Nautilus Drydocks ALVIN (3D printed)

[Subculture]

Unless the cylinder is pressurised by the ballast system, no retainers are needed. If you do need a retainer an alternative would be to use the equipment tray and incorporate a bayonet latch. The tray would need to be made from something strong enough to cope with the tensile forces, wouldn't like to rely on FDM printing for that part without some reinforcement.

[gboxwood]

My dear friends, I haven’t posted in a while but I have been busy with the ALVIN nonetheless.
Let me try to walk you through my process up until this point. There has been a lot of:

• Making up my mind
• Waiting for components to arrive
• Modeling those components
• Assembling as much as possible in CAD to get a feel for the best layout of the components
• Testing on the bench
• Assembling for real

A few surprises along the way but in principle, an enjoyable process.

Making up my mind

I decided to follow the advise to drop the vertical thrusters. I’m glad I did it because I would have not been able to fit them (somehow).
On the other hand, I decided to make this a little bit more complicated than required so I could experiment with solutions and enjoy the boat to the fullest (it does not offer a lot of driving options without any dive planes and additional thrusters).
Therefore I incorporated 3 solenoid valves as per scheme below (pretty much along the lines of my initial sketches about diving/surfacing strategy):

ALVIN-DES-001 by Giacomo Bosso, su Flickr

CAD modeling and components layout

I started with the 2 WTC’s that will make up this system and increased the level of detail as I received the components. I claimed two WTC’s 300mm long to begin with and I can state that the electronic equipment WTC is packed at the moment. The water ballast tank can be shorter (more on this later) but I will leave it at 300mm for the time being (it is easier to shorten it than to make it longer…).
The current status of the 3D model is shown below and is WIP as it gets regularly updated as I move forward with the actual build and validate or update the design as required.

ALVIN-030 by Giacomo Bosso, su Flickr

[gboxwood]

The electronics WTC

This is how the electronics WTC looks like in the CAD:

ALVIN-023 by Giacomo Bosso, su Flickr
ALVIN-024 by Giacomo Bosso, su Flickr
ALVIN-026 by Giacomo Bosso, su Flickr
ALVIN-027 by Giacomo Bosso, su Flickr

[gboxwood]

It develops itself above-and-below a horizontal tray with a certain strategy for which components go above and which ones go below (the more water-friendly ones, in case the cylinder takes water).
ALVIN-028 by Giacomo Bosso, su Flickr

The heart of it all is the motor and water pump combo which required a very integrated mount due to the size of the water pump itself.

ALVIN-029 by Giacomo Bosso, su Flickr
The solenoid valves are also a strategical set of components and they are mounted using a dedicated support.

Senza titolo by Giacomo Bosso, su Flickr

Senza titolo by Giacomo Bosso, su Flickr

[gboxwood]

You can see each solenoid valve has its own electronic switch:

Senza titolo by Giacomo Bosso, su Flickr

Senza titolo by Giacomo Bosso, su Flickr

This is a neat unit I’m using throughout this build (also for the lights) and it wires like this:

ALVIN-DES-002 by Giacomo Bosso, su Flickr

[gboxwood]

Familiarizing myself with the electronics (testing on the bench)

Wireless switch

It was a totally new component for me and I had to figure it out first. It turned out to be a great device and I rewired it to my liking so I did not have to use the flimsy terminal blocks it came with.

Senza titolo by Giacomo Bosso, su Flickr

Senza titolo by Giacomo Bosso, su Flickr

The motor

It is a geared motor with a 140 rpm output at full throttle. I guessed that given the propeller size (120mm) torque was more important than rotational speed… I will see if it works or if it is too slow (I don’t expect to race the ALVIN but I have no sensitivity about motor speeds for these applications). On the bench, I tested it with the manual speed controller it came with.

Senza titolo by Giacomo Bosso, su Flickr

The ESC

I bought 2 cheap ESC’s for both the motor and the pump. I am not happy with them and I ordered Mtroniks’ Viper Marine as a replacement. I could not program them (teach them the endpoints) and as a result, the reverse is too slow to be useful (more important with the pump).

Senza titolo by Giacomo Bosso, su Flickr

[gboxwood]

The solenoid valves

I bought many of them to gather some understanding about their size and robustness, I eventually chose a mid-sized model which feels solid and takes up some space. I have smaller ones for other projects and I will learn as I go.

The lights

I envision two set of lights, the 4 main flood lights and the surface/navigation lights (red green and 2 sail lights, maybe the strobe?).
For the main flood light I wired 4 high power white LED (1W) in series so that I can plug them directly into the 12 volts supply voltage.

Senza titolo by Giacomo Bosso, su Flickr
Everything got bench tested to make sure the principle of operation was correct:

Senza titolo by Giacomo Bosso, su Flickr

[gboxwood]

Assembling the electronics cylinder

The tray is cut out of birch plywood, not the best choice but a good starting point.

Senza titolo by Giacomo Bosso, su Flickr

The solenoid valves block goes in first:

Senza titolo by Giacomo Bosso, su Flickr

Senza titolo by Giacomo Bosso, su Flickr

All the rest pretty much follows until you are offered with a tight mess of components and cables running all around…

Senza titolo by Giacomo Bosso, su Flickr

Senza titolo by Giacomo Bosso, su Flickr

[gboxwood]

Senza titolo by Giacomo Bosso, su Flickr

Senza titolo by Giacomo Bosso, su Flickr

Senza titolo by Giacomo Bosso, su Flickr

[gboxwood]

The pictures from below show an additional support “arc” I added later on to prevent bending of the plywood tray at its weakest point.
The assembled tray with components fits once slid into the cylinder but I see margin for improvement here and there and I’m still surprised by how tight it ended up being (I really thought the 300mm was plenty of room to deal with).

Senza titolo by Giacomo Bosso, su Flickr
Senza titolo by Giacomo Bosso, su Flickr
Senza titolo by Giacomo Bosso, su Flickr
Senza titolo by Giacomo Bosso, su Flickr

[gboxwood]

The front endcap is dedicated to the 4 pairs of inlets and outlets for the pump and 3 solenoid valves.

Senza titolo by Giacomo Bosso, su Flickr
Senza titolo by Giacomo Bosso, su Flickr
The rear endcap is dedicated to the electrical connections and to the drive shaft dry seal (not there yet).
As for the electrical connections, I used 2 waterproof connectors, one being a 2-pins for the main voltage and the other one being a 4-pins for the lights (2 wires for the main flood lights, 2 wires for the nav/surface lights).
I could have used a 6-pins connector but the more the pins, the harder it is to get it in and out.
I will also provide a 3-pins connection for the rudder servo pretty much along the lines of what is advised for in Bob’s book.

Senza titolo by Giacomo Bosso, su Flickr
Senza titolo by Giacomo Bosso, su Flickr

The big hole means there will be an additional piece added to it with the dry seal integrated in it. I need to find a dry seal for a 5mm shaft… any advise?

Senza titolo by Giacomo Bosso, su Flickr

Things I learned after this first WTC I did:

• Space is at a premium (I was expecting it but it was still a surprise given the size of this boat).
• Connectors are bad, even the small ones take up a lot of space, better to skip them entirely if possible/practical.
• Keep cable slack to a minimum.
• Fix everything down as much as possible, in particular the antennas (the risk is high to pinch them while sliding the cylinder on…).
• Overall… an interesting build which is delivering more challenges than I expected (in total honesty).

[gboxwood]

About the ballast tank volume

I took the advice from earlier posts and looked at the amount of weight that the tank shall sink.

From some online pictures I figured the waterline and split the parts accordingly:

ALVIN-014 by Giacomo Bosso, su Flickr
ALVIN-019 by Giacomo Bosso, su Flickr
ALVIN-020 by Giacomo Bosso, su Flickr

[gboxwood]

Then I sliced them with no supports to get their weight (2 examples only, more parts are involved, like the sail):
ALVIN-021 by Giacomo Bosso, su Flickr
ALVIN-022 by Giacomo Bosso, su Flickr

A spreadsheet would then advise for the minimum volume of the ballast tank:

ALVIN-DES-004 by Giacomo Bosso, su Flickr

It can be seen that I could get away with a ballast tank as short as 150 mm… something I will definitely take advantage of after the first tests.

Please also note that I plan to use a snorkel to vent the ballast tank, so I will be able to use all its volume. Again, testing will be required to learn and rule out some uncertainties (but it appears to be plenty of margin to make this work).

[rwtdiver]

gboxwood,

WOW! Some real excellent know how here. It's obvious you have some real knowledge and skills with electronics and fabrication. I can only say, some real nice work here!

Rob
"Firemen can stand the heat."

[gboxwood]

Thank you very much Rob!

I miss all the brass crafting part of this hobby and I try to compensate with 3D printing (which feels like cheating if you ask me).

Are dry seals something which you guys typically manufacture or are there ready made units that can be purchased? I miss such component for the 5mm diameter driveshaft.

Thank you again,

Giacomo