So I went back to square 2 and decided to remake the original brass valve body in lighter ABS plastic using our Makerbot 3D printer.  The beta version looked like this:

Remade underwater solenoid valve with 3D-printed structure.

I re-used the gasket, screws, core, and shaft from the original valve (where the solenoid sits), and replaced the big brass parts with light ABS plastic that is customized to fit onto my hull.  The steel shaft press-fits onto the shaft holder at the top of the photo.  The shaft holder screws into the connector piece, which implements the same idea as the brass valve piece, but instead of the input being a pipe fitting the input is open to the inside of the robot hull.  Note that in the photo I’m holding the valve upside-down, the solenoid shaft will hang down from the top of the robot hull, and the other end of the valve will bolt onto the hull’s top using the existing screw holes, pressing the base of the conical shaft against the hull’s top hole (apparently I’m a terrible photographer too).  When the valve opens, the air inside the hull will flow through the conical shaft out of the hull’s top hole and will be replaced by water flowing in through the bottom hole.

The new valve did in fact work.  The solenoid successfully pulled back the gasket to open the valve when activated, and it almost sealed when the power was disconnected.  The seal wasn’t perfect because it printed over a bump in the 3D printer’s build platform.  The plastic was 277 grams less massive than the original brass, and with a tweak or two the new valve will bolt onto the hull efficiently.

The valve had a few problems.  The steel shaft’s wide screw part was too wide for the hole it was supposed to screw into, so I had to file the hole out until I could press-fit the shaft in.  The support structures on the interface piece are too spindly, as you can see in the photo.  Also the square plate on the interface piece was a little thin, and it sat a little too short.  The corners  of the square hit other structures inside the robot hull, and it needed to be about 1 mm farther away from the nut holders.  In the new version I have tried to rectify these issues:

Updated valve connector structure.

The square plate is now half again as thick, and the nut holders (the bits sticking out near the top of the conical shaft) are 1 mm farther away from the plate.  The supports are a lot thicker, too: they’re 4 mm wide instead of 2 mm, and the six supports that don’t connect to the nut holders flare outward stylishly.

There’s one other issue that I foresee.  The heavy copper solenoid and the steel shaft/magnetic core sit pretty high in the hull, and raise the robot’s centre of mass.  This is a potential problem because if the centre of mass is too high the robot will roll too much in the water.  I can add extra ballast to lower the centre of mass, which may or may not be fine, depending on what kind of buoyancy the rest of the hull has.  I might actually have to do some engineering to figure that out.  Or, I can try it and see what happens.

Pictures and video follow:

A semidemihemisphere with overhang supports attached (but not the walls that the cutaway rests on).

Cutting away an overhang support.

The full AUV hull. Neat!

Here are two videos of a semidemihemisphere being printed. Part 1:

Part 2 (the cool part):

Tweaked version of the AUV model.

I made a few interesting changes to the cutaway portion though:

The cutaway is needed to hold up the hull while it prints.  Without the cutaway, the plastic, heated to its melting point by the Makerbot, will droop and hang or fall.  The cutaway gives the hull something to sit on as the plastic cools.  Here is the cutaway portion, viewed from the side:

AUV hull cutaway pieces from the side.

A – I added a bunch of little bits on the bottom layer of the model.  This is a hack to get the printer to print a larger raft.  (The raft is a thick layer of plastic that the model sits on as it prints, intended to let the model stick well to the build platform.)  I had to reduce the raft margin so that it would fit on the build platform, but with the smaller raft the pieces were susceptible to being knocked around by the toolhead as it goes.  These pieces get raft added around them, which merges with the raft of the actual model so that the real pieces get a nice stable raft.

B – This bit connects the cutaway material to the hull.  The hull is getting pretty tall at this point, so if it gets hit by the toolhead there’s a force applied that is strong enough to knock around the hull and perhaps misalign the layer.  This part (and its equivalent on the other side of the model) help hold the hull steady against the cutaway shells as they get close to being joined together.

C – This bar connects the cutaway shells together to keep them steady.  In a previous model (shown in the last log entry) I made this part too wide, and it took so long to print the horizontal bar that the hull cooled too much and split a little.  It wasn’t anything a little plastic welder couldn’t fix, but the supports don’t need to be very wide to fulfill their purpose.  After shrinking them to 1 mm wide the printer could print one bar layer with just two quick strands of plastic.

AUV hull cutaway pieces from the top.

D – This is the support for the upper tab that hangs freely over the semidemihemisphere.  You can see it from the side at the top of the cutaway side image above.  This is a pretty cool piece that’s anchored vertically on the cutaway shell, and ends up a horizontal flat surface at the upper tab bottom.  It is topped by a 0.5 mm shell that the tab sits on (see E).  I did this piece with a loft from a rectangle on the vertical plane to a rectangle on the horizontal plane.

E – The lattice pattern holds up the hull’s overhang.  The lattice rests on a 2-layer thick platform.  The bottom layer of the platform is printed over open space, and droops badly.  The second layer of the platform is printed on top of that and is pretty flat.  Then the lattice is printed on the nice flat layer, and the actual hull is printed over the lattice.  The lattice is sparse enough that it can be separated from the hull easily with a sharp utility knife (same as the shell in D).  The squares in the lattice are 5 mm to a side (I originally used a 10 mm lattice, but it was too sparse and the hull overhang drooped too much) and 0.5 mm thick, i.e. the width of one strand of plastic.  Note that the top-most (innermost) hull overhang is narrow enough that it doesn’t need to be supported by a lattice.

F – The circular patterns support the circular holes in the hull.  Without these the Makerbot would try to print a circle over the lattice, and it would just collapse into the open spaces.  The square support on the upper lattice supports the receptacle that accepts the upper assembly tab.  Without it, the printer would again try to print the square over the lattice and the first layer would fall into the holes.  With the support the first hull overhang layer is a full surface, and the receptacle square sits on that full layer instead of the lattice.

Here is an old version of the cutaway that had a bunch of problems:

An early, problematic cutaway version.

Problem 1: The bottom layer had these wedges to hold everything together and provide a solid base.  After I started using a raft these were no longer necessary; in fact, they were a detriment because they took a long time to print and the first couple hull layers cooled too much and came apart.

Problem 2: As stated before (C) the cutaway support bar is too wide and takes too long to print.

Problem 3: The lattice is radial instead of square.  The printer couldn’t print the off-angle 0.5 mm lines.  The lattice would be printed near the edge of the lattice pattern, but there wouldn’t be anything in the middle, and the hull overhang would not get its support.

Problem 4: The circle supports in the lattice are too thin (0.5 mm) and they don’t get printed properly.

Problem 5: The innermost cutaway support is thicker than the others.  Once again, there was no benefit to this and it took significantly longer to print than the thinner version in the later versions.

Here is my current design in full, including the support structures:

The 3D model for my spherical AUV hull.

This is one eighth of the hull; I call it a semidemihemisphere.  Eight of these will make a sphere 18 cm in diameter.  I split it up into eighths because I’m using a Makerbot 3D printer and its build platform is limited to a cube about 10 cm to a side.  Also, this way each piece can be identical (or nearly so).

One limitation in the Makerbot is that it can’t print overhangs very well, so I had to add some support structures that can be cut away.  In this entry I will ignore that and focus on the hull design:

My design requirement was simple: I needed a large, spherical hull with a hole in the top and bottom for buoyancy control.  As I said previously, in order to make it big enough I had to split the sphere into eighths.  The eight pieces needed to be bolted together securely, but I wanted to avoid having a lot of nut and bolt holes on the outside of the hull.  (You’ll see that there are several outer holes in the current design, but I have ideas to remove them all that I might implement in a future version of the hull).

The sphere is composed of an upper hemisphere and a lower hemisphere, each consisting of four more-or-less identical pieces like this:

AUV semidemihemisphere hull component.

These two hemispheres are assembled using connectors inside the hull, and then the hemispheres are bolted together from outside the hull.

The upper assembly looks like this:

AUV hull upper assembly.

The tab on the right fits into the matching receptacle on the left to connect neighbouring pieces in a hemisphere.  In my test prints of the upper part of the semidemihemisphere, two pieces snapped together quite securely.  The circular hole in the middle fits a #4 bolt that will connect the valve to the top hemisphere.  It will only be used on two of the semidemihemispheres; I’ll delete it from the model file for the other units so that I don’t have to plug it.

Just below the upper assembly you can see the inner component mounting points:

AUV hull inner component mounting points.

These can be used to mount components inside the AUV, e.g. a microcontroller or a sensor/control module board.  The slots are sized to fit a #4 nut, so that the mounting board can be bolted on easily from below, without having to hold the nut (many of the nuts on the Makerbot itself are held in this manner). A mounting board will be bolted to one mounting point on two neighbouring semidemihemispheres, which will handily hold the semidemihemispheres together.  It might be worth connecting neighbouring mounting points together even if nothing is mounted on them, just to reduce stress in the upper assembly tab.

The flat, right-angled circular protrusions inside the hull are actually an artifact of the support structure needed to print this thing on the Makerbot.  This nod to the machine’s limitations turned out to be quite useful; I’m not sure what the best solution would be if I got this manufactured commercially.

Down at the bottom of the hull you can see two more square connectors to join the lower parts of the four neighbouring units that make up a hemisphere:

AUV inner hemisphere connectors (right).

One side has a hex nut inset, again so that the nut doesn’t need to be held when tightening the bolt.  Two neighbouring semidemihemispheres can be joined at the bottom with a 1/2″ #4 bolt using these inner connectors.

Once the two hemispheres have been assembled, they are connected using the outer joins:

AUV outer hull joins.

The left connector accepts a #4 bolt, and the right connector accepts a #4 nut.  By matching one hemisphere’s bolt holes to the other’s nut holes, the two hemispheres can be joined together.  This is obviously a major flaw in the outside of the hull.  I currently plan to print plugs that can be inserted into holes to smooth out the hull surface.  This is one thing I’d like to improve in the future, using a technique similar to the protruding tab in the upper assembly.  I couldn’t use the same trick here because the bottom surface needs to be flat to sit on the Makerbot build platform while it prints, but a similar piece printed separately and glued in should be quite effective.  I decided not to do that in this version of the hull, because I wanted to get the heck on with things and it’s not that important.

One problem is that when the semidemihemisphere is printed the bottom is not flat.  It gets curled up, the bottom layer sticks to the raft interface, and so on.  I currently plan to sand the bottom layer down and replace it with a cheap-o gasket made out of silicone rubber.