We decided to spend three nights in Port McNeill so that we didn’t feel rushed and Drake had ample time for ball play and walks.

For slow boats, such as ours, the section between Port McNeill and Campbell River is often a problem because the currents can be large and their timing unhelpful. Throw in winds, which can increase sharply during the day with a direction that is almost always on your bow or on your stern and picking when to leave and where to stop is difficult. Last year, the stars, or rather the sun and moon/, lined us perfectly and we were able to go from Port McNeill to Gowlland Harbor (directly across the channel from Campbell River) in one day. This year we weren’t so lucky and needed an intermediate stop. We chose as our anchorage, Blenkinsop Bay. The strong westerly winds in Johnstone Strait made the anchorage windy with some chop, but it wasn’t terrible. By using it and starting early the next day, we were able get a current boost through Discovery Passage beyond Campbell River into the Strait of Georgia. That night we anchored in Tribune Bay on Hornby Island. From there, we continued down the Strait of Georgia to Nanaimo where we transited Dodd Narrows to gain the more protected waters of the Gulf Islands (the British Columbia islands across from the Washington San Juan Islands) anchoring that night in Montague Harbor on Galiano Island.

Once in the Gulf Islands, our long days of travel are generally behind us because the area is compact. We stayed two nights in Montague Harbor before heading to the Port Browning Marina on North Pender Island for two nights. We then crossed the border back into the USA, spending our two nights in Garrison Bay on San Juan Island, one night in Blind Bay on Shaw Island and a night in Griffin Bay on San Juan Island. During that time, we went to shore often to do walks and play ball with Drake. The weather was pleasant.

From Griffin Bay we crossed the Strait of Juan de Fuca and docked in Port Townsend, on Saturday, August 29. We had an appointment for Monday to install a new AIS-VHF antenna and pull a new cable from it, down our mast and into the locker beneath our lower helm where the new AIS (Automatic Identification System) black box installed during our Spring 2026 haul out was located.

From Port Townsend we headed up to Cap Sante Marina in Anacortes for two days to sit out a short spell of windy weather. After that we spent four nights at the Deer Harbor Marina where our yacht club leases dock space as an “out station” for its members. We’ve come to enjoy this leisurely travel schedule towards the end of our cruising season..

We finished our summer cruise with a night at anchor in Prevost Harbor on Stuart Island (and requisite walk out to the Turn Point Lighthouse) and a night anchored in Griffin Bay (with a walk to the Cattle Point Light House). One final, long day across the Strait of Juan de Fuca, and down Puget Sound to Bainbridge Island returned us to where we started nearly 4-1/2 months earlier.

Leg 5 – 23 Travel Days, 907.3 NM, 157.8 Engine Hours

Juneau to Ketchikan via Glacier Bay – July 2 to August 2

We did it all on this leg, sightseeing, fishing, short and long travel days, and waiting for weather and dock congestion. During that time, we ended up spending 13 nights within the borders of Glacier Bay National Park (five of those nights in peripheral areas not covered by the permitting system).

We left Juneau on Thursday, July 2 ahead of the Independence Day long weekend and anchored in Sawmill Bay, far up Excursion Inlet. While within the borders Glacier Bay Nation Park (GBNP), entry is not controlled like the waters of Glacier Bay. With a cool and damp weather forecast, we ended up staying in Sawmill Bay for two nights.

On July 4, we traveled west along Icy Strait past the entrance to Glacier Bay and made our way to the northern end of Dundas Bay (also a non-permitted portion of the park). We settled in for 3-nights of damp and unsettled weather. While outside in Icy Strait, winds rose to 20-30 kts, we saw less than 10 kts.

Upon leaving Dundas Bay with better weather, we headed across Ice Strait for some halibut fishing in Mud Bay. After Marcia caught a couple, we went closer to shore an anchored for the night. The next morning, we returned to the same location and caught an additional halibut. For the night we went east in Icy Strait and anchored in Flynn Cove, where we launched the dinghy, went to shore so that Drake could get some ball play.

Several months earlier we obtained an “advance-notice” permit of 7-days/6-nights for Glacier Bay starting on July 11, but in a few days earlier we also obtained a “short-notice” permit for two days entering on July 9. With those 9-days, 8 nights in the permitted portion of Glacier Bay National Park, we visited both the west and east arms as well as spend both the first and last nights in Bartlett Cove so that Drake could get some ball play and shore time (dogs are prohibited from the shore in GBNP except in Bartlett Cove and then only in limited areas).

In Tarr Inlet, at the head of the west arm, is the tide water Margerie Glacier, which calves ice directly into the inlet, and the Grand Pacific Glacier, whose terminus is several hundred yards short of the water. The Margerie Glacier leads quickly up towards the steep summits of the Fairweather Range, which separates Glacier Bay from the Gulf of Alaska, The Grand Pacific Glacier leads north about 16 miles into British Columbia. We spent about an hour drifting in front of the glaciers just enjoying setting. While drifting, a car size chunk of ice “joined” us. We gave it a push with a boat hook sending it away before we restarted the engine and departed.

We took advantage of our extra days in the park by spending time in the east arm and traveling to the very end of Muir Inlet. It is a lovely section of the park but doesn’t get as much recreational boat traffic as the west arm. It too, has glaciers, such as the Muir and the Riggs but most end before reaching the waters of Glacier Bay.

While returning to Bartlett Cove, like many boaters up and down the coast this summer, we got to share the waters with the 390’ mega yacht Launchpad owned by Mark Zuckerberg of Meta/Facebook. Our encounter was in Sitakaday Narrows, a few miles NW of Bartlett Cove. I did my best to get out of the way as it was doing 19 knots and throwing a large wake. We spent our last night in Bartlett Cove and gave Drake his last play session in the park.

Since Glacier Bay is the furthest north we go, it has become our turnaround point and we are now heading (more or less) south. We decided to take Stephens Passage on the east shore of Admiralty Island into Frederick Sound. For our first night after Glacier Bay, we used Colt Island, about 8 miles SSW of Auke Bay near Juneau. While not particularly scenic, it was convenient and it allowed us to reach the next day one of our favorite anchorages on the west shore of the West Brother Island in Frederick Sound. While in the area, we were going to try some fishing for prawns and halibut. After a couple of days of mediocre results, we decided to head into Petersburg. Along the way, we experienced some strong winds. Even while in Petersburg we had winds in the mid-20’s with gusts in the low 30’s (knots). Fortunately, we were securely tied to the dock at the time.

After two nights in Petersburg, we headed through Wrangell Narrows and made our way into Ernest Sound. We spent a few days in Ernest Sound trying some new prawning areas. While we got prawns in most of them, the hauls weren’t great and we decided to head towards Ketchikan.

This final section, about 60 miles, took us a remarkably long time to travel. From our last anchorage in Ernest Sound, Santa Ana Inlet (photo on left of BC wild fire smoke enhanced sunset while in Santa Anna), we went first to the public dock in Meyers Chuck (photo on right of float plane shortly after takeoff from public dock in Meyers Chuck). We spent two nights here, one of which was on account of weather. Drake didn’t object because he got to play with another labradoodle who belonged to one of the community residents. While talking with them we learned that a fishing area near Ketchikan, District 1, had not opened for commercial fishing as expected and the harbors were full non-fishing boats. Many boats were anchoring out in Tongass Narrows, the channel passing in front of Ketchikan. Subsequently, we also learned that because of construction on the Bar Harbor floating breakwater, some permanent boats were being relocated to slips that would typically be available for transient boats. Rather than risk having to anchor out in the very busy Tongass Narrows, we headed over to Thorne Bay on Prince of Wales (PoW) Island. We spent three nights there, doing lots of walks and play sessions with Drake.

Finally, on August 2, after hearing that commercial fishing had reopened in District 1, we left Thorne Bay and headed into Ketchikan, where we were able to get a slip.

Leg 4 – 24 Travel Days, 787.8 NM, 140.3 Engine Hours

Meandering to Juneau – June 7 to June 28

Meander seems to be the operative term to describe our cruising style this season. Similar to our previous leg out of Ketchikan, we first did a short excursion near Sitka then returned to stay a few more days.

South of Sitka for about 30 miles is a section of coast that has offshore islands that create a protected travel route with many good anchorages. From Sitka, we cruised to Kliuchevoi Bay, which is next to the very popular Goddard Hot Springs. This was our first visit here because we’ve been put off by the crowds that we expected to find here. Perhaps because it was still early in the cruising season, the hot springs were not crowded, and we managed to have one of the two USFS constructed bath house to ourselves on the two days we were in Kliuchevoi Bay.

From here we made a short jaunt to First Narrows Cove, nestled between three islands with a narrow entry channel. It was rocky and had no functional beach, so we then went to Seven Fathom Bay where the USFS has a recreational cabin. In front of the cabin was a grassy area above a shallow rocky beach on which we landed our dinghy. Drake thought the beach perfect, so we suspect we’ll be visiting again. Our last night before returning to Sitka was in Leesofskaia Bay, a short distance from Sitka.

The next morning, we docked back in Eliason Harbor of Sitka, coincidentally in the same  slip as we had five days earlier. While in Sitka we attended one of the evening concerts from the Sitka Music Festival, did lots of walks on the trails around the town, and picked up all the packages we were expecting.

On Friday, June 19, after a luxurious seven days in Sitka, we headed out. We had been communicating with Kathleen & John Douglas on Laysan (a sister ship to Alpenglow), and had agreed to rendezvous on Kuiu Island. We made an intermediate anchorage at Ell Cove on Baranof Island, before meeting up with Laysan. Also joining the rendezvous was Luck Dragon owned by Ed and Carlene (aka, Charlie). Luck Dragon is a “classic” Diesel Duck 462 that is a cousin ship to our sedan style Diesel Duck. We dropped our dinghy to shuttle folks between the vessels for our evening gatherings on Laysan. After two nights in the Shelter Bay area of Tebenkof Bay, we all went our separate ways.

Our next stop was a relatively short distance north to Rowan Bay. There is a USFS dock here but it isn’t advertised as a public dock. We went and checked it out with our dinghy hoping for convenient shore access for walking the roads. As best, I can determine, it is more likely a work float or designed for float planes. The USFS has an extensive road system on Kuiu Island and they may stage some of their support activities from here. There were several trucks parked at the top of the dock. We did a short walk and found what we believe were moose droppings.

We then crossed Chatham Strait to Gut Bay where we dropped our prawn pots before anchoring. The catch the next morning was skimpy so we did not reset but instead headed back across Chatham Strait to Saginaw Bay. John Douglas had mentioned a trail that led from the anchorage at its end to connect with the Kuiu Island road system. This was a “new to us” anchorage and perfectly fine but a long ways back. When I had talked to John about the anchorage, I neglected to ask which stream fork led up to the trail and, of course, when presented with a choice, chose the wrong fork. Something for another visit.

A long day from Rowan got us to Pavlof Harbor on Chichagof Island and a shorter one the next day to Whitestone Harbor. We were positioning to go into Juneau but with a windy forecast, rather the moving to Funter Bay about 12 miles closer to Juneau than Whitestone, we elected to stay two nights in Whitestone. A very early start on Sunday, June 28 had us in Juneau before 10 AM.

Leg 3 – 13 Travel Days, 385.5 NM, 66.7 Engine Hours

Meandering to Sitka – May 13 to June 3

In recent years we have set goals that get us north of Ketchikan quickly (e.g., Little Norway Festival in Petersburg or beating the start of the commercial prawning season). This year, absent such a goal we took a less scripted and meandering path.

First, since it had been a few years since we we’ve been around Revillagigedo Island  (this is the very large island on which Ketchikan is located). The circumnavigation of the island is about 125 miles (not counting diversions for anchoring) but since we don’t particularly enjoy the SE corner of the route, we traveled clockwise around the island as far as Walker Cove on the east channel of Behm Canal then turned around to retrace our route counterclockwise. Besides, Walker Cove, we used Yes Bay and Traitor Cove as intermediate stops in both directions. Traitor Cove was a new stop for us. It has a public dock which we were able to use on the return leg of the trip. After this 6-day jaunt we returned to the Bar Harbor docks of Ketchikan.

After two nights in Ketchikan, we started our northbound journey in earnest and traveled first to Frosty Bay, about 2/3’s of the way from Ketchikan to Wrangell. We’ve gone past Frosty Bay many times since it is only about 4 miles north of Santa Anna Inlet, an anchorage we’ve used often. There is a USFS cabin at the bay but we didn’t have time to check it out before leaving the next morning. We continued the next day to Berg Bay, which also has a USFS cabin. Here we did have time and we launched our dinghy, taking Drake to shore and playing ball with him on the grassy patch in front of the cabin.

From Berg Bay we traveled to Wrangell for the Memorial Day weekend. One of the attractions of going to a dock in a SE Alaska town are the walks we can do. Wrangell has sidewalks along most of their roads as well a few trails, so Drake is always anxious to get off the boat and start exploring the different scents. It being spring-time in SE Alaska, one of the scents are the blooming skunk cabbages.

Up to this point we had not done any prawn fishing. The commercial prawn season started on May 15 but it was now closed in most districts. We decided to try some of the areas in Fredrick Sound that were productive for us last year. We traveled first to Sanborn Canal in Port Houghton. The next day, we dropped our prawn pots on the way into Hobart Bay for the night. We were disappointed by the slim harvest of prawns but attributed some of the shortfall to the octopus that was in one of our pots when we pulled it up.

From here we traveled a little north to Windham Harbor to drop pots. The next day’s harvest was pretty good, but we had to work for it. Both this year and last year, we've been surprised by how strong the tidal current is here, and by its timing. We pull our pots by hand, and the current makes that harder: the boat drifts with the current, but the pot sits still on the bottom. So, as we haul the line in, it's dragging through moving water the whole way — like pulling a drogue. The stronger the current, the harder the pull. We thought we were pulling pretty close to slack (an hour after low) but the current was acting like nealy a one knot ebb current.

Even though it had only been a few days since we had been in Wrangell, Drake was getting antsy, so we headed to an anchorage with a nice play beach, Honey Dew Cove on Kuiu Island. Two nights here seemed to do the trick. To reward ourselves as well as satisfy Drake’s needs, we headed next to Warm Springs Bay on Baranof Island. We were able to get on the public dock which made accessing the trail to Baranof Lake and the hot baths convenient. Again, two nights seemed to do the trick.

An “oh-dark” start from Warm Springs Bay allowed us to get to the DeGroff Bay near Sitka in one day. The next morning, June 3, had us on the dock in Eliasen Harbor in Sitka completing this leg.

Leg 2 – 16 Travel Days, 569.4 NM, 93.5 Engine Hours

Sweet 16 – Bainbridge Island to Ketchikan – April 29 to May 11

We started our 2026 cruising season (our sixteenth trip) on Wednesday, April 29 with a relatively modest first day of cruising from Eagle Harbor on Bainbridge Island to the Port Townsend Boat Haven and a quick visit with the folks at Port Townsend Shipwrights Coop (PTSC).

During our regular maintenance haul out earlier in April, we had PTSC upgrade most of the electronics at our helm stations. Our multifunction displays (MFD’s) at both our upper and lower helms were the builder installed “classic” Raymarine E- series models E80 & E120. Our preference is to replace critical equipment before failure to avoid the disruption and possible extra cost of emergency repairs or replacement. We deemed the radar portion of the MFD’S role critical and decided to do the work now. The newly installed Garmin equipment was operating correctly except for the new Garmin AIS which was performing subpar. We stopped by Port Townsend to see whether a new AIS-VHF antenna would fix the issue. Sadly, it didn’t and our view of AIS targets remains less than we’d like.

While in Port Townsend we crossed paths with a “cousin” vessel, the classic Diesel Duck Shearwater owned by David Cohn. David is an avid and expert fisherman and spends much of his Alaska time in the Sitka area laying in loads of seafood for the winter. He had left from his home in the San Juan Islands a few days us before but returned to Port Townsend when he saw some erratic electrical monitoring equipment not working correctly. His problem was quickly fixed with a new connector.

After Port Townsend, we positioned in Prevost Harbor on Stuart Island in the San Juan Islands Archipelago. We made the short crossing the next day to Port Browning on North Pender Island. It is a Nexus Canadian Border station, and there is a grocery store only a half mile walk from the marina. We wanted to provision with the fresh produce that we weren’t able to take across the border. Port Browning Marina also has a huge grass field where Drake can play ball until total exhaustion (as indicated by his tongue hanging sideways out of his mouth). It being May 1, there was a May Day Celebration being held on the lawn during our visit.

From here we started a steady march northward. On May 2, we traveled 74 miles to Tribune Bay on Hornby Island. On May 3, we traveled a more modest 64 miles to Otter Cove on Discovery Passage but transited the often-problematic Seymour Narrows north of Campbell River. May 4 was 95 miles to Port Alexander on Nigel Island, positioned for crossing Cape Caution. A 93-mile day on May 5 got us around Cape Caution, during which we deployed our stabilizers in the water, and up to Strom Cove on Seaforth Channel, north of Bella Bella. Next day’s 68 miles put us in Khutze Inlet along Graham Reach and catching us back up with David Cohn’s Shearwater. With a forecast for deteriorating weather and the memory of being stuck the previous year in Klewnuggit Inlet on Grenville Channel for four nights, we traveled 89 miles on May 7 to an overnight in Kelp Passage on Porcher Island followed by a short 20 mile day to get us to Cow Bay Marina in Prince Rupert.

After two nights in Prince Rupert (and lots of play for Drake despite periodic rain), the weather improved enough for us to travel on May 10 the 33 miles to Brundage Inlet on Dundas Island, where we again shared the anchorage with Shearwater. The next day, May 11, with an additional 57 miles we arrived in Bar Harbor Marina in Ketchikan.

Leg 1 – 13 Days, 682.2 NM, 104.8 Engine Hours

Monitoring Battery Activity - Part 1 - The Discharge Cycle

We installed our Lithium Ferro Phosphate (LFP) battery bank in August 2021 (Out with the Old, in with the New).  It consists of 10 Battleborn GC2 12V-100 Ah batteries arranged as two serial banks of five paralleled batteries giving us 500 Ah at 24V.  We also installed a Victron Cerbo to help control and monitor our Victron equipment (chargers and battery monitor).  I subsequently installed a Raspberry Pi single board computer running Signal K to record and display data being generated on board (Boat Data).

I’ve now been collecting the data for three years and I haven’t been doing much more than displaying real time data while on board.  As a winter project, I’ve tried to organize and analyze it more carefully.

While at the dock and connected to shorepower, the batteries are not doing very much. The AC needs of the boat (e.g., water heater, toaster oven) are supplied directly from the shore (mediated through the isolation transformer and inverter/charger). The DC needs (e,g, refrigerator, lights) are handled by batteries working with the DC charger half of the inverter/charger.  While actively cruising and underway, the alternator driven by the propulsion engine provides the current for DC loads while the AC loads are taken care of by inverter part of the inverter/charger powered through the alternator.

It is only while at anchor (or occasionally at a dock), with no shore power connection, that the batteries must do work and discharge some of their stored power.  In the three cruising seasons for which I have data (2023, 2024 and 2025), I identified 247 depletions of significance.  I ignored the small discharges that occur when transitioning from shore power to engine power when leaving the dock, and the reverse situation, engine power to shore power, when arriving at a dock. I also ignored short duration discharges associated with events where the engine is off and you aren’t on shore power (e.g., visiting a fuel dock, waiting in a temporary anchorage for currents to subside).

The first chart shows the distribution by duration of the 247 discharge cycles.  Our cruising style is one of motion and we only spend multiple nights at anchor in one spot a dozen or so times a year.  When we do, we run our generator daily to recharge the battery banks. A lot of the 20+ hour discharge cycles are probably associated with multiple nights at one site. Being a slow boat, cruising around 6-1/2 to 7 knots, we tend to put in long days to cover the same distance that faster boats do. The short duration discharge cycles often represent a 7 PM arrival at a destination followed by a 5 AM departure the next morning.

The next chart shows depth of discharge (as measured in amp-hours, Ah) distribution for those same 247 discharge cycles. The same comments as above about deeper discharges being associated with multiple nights at anchor and smaller discharges representing longer days underway apply to the distribution.

The last chart shows the power consumed by hour of the day. There are certain items on board that once they are turned on are rarely turned off.  The big examples are refrigerators and freezers (we have two of each) and all our monitoring equipment (e.g., the NMEA2000 bus). 

The data show that we consume 8-9 amps as a baseline. Because we are primarily operating on DC, while at anchor, we tend to turn on the AC inverter part of the Inverter/Charger only when we need it (e.g., using the Starlink antenna). Turning on the inverter tends to increase our usage by an additional 4 amps (we rarely use the inverter to run a large AC load like a toaster oven or electric kettle).

The last major DC load is the Kabola furnace.  On cold mornings, we often heat the boat up with the Kabola.  It will easily use 8 amps as the pumps and blowers kick on but after things warm up, the load usually drops to around 4 amps. There is a temperature dependence to our usage.  Cold weather will result in longer and more frequent operation of the Kabola while hot weather causes higher duty cycles in our refrigerators and freezers. The highest usage is in the early evening when we would have AC power on in order to watch streaming TV via the Starlink antenna and set the Kabola thermostat up to keep the boat comfortable.

When comparing our amp or amp-hour numbers, remember that we are operating at 24v DC.  The equivalent numbers for a 12v DC system would be double.

 

Monitoring Battery Activity - Part 2 - The Charge Cycle

The “yang” to the discharge cycle’s “yin”, is the charge cycle. We have three ways to restore the batteries after they have been depleted.

Shore Power – We have a Victron Multiplus 24 3000/70 inverter/charger that when provided AC power can recharge the battery.  While our boat is wired to accept 240V-50A, for simplicity, we only connect to shore power via 120V-30A

Alternator – We have two alternators on our Lugger 1066T diesel engine, one for the start battery and the other for the house bank battery. The house bank alternator, a Leece-Neville 4740JB (24V-200A), is regulated by a Balmar MC-624.

Generator – We have a Northern Lights OM773LW2 – 9KW generator. This provides AC power to the Multiplus Inverter/Charger but, in addition, we have a Victron Skylla-I 24/100 charger attached to the output of the generator.

The “State of Charge” (aka SoC) is often the metric used to determine the status of the battery.  It runs from 1 (or 100%), when the battery is “full” to 0 (or 0%), when there is no power available.  For our battery bank, each 1% change in the SoC is about 5 Ah.  With a 24V battery bank, that 5Ah of energy would power the base load of our boat (8-9 A) for about 35 minutes.

For this analysis, I wanted to see the charging characteristics all the way until battery full and so I eliminated cycles that didn’t go to at least 99% SoC.  I wasn’t quite as concerned about the starting level of the discharge, though.  I ended up with 219 charge cycles that went to completion, 7 for shore power, 27 for generator and 185 for engine.  The small number of shore power charging is not surprising since upon arriving at a dock, in most circumstances, would already be full.

The chart below shows the aggregate charging profile for each method of charging. While the monitoring system collects data continuously, it only preserves the data for the long term once every six minutes (one-tenth of an hour), taking an average of the values since that last time increment.  To allow the comparison of charge cycles with different durations and different battery depletions, I used the charge cycle’s end point as a reference.  At any given time point, I averaged SoC for those charge cycles that extended to it. 

A characteristic of LFP batteries is that they accept high levels of charging current until nearly complete.  My observation is that the charging is nearly constant, limited by what the charging source can provide, until about 98% SoC.  At that point, the current acceptance rate decreases rapidly.  At 99% SoC, the battery monitor, a Victron BMV-712 in our case, may say it is “close enough” and report a 100% SoC. 

On the chart, I’ve had Excel compute the linear regression line for each category with the Y-intercept being set to a 100% SoC. The X- coefficient in the equation represents the slope of the charging curve. An average hourly charge rate can be computed from the coefficient by multiplying it first by 500 (the number of Ah in our full battery bank) and then again by 60 (the number of minutes in an hour).  For shore power that calculation suggests a 48 amps per hour charge rate. For the engine it is 96 amps per hour and for the generator 129 amps per hour.  These charge rates are lower than the actual output of charging source because they are net of the charging source’s output and whatever DC loads the boat’s system requires.

 

Monitoring Battery Activity – Part 3 – The Aging Process

Batteries are such a solid (often made primarily of lead, after all), quiet and passive object, that their “wearing out” seems hard to believe. Underneath their cases of rubber and plastic, however, is a cauldron of electro-chemical processes that can turn motors, power lights, heat water and make life more comfortable. After doing that, providing the batteries with appropriate electricity power can reverse those electro-chemical processes so that the cycle can be repeated.

Our current lithium ferro phosphate (LFP) batteries are the third set of batteries we’ve used. The previous two battery banks were both made up of AGM lead acid batteries. For AGM batteries, typical lifespan is 500-1,000 cycles and four to eight years. Our two AGM battery banks lasted about six years. One of the claims of LFP batteries is greater longevity. Our Battleborn Batteries are rated at 3,000 to 5,000 discharge cycles so we are hoping they will last the rest of our ownership of Alpenglow. Ignoring the small discharge cycles accompanying the transition between shore power and engine alternator, we have had 261 discharge cycles of more than 50 Ah in years 2023 through 2025. Assuming our first full year, 2022, with the new batteries was similar, I estimate that we’ve had fewer than 400 significant discharge cycles so far.

While batteries can fail suddenly, normal “wear and tear” usually shows up as reduced capacity. Our LFP battery bank is rated at 500 amp-hours (Ah), but we have never intentionally consumed that much energy (more about an unintentional discharge later) before recharging. Searching the Internet for methods to determine a batteries capacity usually give solutions that are impractical for “working” batteries because they require they usually require the such the isolation of some or all the batteries from loads for a period of time followed by monitoring the battery while under a constant modest load.

The method I am using to assess the battery bank is monitoring the bank’s voltage over the course of many discharge cycles. While each discharge cycle will be of different duration, depth of discharge, and rate of discharge, I am hoping the aggregation of data will reveal patterns and trends.

In my dataset, for the cruising seasons 2023, 2024 and 2025, I identified 201 discharge cycles that started from a nearly full battery (<2 Ah of discharge) and resulted in a discharge of more than 50 Ah. I had 37,758 measurements of the battery during those discharge cycles. Each measurement included the battery bank’s voltage, state of charge, amp-hours discharged and rate of discharge. These measurements were the average of instantaneous values during the previous six minutes (five minutes in the case of the 2023 cruise year). The discharge cycle ended when the next charge cycle began. In most cases, the charge cycle was the main engine starting, but it could have been the generator starting or even the shore power breaker being switched on if I was purposely cycling the battery bank while at a dock.

Besides the aggregated data from three cruising seasons, I also have one case of the total discharge of the battery bank. This unfortunate and unintentional “test” occurred over about 4-1/2 days during the New Year’s holiday of 2023/2024. The shorepower to the boat was turned off and the battery bank depleted until the battery management system (BMS) incorporated in the individual batteries shut things down to prevent any further depletion. The boat sat for over 3 weeks with dead batteries until I returned to the boat near the end of January for my regular mid-winter visit. I restored shore power and the inverter-charger then recharged the battery. In a call to the battery manufacturer’s technical support, they told us that the accidental complete depletion ought not have any long-term impact on the battery.

Below is the graph of the house bank battery voltage versus the amp-hours discharged for that unfortunate incident. We had most things shut down so the typical discharge was only a little over 4 amps. The discharge curve is relatively flat, although it steepened at about 150 Ah discharged until about 180 Ah whereupon it flattened again, creating a slight “S” shape. A few bumps in the curve are when the diesel furnace came on to heat the boat (we keep the thermostats at 45°) and increased the discharge rate, The discharge curve steepened again at about 350 Ah discharged, then plummeted after about 470 Ah. The BMS turned off the batteries at 509 Ah after 4-1/2 days.

The next graph shows the aggregated discharge curves for the 201 discharge cycles in the cruising seasons of 2023, 2024, and 2025. The scales are different because the range of voltages and amp-hours is more constrained. I have also included the relevant portion of the 2023 total battery discharge as a reference curve since its low, steady discharge rate is closer to what a bench test might be.

Two things strike me about the curves. First, the steepening of the discharge curve (the “S” bend) at about 150 Ah is consistent across all discharge cycles and not an artifact of a particular cycle (e.g., the discharge rate increased because of a device being turned on). It must be inherent in the battery, either something to do with the BMS or the batteries construction (e.g., as the discharge continues additional/different internal portions of the battery are called into action). The second observation is that as the battery is aging, the discharge curve is lowering (i.e., the battery voltage is less for the same level of amp-hours discharged). In the flatter portions of the curve, we are only dealing with 10 millivolts (mV) or less. In the steeper portions of the “S” bend, it is much easier to see because it is closer to a 50 mV difference between the 2025 curve and the 2023 curve. The aggregate data at higher amp-hour discharge is noiser because I have fewer discharges going that deep and therefore the varying loads (e.g., furnace, inverter, Starlink) that come on or drop off can swing the voltage.

We are still very happy with the LFP battery bank we installed in 2021 and have no regrets. They make cruising more pleasant in not having to worry about managing loads or charging schedules. Compared to our previous AGM batteries they charge dramatically faster and reduce the time we must run the generator when we are spending multiple days at anchor. As to their longevity, from our previous experience with AGM batteries, I believe I would be noticing far more reduction in capacity in the AGM batteries at this point, four cruising seasons, than I’ve detected in the LFP batteries.

All of the measurements I’ve used in the analysis, battery voltage, state of charge, amp-hours discharged and rate discharge were obtained from our battery monitor, a Victron BMV-712 installed in 2021 as part of the LFP battery installation. The BMV-712 directly measures the battery voltage. The current going into or out of the battery bank is computed by the BMV-712 via measuring the tiny voltage drop across a 500A shunt (a known resistance) and applying Ohm’s Law. The BMV-712 computes the other values from tracking the current flows over time. At the time of installation, the BMV-712 was configured with the appropriate values for our battery bank.