The goal - to replace this:
Watermota SeaWolf, circa. 1972
Thoosa 7000HT, circa. 2020. Click link for more details and documentation.
(OK, without the dirt either...)
The motor is a Thoosa 7000HT, rated for 7kW continuous output, running on 48V DC. It is reversible, and capable of some "regenerative braking" when under sail.
The batteries are ReLiOn RBGC248V LiFeO4 (LiPo) golf cart batteries, rated 30AH at 48V, with integrated battery management electronics and CANbus interconnect. Two batteries in parallel is the minimum to run the motor at full speed.
For shore power, there is a GX4820 48V 20A industrial charger.
For solar power, there will be ten Solarparts semi-flexible panels, nominal 100W but actually rated 74W at normal operating temperature. There is a Smart Solar 150/70 MPPT charge controller with a remote display. The panels produce 16.5V at the maximum power point, the batteries need at least 55V to charge, so there will be two strings of five panels each, producing about 80V.
Sketch of motor installation (PDF)
Click thumbnails for larger image
Motor with Mk 1 temporary shaft extension
New instruments - motor and solar monitors
New throttle, keyswitch and monitor
First two batteries under cockpit sole
Solar charge controller
Motor, controller, circuit breakers
The starter motor on the original Watermotor jammed in 2020, but I was able to remove it, take it apart and free it up. This year it did the same thing, but I'd had enough - I removed the entire mess and temporarily fitted the electric motor while at the mooring - lowered the engine onto my little catamaran with the boom, took it to shore and dragged it up the beach with my truck.
My initial idea was to make a temporary propeller shaft extension, just good enough to let me motor out to sea, and sail to a boatyard. My first attempt worked - everything ran, I was able to move under electric power (at least, after removing a heavy weed growth from hull, propeller and rudder), but the plastic broke when I applied more throttle to counter a current. Fortunately, there was favourable wind so I sailed back to the mooring. The second temporary extension was in steel pipe, but I had some wobble which would probably have damaged the stern tube if I'd gone very far. The current plan is to get a complete new shaft and coupling fabricated, which will take at least a couple of weeks, then careen Alastor to fit the new shaft. Then I can proceed to the boatyard in a more normal fashion for haul-out and painting.
I tried to remove the propeller at the mooring, using SCUBA gear. I'd made up a custom prop puller, but nothing doing. So to plan B - remove the shaft together with propeller, then replace it temporarily with the original bronze shaft I still have from around 1990 to keep the water out. Doing that in situ would probably have worked, pushing the old shaft out from inside with the even older one, but it was easier and less risky to careen the boat and do it out of the water. I managed to finally get the prop off the shaft with the aid of a heat gun (bronze has a thermal expansion coefficient of 12 compared to 7 for steel) in conjunction with the prop puller.
Solar panels temporarily mounted on mainsheet slide. The final plan is to fit a second track alongside the mainsail track so I can hoist both at once, or at least not have them interfere with each other.
Mk II temporary shaft extension
Mk II motor mounts, replacing Mk 1 wooden ones
Careening Alastor in Horton Bay to swap propeller shafts.
The propeller puller
The original Watermota gearbox had a 1 inch shaft, same as the propeller, and a keyed flange secured with a nut (and really well jammed on, almost an interference fit). The propeller shaft had a mating keyed flange, secured against reverse thrust with a couple of bolts bearing on dimples in the shaft. Dafoe's new flanges are better engineered, with an alignment ring between the flanges, and secured with through bolts that engage with indents machined in the shaft. The new longer propeller shaft had the indents already in place, but they didn't want to touch the Thoosa shaft, saying it might be hardened or something. I managed to carve the indents with a file, and using a twist drill as a ream to get the right shape. I also lapped the flanges onto both shafts using a bit of polishing compound, as they were really stiff - I didn't want to have to bang them in place with a hammer.
Thoosa motor with new coupling flanges
Thoosa motor with indents filed in shaft
My mast hardware that had been on order for a couple of months arrived, so I was able to fit that. It was made to take countersunk bolts but I wanted to use pop-rivets, so I ground a twist drill flat and used that to machine the bolt holes to a cylindrical shape.
Solar panel track with temporary spar.
I contrived to thread a cable from the engine compartment under the portside bunk and up the mast, for the solar panels. There are two strings of five panels each, sharing a common negative wire. The two strings are in parallel, each with a separate circuit breaker. That gives a suitable voltage to feed the charge controller - it has to be somewhat more than the battery voltage (50V), but less than the maximum allowed (145 volts). Each panel gives 16V at maximum power point so that's 80V per string. This arrangement also lets me hoist half of the panels and still charge. When drilling the hole in the mast, I found I'd gone into an internal conduit or pipe that contains the cables for the masthead antenna and light. I hadn't realized that was there, but it worked out OK in the end. I was able to run a fish tape down and out the bottom of the mast, and pull a steel wire through, and then the actual cable, well lubricated with washing-up liquid.
Solar power cable in mast, with connectors for two strings.
Final arrangement of four batteries in parallel, 120Ah at 50V.
The batteries are lithium iron phosphate, with integral
battery management. There's a CANBUS connection daisy-chained
between the batteries to synchronize the BMUs and allow them to be
controlled from one button. I'm hoping to get a remote monitor
and switch - removing the cockpit hatch is hardly convenient.
(The batteries can be turned off, and the actual state of charge
monitored on a pair of LEDs, as opposed to the motor controller's
idea based on being told when the batteries are full and measuring
the current used).
The battery shelf is plywood, shaped to fit the hull aft, supported on wooden rails screwed to the plywood locker sides. The acrylic cover is mostly to avoid "bad things" were a metal object to fall across the terminals. (I believe the batteries have internal protection, unlike a lead-acid battery, but I'd rather not test that).
Setting up to align the motor. The weights and pulley are supposed to balance the unsupported weight of the prop shaft.
The Watermota had been leaking oil for some time (years). I hadn't realized quite how much. Most of the oil was in the bilge under the engine (the oil pan overflows with rain water that gets past the cockpit floor seal, and weepage from the stuffing box, carrying oil with it).
I've ordered material to fit a second track on the starboard side, so I'll be able to collect power sailing east as well as when sailing west.
The Great Battery Connection Question - my investigation into whether I had connected the batteries "wrong", or whether there was a better way.
I'd been thinking about the possibility of flying the two rows of solar panels facing in different directions, so that I'd get some sun from both port and starboard as the boat turns. I fitted a couple of blocking diodes in series with each string so that the panels on the sunny side won't discharge through the ones on the shaded side. I'm not entirely sure they are necessary, but they can't hurt.
Blocking diodes in the circuit breaker box. Also shown are the Anderson connectors on the motor controller.
Inside the motor controller - another motor controller
The extra connectors seem to be connected to the battery cables, but the other side of the shunt. That means that the monitor will know about current that flows through them.
I had had the solar charge controller connected directly to the battery bank. That worked, but the monitor didn't know the batteries were being charged. Clearly the thing to do was to connect the charger to the "charger" terminals, but I didn't know what the connector type was, though it looked vaguely familiar.
A Google reverse image search with the hint "connector" suggested that they were an Amp or Anderson connector. I ordered some, based on images in an online catalog, but didn't check the dimensions. They were much too small. More online searching found a blog from another Thoosa user in 2009 saying that they were 75 amp Anderson Powerpole connectors. I ordered a couple of pairs, but the catalog showed them in the wrong colours, so I have blue instead of black. Never mind, it works. Now with the solar controller connected that way, the monitor shows the batteries charging as well as discharging, and its "dead reckoning" scheme of estimating battery status has a chance of working. The other thing I needed to do was to configure the monitor with the correct amp-hour rating of the battery bank, which is 120Ah (30Ah each). I had previously set it to 60Ah when I had just 2 batteries, so it was reading low.
I'd been thinking about how to improve on the wooden spar (with steel eyebolts) as a support for the panels, when I spotted a telescopic boathook in the chandler's. I bought that with a view to hacking it up. After a few hiccups I ended up with something that I hope will survive a while in service. Instead of the length of cord securing the wooden spar to the track slider, I have a stainless shackle, with stainless loops pop-riveted to the aluminium. It will no longer telescope shut, but it will come in half for easier storage.
I've decided to replace the current halyard and bridle on the spar (the "head" of the "solar sail") with a throat halyard and a peak halyard (using gaff-rigged terminology). The peak halyard will be the current halyard to the top of the mast, while the throat halyard will go to a block at the top of the solar sail track. I'm waiting for more track material before getting my big ladder aboard.
Maiden electric voyage - Princess Margaret Marine Park on Portland Island
About 14nm each way. On the way out I had some sun but not much wind in the morning, setting out from Mayne Island, and hoisted the solar panels, motoring slowly so as not to use much more power than I was generating. In the afternoon the wind came up so I sailed for a while, anchoring off Portland Island. On the return trip, Alastor was finally re-launched about 4:30pm so I sailed to Portland Island again. The next day was raining with very little wind (or sun) so I motored essentially the whole way, at just under 2kt, arriving with something like 60% charge left. The tide in the Horton Bay narrows was against me so I cranked the throttle open a bit to get 2kt over ground, maybe 4kt through the water, for a few hundred metres. I was intending to do some speed trials to get a power curve, but it was too miserable in the rain.
While at the boatyard I took the opportunity to pressure-wash the interior of the engine compartment and bilges. 40 years worth of gunk, and there's still a bit of oil draining down from somewhere.
I also fitted a pair of Anderson connectors to the GX4820 charger, so I could plug that in to shore power and charge without taking the cockpit floor up.
Another issue that came up was that something in the motor controller, maybe the monitor itself, is showing a parasitic load of about 200mA when the controller is "off". That's almost 5Ahi, or 4%, per day. I found a battery switch in a secondhand store and fitted that in series, so I can disconnect the batteries instead of raising the cockpit floor and turning them off individually. One problem with that - if they aren't full, the monitor loses track of the charge state for its dead-reckoning.
Glueing new rear mounts.
I salvaged a couple of the big chunks of GRP I'd pulled off the old rear mounts. I then expoxied them to the hull, using the new rails as a guide. The idea was to glass over the joints for a bit more strength. Which I did. Unfortunately, I didn't use quite enough hardener for the temperature. Also it rained that night, and some water got in through the cockpit floor (not bolted down as I need access to the batteries). That wetted the new glass, so a couple of layers never set. I've pulled them off. The lower layers seem OK but I'm waiting for a bit dryer weather before sanding them and trying again.
I found the stern navigation light wasn't working. It seems that the engine block was used as a common connection point for a lot of 12V ground connections, not all going to the same point, so when I pulled the engine out I'd cut the 0V wire to several services. I thought I'd found them all, but must have missed the one to the stern - one of a number of black wires disappearing into the original black-taped wiring harness.