Project "Jules Verne" - Creation of the Russian ROV

[Kant39 (Denis Stein)]

Greetings to all! I have been nurturing the idea of creating a ROV for a long time, since I live on the Baltic Sea, between Poland and Lithuania. Of course, the Baltic Sea has muddy water and this sea is not deep, but in the Kaliningrad region there are large lakes with a maximum depth of up to 70 meters with very clear water.

I have been developing flying drones and shooting wildlife and nature for over 10 years, my drones have climbed over 5,000 meters and captured gorgeous clouds. But in connection with the events in Ukraine, flying drones has become pointless and dangerous. Since I have been involved in radio electronics since childhood, I cannot sit still - I need to create something. I decided to create an underwater drone - ROV, research class. I want to dive underwater and explore the underwater world.

In connection with the events in Ukraine, Russia is increasingly turning into North Korea and therefore, due to various sanctions, many things were not available to me.
Now, I can buy practically nothing from the USA, Europe, although I actually live between European countries. (The Kaliningrad region is an enclave - separated from the greater Russia by the countries of Poland and Lithuania). I used to be able to buy on Ebay, but now it’s almost impossible - no one wants to do business with Russia and this is expected…

So, I wanted to use a lot from BlueRobotics, but this path turned out to be closed to me and also became very expensive - the standard of living in Russia fell very much. As a result, as an engineer, I decided to do absolutely everything with my own hands, since I have no other choice.

I called the working project “Jules Verne”, dedicated to the famous French writer who gave the world the Nautilus and many interesting books.

I decided to start building the apparatus by creating the necessary components. I see no reason to immediately create a general drawing of the entire apparatus, since at the initial level there are too many unknown points and many things will change their design.

It is very good that before the start of the war I managed to build my own CNC machine and bought a cheap Chinese lathe and greatly altered it. As a waterproof case for batteries, I decided to use a thick-walled polyethylene pipe for gas, water. I planned to machine end caps with sealing rings.
After buying a lathe, I tried to turn the inside, as the surface inside is very uneven. I quickly became convinced that in its original form, the Chinese machine is no good for anything and needs to be heavily redone.

I completely dismantled the machine, scraped the working surfaces of the machine to improve rigidity. I also installed two stepper motors, installed a control board, introduced two electronic rulers - so I can see the movement of the axes on the big screen with an accuracy of thousandths of a millimeter. As a result, I can operate the machine accurately and conveniently, set any feed, cut any thread without having to change gears (I got an electronic gearbox).

And this is my CNC machine with 4 axis. I specially made a high vertical axis Z, in order to be able to process normally in the rotary axis.

[Kant39 (Denis Stein)]

So, for starters, I decided to come up with my own sealed system (Penetrators) for wires, sensors, motors, lights and any other external devices. I came up with such a system of two types. A short penetrator for an underwater light (made of aluminum alloy D16T (2024 in the USA) with cold anodizing) and a long one made of marine bronze (CuAl9Mn2). They have silicone o-rings. The main difference of my design is the presence of two round printed circuit boards. The channel inside the threaded cylinder was filled with 703 sealant, which has very good adhesion. The boards were connected to each other with a 1mm copper wire and soldered. Finally, the silicone insulated cable is soldered to the PCB and sealed with 703 sealant. As a result, we get multi-level protection for water flow.
All this was done on a lathe and CNC machine. I made cutouts in all parts for a 15 and 16mm open-end wrench, for secure fixation.

These are blanks that I machined from marine bronze for penetrators and nuts.

And these are aluminum blanks for penetrators. alloy - D16T (2024)

And these are ready-made penetrators

These are 3 and 4 pin penetrator boards. (3 pins for motors and lights, and 4 for sensors and other devices)

The short penetrator is screwed with its part into the body of the dive light, it does not need a nut. The body of the dive light will also be made of 2024 alloy. I did not try to make the penetrator for the light from marine bronze due to the possibility of electrochemical corrosion of two pairs of different metals. The long penetrator is fixed with a nut on the other side.

Silicone rings on the nuts are actually not really needed, since they are inside the ROV, however, when tightening the nut, these rings will not allow the nut to unwind from vibration, shaking. In addition, such a nut will be easier to unscrew if necessary (it will not stick to the surface of the end cap).

My design penetrators has the following advantages:

• in case of damage to the cable jacket, the device will not be flooded with water.
• you can repair and replace any external part of the ROV without dismantling the sealed box with electronics (you just need to remove the sealant and cut and re-solder the wire)
• several levels of leakage protection (joint work of the sealing O ring and sealant 703)

[EliotBR (Eliot)]

Hi @Kant39,

Definitely an impressive setup you’ve made for yourself, and it’s cool to see your designs and ideas, and the progress you’ve made so far!

The PCB-capped penetrators are a nifty idea, and you could potentially adapt them into some form of connector down the track.

Sorry to hear the negative impacts of the nearby geopolitical conflicts on your life and opportunities. If there’s to be one positive from the situation, at least you’ve managed to maintain motivation and make the most of what you do have available to innovate in an extra limited design space - it’s impressive and inspiring

[Kant39 (Denis Stein)]

Hi Eliot! Thanks for the words of support! In fact, often hands down, because you do not see a good future. In addition, those men who remained in Russia live in constant fear that at any moment they may receive a mobilization agenda for the war that is going on in Ukraine - after all, a lot of people are dying there, both from Russia and from Ukraine.
Of course, many Russians left for Kazakhstan, Georgia, Turkey, but this is possible only for people who have a reserve of money. In Russia, the average salary is $350. Being an engineer, I get more, including due to the fact that I repair Canon, DJI at home, but still this is the minimum for the survival of a family of 4 people. Nevertheless, one has to live in hopes for a better future, even though my mind says otherwise.

[Kant39 (Denis Stein)]

So, when I decided on the design of the penetrators, I began to design an underwater light. When creating the underwater light, I forgot that the Baltic Sea is actually not very deep - in my mind I swam in the Mariana Trench. This can explain the incredibly large margin of safety for the water pressure of my underwater lights. I work at Solidworks and I ended up with my build design.

I’m starting to make a part on a lathe!

Preparation is done! Now it’s up to the CNC machine!

Cut the thread by hand - so calmer. There is nothing worse than breaking a thin M3 tap in a hole.

In parallel, I made a high-power LED driver circuit board. I plan to use LEDs with white, or rather yellow, because the less blue spectrum in the light, the better it spreads in the water. In addition, I will use UV LEDs. In ultraviolet light, many precious stones are visible - amber, diamonds. Of course, there are no diamonds in Kaliningrad, but here is the largest amber deposit in the world! Diamonds under water can be found where there was active volcanic activity - for example, the Far East of Russia, Kamchatka.

I made the central part - the mount for the LED. It has many holes - wires for the LED and for fasteners pass through them. This part, in my opinion, will very effectively transfer heat to the massive outer part of the underwater light through a good heat-conducting paste.

[Kant39 (Denis Stein)]

After I completely made the main body of the underwater light, I seriously thought about how to protect the metal from sea water corrosion. Unprotected aluminum alloy won’t last long so I decided to anodize the aluminum. As you know, there are 2 types of anodizing - warm and cold (hard). Warm anodized is produced at room temperature or above 0 degrees Celsius. Warm anodizing is not suitable for us, since the coating is not very dense and, most importantly, it will not be able to withstand scratches from sand and other abrasive particles. I chose cold (hard) anodizing, despite the fact that it is much more difficult, but the coating will be corundum - what sandpaper is made of.

To be honest, I was afraid to spoil the made bodies of underwater lights, since it took a lot of time for each of them to work on different machines. So I decided to practice on something else. At this point, I decided to do another small project that would help me further my hobby. We are talking about the manufacture of a miniature grinding machine with a diamond disc and very low runout. I dreamed of having this machine for a long time - so that it would be convenient to sharpen any drills, cutters, knives and anything. The problem was that such machines are not for sale (there are large, uncomfortable and with a lot of runout), so I decided to make such a machine and at the same time try out cold (hard) anodizing.
So, let’s leave the underwater lantern aside for now - we will return to it a little later.

As usual, I built the design of my grinder in Solidworks.
So, the key requirements for the machine:

• 3-phase brushless motor. No brushes - the main thing is reliability. I chose the motor from the drone T-Motor MT4008-12 600V, power 450W.
  • Incredible rigidity and strength. It should be made of 2024 alloy 30mm thick.
• The presence of a power supply port (8-35 Volt), as well as two USB ports for connecting flexible small lights. Good light is the main thing for quality work!
• Presence of a reference platform with markers of given angles. This will allow you to accurately sharpen the drills.
• Of course, the speed controller!

After developing the design, I began to cut out the parts on my CNC machine.

Simultaneously with the manufacture of parts of the grinder, I began to prepare for cold (hard) anodizing. It turned out to be very difficult. I planned to use automotive battery electrolyte and dilute it with distilled water. During quick tests, it turned out that I needed to cool the electrolyte to at least minus 18 degrees Celsius! In the Kaliningrad region, such temperatures rarely occur in winter, and not even every year! Then I bought a copper tube and made a cooling circuit for the electrolyte out of it. I covered this tube with a layer of varnish so that the copper would not interfere with the electrochemical reaction. Freon will flow through this tube using a refrigeration compressor!
We must not forget that cold (hard) anodizing is carried out with a high current density, which means that the part, like the electrolyte, will quickly heat up. That is why rapid cooling with freon will help here.

From a sheet of lead 3 mm thick, I cut out pieces and soldered them like the inner shape of an aluminum pan. This will be the electrode. I also made sure that it is extremely important that the electrolyte is mixed in the container, otherwise the process will be uneven and not of such high quality. It is for this purpose that I designed my peristaltic pump. This is where my 3D printer helped me. Printed with PETG plastic. This pump can easily pump harmful acid through the silicone tube and it won’t hurt the pump in any way! In medicine, this is how blood is pumped.

To tell the truth, the first experiment turned out to be defective due to insufficient current and I had to remove the resulting layer of solid anode from the part using caustic soda.

It worked on the second try! Both parts have received a hard coating that cannot be scratched with a sharp knife! After anodizing, I boiled the parts in boiling, distilled water for an hour. This is necessary to “close the pores” of the anodic coating. There is a structure similar to open cones, when boiled in water, aluminum hydroxide is formed and they are sealed tightly. Interestingly, the two parts turned out to be slightly different colors! I didn’t use any dyes! It’s just that during anodizing there was a slight difference in the temperature of the electrolyte (the freon did not have time to cool much).

The casing, the body of a miniature grinding machine, I printed on a 3D printer. With pleasure I assembled my designer! This is my favorite stage!

So, the machine turned out as if from one piece of metal. Incredible rigidity and complete absence of beats. I’m sure this thing will outlive me!

So, I got my experience with cold (hard) anodizing and now I can get back to my ROV! Or rather underwater lighting. Summarizing, we can highlight the main things for successful cold (hard) anodizing:

• The electrolyte temperature must be between -18 to -10 degrees Celsius. At minus 5 degrees, the anode film already becomes not so dense!
• Be sure to mix the electrolyte.
• With cold anodizing you will need a much higher voltage than with warm anodizing. This is due to the fact that during cold (hard) anodizing, the film is so dense that it has a higher resistance to electric current! You will need a source up to 70 Volts and current up to 15 Amps! With warm anodizing, as a rule, 12-20 volts maximum is enough.
  -When anodizing, the electrolyte heats up quickly, so the maximum anodizing time is 1 hour.
• The temperature of the electrolyte cannot be colder than minus 18-20 degrees, because already at minus 19 degrees, sulfuric acid with distilled water turns into ice! You can make more sulfuric acid, and make minus 25 degrees Celsius, but the electrolyte will be unnecessarily aggressive.
• Automotive electrolyte is aggressive not only for eyes and skin, but if it gets on clothes, these clothes will then turn into holes and tatters! But you will not see it right away, but only after washing! My clothes looked like I had been shot 10 times with 12-gauge buckshot! All my relatives laughed at me and asked - who shot me so many times!

To be continued…

[EliotBR (Eliot)]

Fascinating updates - thanks for sharing!

That light design does look a little overkill - hopefully they’re not too heavy for the ROV to lift!

Cool to see the different tools and processes come together to make and treat the various components. I don’t have experience with anodisation, so learning more about what’s involved was very interesting

[Kant39 (Denis Stein)]

Thank you Eliot! The weight in air of my dive light will be 340 grams. I guess that’s not really a lot, given some points:

• ROV, I would like to make a heavy class based on 8 engines
• According to my idea, the ROV should have a little positive buoyancy so that it slowly floats on its own. I need this to lift less dirt, sand from the bottom of a lake or sea. That is, in the water, with the help of a load or elements of positive buoyancy, I can achieve balance.

The only moment where the excessive weight of the parts can have a bad effect is the carrying of the device. But so far, the weight does not look unnecessarily large. For a long time creating drones for flying, I fought for every ounce of excess weight, as this is precious flight time. There they had to act at the expense of reliability and abandon some solutions (folding legs, body elements). In the case of the ROV, I decided to act in favor of reliability.

[Kant39 (Denis Stein)]

So, after gaining experience in hard anodizing, I started anodizing underwater light housings. I have made many different fixtures for the part to be immersed in an electrolyte solution. These parts not only have to hold the workpieces, but also reliably transmit a large current. If there is bad contact somewhere, then this place will heat up and in this place it will corrode the workpiece itself and the attachment to it.

Having even a little experience in anodizing, everything worked out well for me. Here is one part with a hard anodized coating, and next to it is a bare part immediately from the machine.

It would seem that everything is fine, but I found one bad moment. In principle, it would be possible to ignore him, but I’m used to doing everything in life well. My now deceased father always told me - “do only well, or do not do it at all.” The fact is that my cases have many blind holes with M3 threaded threads. With the help of a high resistance meter, I found that inside the thread channels, the anode film is not so dense and it is uneven. Of course, this is due to the fact that electrolyte does not circulate inside these small holes at all, and even air can remain there, which will not allow high-quality anodizing of the surface inside the holes.
I can’t leave it as it is, as sea water getting inside the threads can corrode the metal. Of course - the Baltic Sea is the freshest sea in the world, but even fresh water can corrode aluminum, so this problem had to be resolved.

I made my next development - a device for hard anodizing deep holes. The principle of operation of the device is quite simple - the two halves of the parts are interconnected by silicone tubes. In each part there are holes - sprayers. A thin silicone tube is glued into them, which is then inserted into each hole in the body of the underwater lamp. Electrolyte is fed into one part of the nebulizer through a tube through a peristaltic pump. The electrolyte continuously flushes each of its holes and as a result we have an ideal anode film.

So, I simply installed the already finished underwater light housing into the fixture and redid the anodizing process. There was enough time of 30 minutes to be able to easily screw stainless screws into these M3 holes. With a longer exposure in the electrolyte, a thick anode film will not allow the screws to be screwed into the thread.

Finally, I made the front glass from cast acrylic, 12mm thick. I cut the glass profile, holes, chamfers for screws on a CNC machine. On a lathe, I machined the ends of the glass and polished, also chamfered the edge.

Now it’s time to move on to the final assembly of underwater lights. The color of the housings of underwater lights is slightly different due to the difference in the temperature of the electrolyte, but the mechanical properties turned out to be the same.

On the front, you can see the red ring, it is a 3mm high-strength silicone O-ring, flattened under the pressure of thick glass. So this seal will perfectly withstand high water pressure. I printed the yellow caps on my 3D printer. They are needed so as not to damage the glass of the lamps when transporting the ROV in the car and carrying it.

The internal protrusion in the protective cover itself is necessary so as not to scratch the glass by the cover itself when turning (remove and put on).

In total, I made 5 underwater lights. 1 white light (very warm spectrum - yellow), and 4 UV lights. All 4 UV lights have different wavelengths of UV light (I signed each LED with a marker). I plan to submerge all the lanterns in water and determine what wavelength is best for the best visibility of amber. This semi-precious stone glows differently at different UV wavelengths. As a result, I will know which LEDs are better for me to put. In total, I plan to use 4 underwater lights on my ROV - 2 white and 2 UV. Of course, they will all be controlled by the PPM bus, brightness independently. I will have 1 underwater light as a spare.

By the way, when assembling underwater lights, I made sure that the UV LED quickly sets fire to a match with its light emission. I additionally made sure that my cast acrylic glass did not interfere with strong UV radiation and did not heat up.

For fun, I also did a static calculation with different pressures on the 12mm front glass. The picture below shows the deformation of glass under pressure at a depth of 5,000 meters (7090.67 psi). Graphically, in the picture the bending of the glass is exaggerated (adapted for visual perception) - in fact, the glass will bend to a maximum distance of 0.998mm.

Below, the deformation of the glass at the bottom of the Mariana Trench (depth 11000 meters, 15599.48 psi) is shown. It will be 2.196mm in the center. It is difficult to say whether the underwater lamp will withstand such pressure - on the one hand, for acrylic glass, such deformation should not be destructive, but on the other hand, there is a penetrator. Although the penetrator has a small footprint, the sealant will strongly deform and affect the printed circuit boards, even though there is also sealant behind them. If I decided to lower my underwater lantern to the bottom of the Mariana Trench, I would fill it with oil inside, for example.

To be continued…

[ljlukis (Larry Lukis)]

Wow. That’s all pretty amazing.

[damonblue (Damon McMillan)]

Wow! Inspiring. Denis, I think you are my new hero. All of this is so cool!

[Kant39 (Denis Stein)]

Thanks Damon! It’s very nice to read this!

[jim_p (Jim Poulmas)]

Hi Denis,
This looks great. How did you machine the channel for the o ring? I have a lathe and I was thinking of using a parting tool and slowly walk it into the part to cut the channel. Or maybe just getting a blank lathe tool and grinding it to size. Stainless is easy to weld but difficult to machine. (For me anyway, I’m not a machinist.)
Jim

[Kant39 (Denis Stein)]

Hi Jim! In fact, I am also not a professional turner, but only a self-taught amateur.

For underwater lights, I used a standard cutter MGEHR1212-2 with a 2mm cutting insert. I machined this insert on a diamond grinding wheel for a specific shape. With this cutter, I made a rectangular groove for the 3mm seal ring.

For underwater penetrators, I used a different cutter. Since I use a 2mm thick seal ring in penetrators, I took a cheap cutter and gave it the desired thickness of 2mm and made it rounded. Due to the fact that the penetrator has a long overhang, I turned the cutter the other way around and thus machined the groove.

Note also that I use red silicone rubber as the O-ring material throughout. The fact is that silicone rubber withstands cold and hot temperatures better and does not flatten over time, such as the popular NBR. In general, a rectangular o-ring groove profile is generally better suited. For example, on my underwater flashlight you can see how much the red silicone rubber flattened due to the high down pressure - when compressed, the silicone completely filled the rectangular groove profile. If the groove were round, then the silicone would not be able to shrink so much and it could be damaged. In the case of a circular groove, it would have to be made much deeper. This is how the groove on the penetrator is made - deeper than would be required with a rectangular groove profile.
In any case, it is better to try to make a groove on a rough workpiece, check everything and then do it on the real part.

[jim_p (Jim Poulmas)]

Thank you Denis!