Nice and shiny

Well, the sanding is done and the varnishing is underway!  It is always a really satisfying process to start with the dull sanded finish and instantly see the final result.  I have two coats on and will apply a third later in the week.  Overall, I am happy so far.  It isn't perfect, but I'm getting better.  I only wish it wasn't at the very end when I am more impatient to get it back on the water.

I have ordered the final parts for the new gearbox, and I'll get hopefully make progress on this soon.  The weather has been perfect for boating, but I guess having nice weather for varnishing isn't a bad thing either.

Sanding at last

When I first had the new boat assembled last fall, I knew my warm days were numbered, so rather than miss another week doing the sanding and varnishing, I decided I would wait and do that in the spring.  Well, warm weather is here at last.  I have put the garage back into workshop mode, and I have even taken the catamaran out for a cruise since the new boat was disassembled again.

I have been able to use my fancy new sander.  I also decided to try and hook up our indoor vacuum cleaner as a dust collector, since the shop vac has an oversized hose.  It worked remarkably well, and I'm sure as long as I change the bag when I'm done, it won't be any worse off.

The finish quality should be another incremental improvement over previous boats.  At least I would say the sanding results are an improvement.  I will be varnishing when I return from a trip next week.  Someday I will hopefully get professional quality results, it is a good challenge.  Maybe someday I will even get a spray gun.  Until then my brushing and rolling will only get me so far.

I have also spent some additional time on the drivetrain.  My previous posts showed a few images of a new all bevel gear design.  I would still expect this to be an improvement, but I have moved towards yet another iteration.  This one is a hybrid chain and bevel gear system, but it is still housed in a similar sized box.  The crank arms attach to a first shaft with large ANSI 35 sprocket.  An intermediate shaft has both a small sprocket and a large bevel gear.  This has the advantage of a higher gear ratio, I should hit 1:6.5 with this design.  It also uses a 1:2 bevel gear set which is significantly cheaper than a 1:4 set (enough that the cost of the sprockets is still cheaper overall).  Lastly, this means that the torque on the bevel gears will already be reduced by the chain gear ratio, so this will improve gear life.  With the lower torque, I will likely test a slightly smaller diameter shaft tube (5/8"), and I am eager to test a prop with carbon fiber blades made by Rick Willoughby that has been sized for this approximate gear ratio.

My budget will hopefully be stretched further by doing a greater amount of machining myself.  I have found a local "makers" organization with a shared workshop and inexpensive dues.  This should allow me to do most of the milling myself.  I have found a space with power for my welder, so if I brush up with a little practice, I can do this as well.

Cabin fever and new designs

The lakes might be frozen here for the next few months, but at least the holidays have given me some time to kick around some ideas I might be able to try in the spring.  I get to wait and see how much is in the piggy bank once I'm ready.  I've also been able to watch what others have been up to.  Rick and Mike have both tried new ideas and have had positive results.

I don't expect to do any major work on the hull.  My focus is improving the gearbox stiffness and maximum torque and increasing the drive leg durability with a stronger chain.  I have some new designs worked out in CAD, and I'll think it over for a few weeks before doing anything in case I change my mind.

My original gearbox used a 4" x 6" rectangular aluminum extrusion.  This was oriented so the top and bottom were open and had cover plates.  Under load, the only features to prevent the sides from flexing out were the cover plate fasteners, and clearly this wasn't sufficient.  The new design uses a 4" x 4" extrusion with the fore and aft sides open.  The aft side with the output shaft will have an extruded section welded on as a cover plate and and also serving as a mounting area for the shaft bearings.

The second CAD image shows a cross section with the top cut away.  This shows two additional machine screws positioned as close to the gear interface as possible to provide additional reinforcement against flexing.  The final forward side is covered with a third piece of extrusion sized to nest inside the main housing.  This cover has a cutout for the large gear and is screwed to the housing near the crank shaft spindle.  This should effectively create reinforcing ribs on the side opposite the output shaft.  The crank shaft is also slightly larger diameter and is available as a ground high precision rod, so that I can more easily achieve a tight clearance with the gear.

I am working on new designs for the drive leg as well, although this is not as critical.  I considered various options of using a double chain loop or upsizing a ANSI 35 chain.  However, potentially the easiest option to increase durability is to create a sealed assembly so the chain can be plain steel rather than stainless steel.  The stainless steel chain has a significantly lower strength, but I had chosen it to simplify the design and because the miles I put on the boat shouldn't wear it out anytime soon even if it runs in water not oil.  Production drive legs such as the Wavewalker, use a mineral oil bath to lubricate parts and help prevent water inflow.


The new leg design uses two standard bicycle bottom brackets with machined end caps to house shaft bearings.  I can add o-ring grooves and o-rings to the shafts to seal against the bearing ID.  The top and bottom each have standard tubing welded at right angles and these nest similar as before.  In addition, I can now order steel chain in a continuous riveted loop (no connector links) because the bottom bracket shells are large enough to slide the chain loop on over the sprocket.  The current leg was too tight to allow this and required a connector link.  The continuous loop is rated at approx. 170 lb. working load and this should be more than sufficient.

Seasons changing

Everything is back up and running, and I've notched a few more successful outings.  I had my first windy trial this week.  It's a small lake so waves were not bad, but it was maybe 25 mph gusts.  After getting so used a catamaran, I admit it was a little unnerving at first to feel the wind rock the boat.  By the end of the day, I was much more comfortable.  It will still take some getting used to, especially when I do hit some bigger swells.

All things considered, I would have to say the project has been quite a success.  There are a few bugs to work on, but I've had fun on the times I've been on the water.  Hopefully I don't make anything worse when I take it apart to sand and varnish in the spring.  I'll work on some of the bugs before I put it back together.

The first priority will be the gearbox.  It is currently fine for cruising, but not for anything needing more power.  I suppose I am glad I opted for less expensive straight bevel gears rather than spiral gears.  If the box isn't stiff enough, the more expensive gears would have met a similar fate: worn, rounded teeth.  I have some thoughts on reinforcing the housing with cross pieces and also cover plates with a lip on the edges to help prevent pedaling forces from flexing the housing.  If there is an improvement, maybe then I will upgrade to spiral gears.

Next up is the prop.  There is a yet unexplained difference in speed at a given blade pitch compared to calculations, so I have had to modify the blades to make things work well enough.  I suspect my first attempt will simply be to remake the blades and attempt a more accurate foil shape when grinding.

I might also go ahead and replace the full shaft, I'm not sure.  It isn't that expensive, so it might be nice to try one that wasn't accidentally bent before assembly.  I'm still undecided about the drive leg.  There are several options to increase the strength of the chain loop.  The current setup can handle more than the gearbox, and I think it would be fine for distance race speed, just not short full sprints.  I might also try a leg fairing that slides over the drive leg tube rather than move with it.  The steerable prop is nice and it works well, but I think it limits my options on making the chain loop stronger.  I doubt I will decide until spring.

On the boat side, I don't think there is much I would change.  During the design I wasn't sure how far I would place the gearbox from the seat to fit my height.  My other boat has a profiled recumbent seat and this one has a sharp angle, so comparing the two was difficult.  Now that it is set up, I can see I could have shortened this about 3 inches and decreased the windage of the gearbox support.  There might be a few other changes to reduce windage, but it all amounts to fine tuning.  I have modified the stabilizer mount: reattaching the two halves and I also added a cross pin between the vertical pivots to reduce flexing.  It has made a big difference, so for now I will leave this alone.  I really like the vertical adjustment and the ease of setup when launching.  I imagine starting over I could shave off some weight, but it would again be minor changes.  As I mentioned before, in a new boat I would probably make the stabilizers out of foam sheet and also reconsider the weight of cloth in some areas.  It might all amount to about 5 lbs.

Last but not least, repurposing the trailer for the cat works OK, but I will eventually make or buy a better setup.  Ideally something small I could stow on the boat, so I wouldn't even need to walk back up the shore when launching.

In the mean time, the garage needs to be turned back into a garage.  I don't think I would get much support for my hobbies if it meant scraping frost off the car windows.  I've already put the cat up in the rafters and cleaned up my tools and mess.  Hopefully I can get out on the water again before the cold water deters me, but there is a chance this is it.  At least the outing this week was a nice sunny (but windy) day with fall colors.

Hopefully some of the people reading have a nice warm shop or are just getting ready for summer!  If you have a project, it would be great to see it.  Turns out this blog stuff is pretty easy...

Hydrobowl outing

I've been reading about Hydrobowls for years and was hoping I would finish in time to make it to the event about 6 hours from Minneapolis.  It is a multi-discipline race with overall winners based on points over 5 races.  It is intended to be a measure of overall performance, and I would say it does a reasonably good job at it.  The races are a 100m flying start sprint, a slalom, a 100m drag race from standstill, a 2k criterium, and a static thrust test.  I wish that there was a race where a fast boat could go fast over a distance without the need for agility, but I suppose this is because I found that my boat was slowest cornering boat of the group!

 It is an informal competition, but several of the racers have been going for years, or decades, and have tailored boats to the event and refined over the years.  My boat held it's own, but the drive definitely suffered some wear and tear.  Before the start I had adjusted the angle of the prop blades to increase the pitch - the distance you are supposed to go each rotation.  I ended up with a better feel on the sprint, but the long distance events had me wishing for a faster cadence.  I suspect I will go back to somewhere in the middle.  The final points totals are pending, but I was somewhere in the middle.

The course is a 100m set of buoys, with two parallel rows.  The rows are about 20 meters apart.  The slalom weaves between the 5 or so intermediate buoys along the 100m, but you have to span the 20 meters each time.  The 2k criterium is 8 1/2 laps around the buoys.  This basically means you need a boat that can corner with a 10m turning radius to have any success on these 2 of the 5 events.  I would say my boat is more like 20-25m radius.  This means the slalom needed some reverse thrust with a paddle to make the turns, and the criterium for me was a lot longer than for everyone else because my oval was about twice as wide as the buoys.  My boat turns just fine for cruising a lake, but the 20 degree limit on the rudder rotation means it can't compete on agility.

The drive showed some signs of trouble.  I had anticipated both, but there is no good way to know for sure until you just try it.  The gearbox skipped teeth in the gears while sprinting.  This could be both due to the large size of the housing and the ability of it to flex, but I suspect it is also a result of the bearing bores coming back a little oversize from the shop.  The shop ran the reamer at a high RPM and this caused vibrations and a hole that was a little too big.  I tried to close it a little by peening edges with a punch to close the opening a little.  But it is still not a great fit and allows the output shaft too much play.

Also, I have been using the 25 pitch chain for years without issue on my cat, even after sprints, but this older drive is geared a little higher and also uses 12 tooth sprockets on the prop shaft rather than 9 tooth for the new boat.  It doesn't seem like a huge difference, but it means the chain has about 25% more tension during the same pedal force.  Given the demands when sprinting, this pushed the boundaries a little too much and the chain started to stretch and get too loose and could get jammed up.  Before I could adjust it properly I limped through the slalom and drag race.  I wasn't too surprised; I had debated sticking to 12 tooth for a while.  I would say the drive would manage just fine for years under normal use, just not all out sprints.

Hard to say where I will head from here in terms of design.  The Hydrobowl's future is uncertain since the organizer moved away from the area.  It is a fun challenge to tackle an all around boat for these events, but it is hard to say if the events will happen again soon (unless I host one in Minneapolis!).  Otherwise I could throw on a new chain and cruise the lake quite content with the way thing are.

This is with the caveat of fixing the stabilizer mounts.  As I mentioned, I was concerned about the stiffness after splitting the bracket into two halves.  This was a well grounded fear.  I've found that the vertical supports behind the seat will rotate slightly and also flex the bracket such that the long supports angle 5-10 degrees (estimating) towards the stern.  This means the stabilizer drags at an angle and plows the water, causing more drag and more flexing.  During the drag race I crossed a wake and this set off this feedback loop until I almost went for a swim.  I will definitely modify this.  I am very happy with the vertical adjustment part, but I will join the halves into one.  I will also remove the pivoting pieces from the bracket and use a short horizontal double ended nesting tube on the same bracket that connects to both long arms.  This will eliminate the flexing.  In order to achieve my goal of self contained stabilizers for walking to the lake, I think I will add separate short nesting tubes pointing towards the stern and I can remove the arms from the "in use" tubes and click them onto the "transport" tubes.  My timeline for this is uncertain.  Fall approaches and it is time to build up a good store of outdoor memories to last through the winter.  Fortunately the overall boat is quite sound.  I can work on the small stuff like mounting arms and gear boxes in my basement over winter.  I'm not sure I will, but at least I could.  Come spring, I will get the boat sanded and varnished and actually "finished".  Hopefully I will post some more nice cruising photos this month, and then I'll have to wait and see.

Maiden voyage!

I had my first trials yesterday and today.  Generally, they were successful.  I had a few bugs to work out with the steering cords and chain tension, but I was on the water doing nice laps after a few false starts.  A priority will be grinding the prop to add the lift from a proper profile.  Also, I may adjust the pitch as I was spinning out on the pedals before getting to max effort.  A future priority will be sealing off a greater portion of the stern hull.  You will see the stern sits low due to water that enters a fairly large drive leg stern partition through the drive leg opening.  This will be an easy modification once I take the time to work on it.  It's been a long day, so I apologize for a lack of a storyline on the trials.  This will have to be a photo post for now.

The rest of the drivetrain

I can say now that the scope of my summer project has been changed slightly.  The boat is ready to hit the water (it already has actually, but I'm behind on posts), but it is not finished.  I have yet to do any final sanding on the hull; and no varnish yet, either.  I will do a few more things to prepare for the Hydrobowl this coming weekend, but I will focus on the prop and steering.  Varnish will probably wait until spring.  Right now I want to enjoy it for a few weeks before the weather turns.  And winters are long here...

The final area of the boat I have not focused on in posts is the rear end of the drive.  It is a variation of the flex shaft design Rick has refined lately.  I like a lot of advantages of the "standard" flex shaft, but I have had an idea in my head for a few years, and I really wanted to see if it would work.  Now that it is mostly done, it is easier to explain.  I am using a flexible shaft, for the full length of the boat, but it does not extend below the hull as in Rick's system.  I am running the shaft under the seat and along the deck line to the stern where it joins with a separate drive leg.  The drive leg uses a 1/4" pitch chain common on twisted chain drives, but the chain does not need to twist since the gearbox has already changed the pedal motion 90 degrees.

After a lot of searching I was able to find inexpensive stainless 9 tooth sprockets for this chain.  The max shaft size was 1/4" but I wanted to drill them out for 3/8" shafts.  This meant there wasn't enough material left in the hub for a cross pin, so I found a local bicycle fabricator who silver soldered the sprocket on, and it looked pretty secure.

I realize now that my photo collection is incomplete for the drive, but I can add photos later if there are any requests.  The shafts are mounted in stainless bearing and I leave it all open to water.  This is the same approach I used on my first boat, it is 7 years old now on the original parts with no corrosion on the stainless.  Good enough for me.  The chain runs through a 1 1/4" OD aluminum tube.  This is the smallest possible size drive leg I could figure out how to make.  The round tube is welded to a 1 1/4 rectangular tube which houses the prop shaft and bearing.

The drive leg tube has a foil fairing based on precise plywood pieces cut from the CNC batch.  I sandwiched foam and wood to maintain and accurate shape.  I precut a (slightly undersized) hole in the wood pieces, and once this was all together, I cut it into front and back halves along this hole centerline.  I then wrapped sandpaper around scrap tubing and sanded away the foam from the inside.  This allowed me to glue the fairing to the drive leg and essentially not add any width to the tube except for a layer of glass cloth.

 One of the key reasons this general design appealed to me is that the prop is steerable.  The drive leg tube nests inside a square tubing with flange bushings.  The drive leg tube has a slot for the upper drive shaft, and the clearance in the slot allows for about 20 degrees of rotation each direction.  It can be increased slightly if needed.  The square tubing is bolted to the inside of the hull.

 One of the primary concerns with drive legs is weed catching and collisions.  Generally, these are not issues where I will be boating for the near future, but I added a small strut to help deflect weeds and ramp over obstacles if I hit something.  I will eventually file this strut to a nice profile.  The twist isn't ideal because it will add slight drag, but it is the fastest approach for now.  The bottom of the strut is bolted with a shoulder bolt to the drive leg bottom so the leg can pivot about the strut.

There are plenty of downsides to this design relative to using only a flex shaft under water, and I was aware of them and willing to try it anyway.  I accept it is not an ideal solution for most home brew boats.  But there are a few advantages, and I will focus on these.  Because the shaft only has a small amount of curve, I can use 3/4" diameter (.058" wall) tubing the full distance, and maintain the best possible torsional stiffness with a minimum of weight.  I calculate the shaft at about 44 Nm/rad stiffness, significantly higher than cases where a thinner shaft is used underwater to accommodate the increased bending.  There is no need for a rudder of any kind, and the strut fairing can be more effective since it is not surface piercing.  This is in addition to the steering from prop thrust.

Beaching is more of a gray area.  The strut is currently fixed so it will hit when beaching.  The lake I am on is deep, so this is not a concern for me.  The downside is that the strut is fixed, the upside is that the shaft is protected so beaching will not bend the shaft, which can happen when not using spring steel.  In fact, the second reason this design was appealing was that I intended to make the drive leg beach-able.  My idea was to detach the stern portion of the hull and add a transom.  Then I could use the deck panel or a portion of it to flex up with the shaft.  It would essentially be a long hinge flexing/pivoting with the shaft.  Since the shaft does not flex a large degree in steady state use, it could accommodate more flex with beaching.

In the end I decided this would take more time, and I would prefer to proof the general concept and reliability before I cut into the boat.  If I decide not to continue with this drive leg it is currently easy to undo.  If it works out for a year or two, I can still use it as a test bed for a steerable and beach-able flex shaft drive leg.

The complete drivetrain is very low friction.  I can easily rotate the shaft with my fingers and maintain motion for about a prop rotation; however, I managed to bend the shaft slightly during my experimenting and it has a slight pulsing as it rotates due to this bend, so it always stops at the high resistance point when I try to spin it.  It is not something I can feel with my legs.  I replaced my original bushings with ball bearings since they are protected from water.  I would consider it relatively light weight as well.  The drive leg adds weight but there is no rudder and the shaft has only 12 inches of steel.  In the end, the full assembly with full shaft, shaft bearings, drive leg and rudder control is just under 4 lbs.