Building a submarine and diving to MAX DEPTH

Get your free training on how to build CPS 5: www.cpsdrone.com/
Let us know where else in the world we should take this drone! If you own one of these cool places, contact us at [e-mail]
The music in the video is from Epidemic Sound (I love their service). Use our link for a 30-day free trial: share.epidemicsound.com/prtrs9
If you feel like this video has improved your life, share it with:
- Your dad, mom, brother, uncle and your entire family (if you have one)
- Your life partner (boyfriend / husband / girlfriend / wife)
- Your 10 best friends
- Your drink buddies
- Your university / high school teacher
- Your coworker / coworkers
- Your boss
- All of your employees (if you own a business)
- Your university / high school colleges who you don’t consider friends
- Your hairdresser
- Your therapist
- Your doctor / dentist
- A random person on the street (you can use this video as a pickup line)
- Your dog / cat / rat / fish (this will get us some extra views)
- JJJreact
0:00 - Upgrades necessary
1:32 - Tether
7:00 - Endcaps
12:20 - Roadtrip
13:49 - THE MINE
15:40 - In the mineshaft
22:48 - Results
____________
Music:
Get Right Here Right Now by Fatboy Slim and over 1M + mainstream tracks here go.lickd.co/Music
License ID: XQZV1W9O5gO
♪ Sunset (Prod. by Lukrembo)
Link: • (no copyright music) j...
Fiber optic light animation: Vecteezy.com

[@PeterSripol]

Off to the Titanic now!

[@ockay16]

A bigger tethered drone is already on it's way, should be there tomorrow.

[@sUASNews]

An old pixhawk in there, made me smile. Great project, well done.

[@jordanmeyer66]

Do you think filling the electrical container with mineral oil would help acheive deeper depths ?

[@CPSdrone]

Btw, see you at OpenSauce Peter!

[@krisloaf5473]

the legend speaks

[@JayManty]

Good to see you're using a Playstation controller as a steering device. State of the art for actual submarines these days.

[@TheHadi545]

Lmao great comment

[@CAMSLAYER13]

Not even just the shit ones. Game controllers are a popular input device. They even use them for military drones

[@IdiotWithaMultimeter]

@@CAMSLAYER13good to know we’re using military grade control systems

[@OsirisThaMystikal]

@@CAMSLAYER13gotta appeal to working class teenagers somehow

[@maldy4604]

Lmao I thought the same thing and wanted to look for a comment about it

[@leonides4377]

I like how the controller is better than the one they used on the ocean gate sub.

[@SigmaShortsPL]

Allegro i wszystko jasne. Powodzenia chłopaki! Fajny content robicie!

[@groszek5413]

Allegro jest też w np. Czechach bodajże

[@groszek5413]

A nie czekaj factico

[@williamjacques3561]

Quick idea for the pressurised tube, fill it with non condutive oil. Therefore the tube shouldn't really be able to colapse

[@LostDeadSoul]

Heh, I came back to check if anyone suggested that. Transformer oil is the oil. I think I remember something about mineral oils like baby oil and motor oils which would work for a limited period. Maybe also have a rubber sheet in one end cap. It could flex if any micro bubbles should crimp within the pipe, acting like a balloon. Except only micro bubble flex. It would remove any pressure differential and keep the oil and water separated.

[@williamjacques3561]

@@LostDeadSoul well the idea i had was that having a fluid already inside the tubes, they can't collapse. Also, the fibre cable should of been one of those underground purpose cables, also, adding an surface primary power supply wouldn't be a bad idea, thus longer exploration are possible.

[@PrecioustheMovie1]

Fuckin brilliant idea

[@james10739]

I'm not sure if there is anything to hold it closed but it seems like if you wanted to go father or either way if you could pressurize the container so if you pressurized it to 200m then you should be able to go to like 500m

[@CPSdrone]

We must test that eventually

[@BartJBols]

You can also improve the flashlights by angling them away from the camera direction. this will prevent over-illuminating what is directly in front of you, but still allow light to bounce around and illuminate what you are looking at ahead.

[@CPSdrone]

Good tip, thanks!

[@ctrl_x1770]

How about situations where the light cannot bounce towards the front, making viewing difficult?

[@AchievementDenied]

@@ctrl_x1770 toggle-able lights with a 3-level dimmer?

[@0xddcce1]

yo that thumbnail legit looked like a monster from a horror movie 💀

[@RedTrex9]

I saw that too

[@fridaycaliforniaa236]

Funny how KZbin made me become an expert in deep sea exploration in just a few days ^^

[@SarahKchannel]

A tip for the cable. The cable should be made naturally slightly positive buoyant. Not too much or the drone will struggle to dive. You can do this by adding a strip of neoprene tape ever 2.5m, secure it with clear heat shrink and while at it - put a distance indicator under the heatshrink. 2.5, 5m, 7.5m, 10m etc.. It will help you keeping the cable away from the drone. Also the distance indicator help to just get a non electronic idea of distance or depth. Ideally this markings are something very luminous, like neon color.

[@xbox360Rob]

Top advice 🙂

[@CPSdrone]

Thanks for the tip with the neoprene tape! The buoyancy wasn't a major problem in this specific use case since it was always vertical, but normally that's a bigger issue when the tether tangles on the bottom. We thought of buying a neutrally buoyant fiber optic tether build for ROVs, but they cost quite a lot more and we figured that it's not necessary for now. Markings would also probably prove to be useful, thanks again

[@williamjacques3561]

Or maybe just tie wrap mini bouyies on the cable

[@SarahKchannel]

@@williamjacques3561 not a very good idea. You wanna wrap the cable on drum or figure-8 it on a surface. Those buoys will snag the cable above and underwater...

[@johanarnfinnlvold5989]

​@@SarahKchannel Figure-8 is preferrable, as it will keep the cable straight without kinks. Neoprene will compress under pressure. It would be better to just use some buoyant rope. and tape it, or even better, splice it into the rope. Another thing I found out back when I was a commercial diver, is that it is nice to have a small weight 2-3 meters away from you on a buoyant umbilical, as that will dampen the pull of the umbilical. (or a small buoyant object on a negative umbilical, for the same reason)

[@Sebastian1998844]

A fellow (mechanical) engineer here: next time when you're 3D printing, try testing how the printing orientation might influence the maximum pressure the caps are capable of withstanding (if you still wanna use FFF) Additionally, in order to avoid these issues in the future and don't invest too much time in prototyping and further testing, go right ahead and use SLS instead, vacuum it and apply epoxy (just like you did) and cure it. This might not be even necessary for SLS, but just for experimenting. That'll avoid any structural problems. To top it off, try using an additional plastic lid that will distribute the pressure axially a little better and provide better water protection. I have the feeling the lid experienced all the pressure and that was one of the reasons why it detached so often. Do a little FEM simulation where the maximum concentration of stresses might be taking place and see how you can reinforce it a little. I wish you lots of luck!

[@Sven745]

Or maybe they could use 3d print as a mold, and make the end caps entirely from vacuummed epoxy... The output would be a homogeneous material, and not structure prone to cracks like 3d printed plastic. Just an idea of the top of my head, but maybe I'll help somehow.

[@Sebastian1998844]

​@@connorswaven only the outside? I'm not a fan of only reinforcing the outside. Since the sub will be experimenting with some significant pressure (relatively speaking), I'd prefer to vacuum it and apply epoxy. You're practically replacing all the air molecules from the lids with epoxy. It's a little overkill imo but I'd do it anyways, merely to guarantee that the camera and the electronics are protected. Some rubber to better distribute the stresses axially might be suitable too.

[@Sebastian1998844]

@@Sven745 That's even better! I didn't think about it! Yeah, having one material doesn't sound too bad tho.

[@Chanextr]

Could you please explain why did you suggest SLS exactly? I'm usually working with LCD printers and I would like to know what structural difference the parts printed on SLS and LCD will have. how will it affect their ability to withstand loads?

[@Sebastian1998844]

@@Chanextr from experience. I recall using LCD for a few parts as part of an experiment with dynamic loading (mostly pressure). The results were acceptable but far from optimal. Cracks were evidently there, we found a few droplets in our electronics. The vacuumed environment protecting the electronics was damaged and we could see humidity started to form. We then changed to SLS and none of those things happened. Now, it's important to say that specifically for dynamic loading a more "flexible" material is recommended. However, the results we obtained with SLS exceeded our most optimistic expectations. Hence my suggestions.

[@calummacgregor589]

For the cable, be sure to use a central cable that is thick so tension anywhere is a long the central cable rather than the fiber optic cable. Also, try to have the spool go directly over the dive area next time to ensure it has minimal chance to get caught on something.

[@Marcox385]

It's such a delight to see this kind of projects and documentation about them. This is how engineering should be sold as some way of living, taking science foundamentals and turning them into fun experiments. Greetings from Mexico!

[@MaxLai_0104]

Submarine controlled with a game controller... Wonder how that'll turn out.

[@FB94121]

Well incase you didn't know controllers with joysticks are used in every submersible and by the US military. They are the best thing to use.

[@Abstract.Noir414]

Considering what youre talking about made several attemps to the titanic before i think thats a silly argument, the demise had nothing to do with the controller

[@MaxLai_0104]

@@FB94121 Yes, but the Titanic was controlled with a $43 dollar controller from GameStop.

[@fluffylittlebear]

@@MaxLai_0104 The U.S. military controls drones with Xbox controllers. They spent millions trying to create their own, and in the end, settled on using Xbox controllers because it wasn't possible to create a better controller.

[@FB94121]

@@MaxLai_0104 it doesn't matter. You know how many hours of research and testing went into those controllers. And functionally they are perfect. That's why many companies use them because you can't create something better, these have been tested for many years and can't be improved anymore

[@BartJBols]

The issue with loop radius in the glass fiber is that it becomes more an issue once your cable is longer. Every tight turn you have cuts a serious distance off your loop.

[@user-ox5ye1ju7b]

Be faithful in small things because it is in them that your strength lies.

[@CAMSLAYER13]

Its crazy that the human body can handle deeper dives

[@FinnFann]

Being more than half water has it's advantages. Downside is the other effects we have to deal with, though. At a certain depth you have to start using helium instead of nitrogen mixed in I think? Because the nitrogen starts to act as a sorta dissociative drug. Then there's coming back up that's the problem. You go up too fast bubbles form in your bloodstream (the bends) among other things (progressively more horrifying the faster you go up).

[@en2oh]

there is a degree of irony here, given the current loss of the Titan Submersible. Your work is brilliant! I would suggest changing the solitary fiber cable to a true umbilical cord that is neutrally buoyant

[@DrippyG9]

I literally see every kind of submarine videos and deep sea vehicles on my fyp, the algorithm is trolling us right now

[@jacobbailey2950]

@@DrippyG9 I got recommended jacksepticeye's video "Experiencing The Titanic Sinking In VR"

[@oliverwithers9736]

Tbf this is only 250 meters down, the Titan sub was pressure rated for 1500 meters down but was going down to 4000 meters (for some fucking reason)

[@francesco5254]

"there is a degree of irony" With all due respect sir, there's not a degree of irony. There's a degree of carbon fibery.

[@jamesthebirdman9412]

@@DrippyG9yea I’m getting every homemade submarine video ever. And they are all from like 2 weeks ago

[@caleb5309]

You could fill the cylinders with mineral oil. should help with the compression.

[@olioli8754]

Nie wiedziałem, że w Polsce są takie projekty jak ten.

[@marsmars2895]

ja się skapłem po kopalni pokazanej na początku ;D

[@MaxLove333]

wiadomo ze polakow zdolnych latwiej odkryc zagranica lub w innych wersjach jezykowych :)

[@proq06]

@@MaxLove333 ja po macie do krojenia od dousy

[@Cheeky_Goose]

I do recommend printing your cap in the vertical direction instead of the horizontal, so that it doesn't separate along the layer lines. Also, there are two things you guys can try. 1) try printing in Nylon which is more challenging to print with, but the layer adhesion is good. 2) Try using Cura's mold-making feature to create a mold for the epoxy so you can make an end cap entirely out of epoxy. Would this solve your problem?

[@andreass.7152]

Just use steel

[@kostarak3160]

@@andreass.7152 OceanGate would hate you

[@andreass.7152]

@@kostarak3160 haha 😂

[@SaHaRaSquad]

For maximum durability they could use that trick of encasing the 3d print in fine-grained salt and then re-melting it in the oven, it completely merges all layers and it turns into one solid piece.

[@t05s]

I was thinking like a dental resin or something

[@barabolak]

I hope you read this. Some fancy watches are filled with oil which allows them to withstand very high pressure. You could fill the submarine with mineral oil as well. Your electronics would still operate just fine, since it's not conductive, and there would be no need to 3D print really thick caps, and use thick cylinders as long as there's no air inside

[@AlexJoneses]

Filling with oil doesn't magically make it survive more pressure. It just stops electronics from being short circuited in water

[@mrtechie6810]

This.

[@guillemlluciagris5072]

The first ultra depth submersible was filled with gasoline as it is incompressible. Managed to reach the Mariana Trench in the 1960s with crew onboard

[@barabolak]

@@guillemlluciagris5072 I'm pretty sure all fluids are incompressible. They used gasoline to make the sub buoyant because it's like 30% less dense than water. I just learned this thanks to you. Very interesting stuff!

[@guillemlluciagris5072]

@@barabolak you are absolutely right. I forgot to mention the buoyancy bit. That was indeed the reason why gasoline was chosen. The submersible was named Trieste, the Wikipedia article is super interesting

[@ediblepigz]

I work with fiber optic cable like everyday. In my experience they are very easy to break due to how thin the ones I use are. The ones you guys had was actually pretty thick looking

[@geriangoofy]

what a nice time to make a submarine

[@sealightube333]

Great video! I would suggest to slow down the printing for the endcups, increase the printing temperature, slow down the cooling fans, by this you will get better layer adhesions. And there are lot of gaps in between the layer lines I see also possibilities to improove it.

[@CPSdrone]

Thanks! The bad layer adhesion is also due to the fact that it's printed in ASA filament. Also, I believe that most of the strength comes from how well the epoxy bonds with the print. We'd need to see if endcaps printed in PLA/PETG actually make it stronger.

[@Corvus.2606]

@@CPSdrone maybe resin printed end caps might be more reliable for such depths? I understand it makes it less accessible for people who don't have both types of printer, but it's both harder and has better layer adhesion

[@lotoox7299]

@@CPSdrone have you considered machining them from aluminium? They look to be fairly simple cylindrical shapes that could be easily produced with a lathe.

[@AlJay0032]

Or just have them made from aluminum.

[@sandervanhee3828]

Or add more ridges on the inside of the printed parts so the epoxy can grab on to them and bind them. You can also add fiberglass in the epoxy

[@youruncleted]

my dude posted this video in the perfect moment, he must have inside info

[@t0x1C698]

okay so literally had this idea for a year now for my FYP, and this dude has done it the exact way I was thinking.

[@lal12]

For one layer adhesion doesn't look that good, so tweak with your settings or use another filament. You also could inject the 3d printed part with epoxy. Use Gyroid infill and fill it up, maybe by using the vacuum chamber. Or another idea I would try is to pressurize the chamber itself, either by using a CO2 cartridge (sodastream or similar) or by filling it with oil and adding a pressure exchange.

[@tarcisiobatista5595]

Also use waaay lower layer hights to get more streng, whats cnc kitchen video about layer height and parts streng to get a better underatanding. In the video looks like the layers you are using are lile 0.3 or something and would go for lile 0.06

[@visionofmalkav]

Pressure at 270m is over 380 psi. The average consumer CO2 cartridge for things like that is around 120 psi. Paintball tanks can go up to 800-900 psi but they're MUCH larger and because of the huge pressure difference you'll need a two-stage pressure regulator to safely ensure you don't blow the entire device apart. Flooding it with an incompressible diaelectric fluid like mineral oil might be a better idea.

[@lal12]

@@visionofmalkav yeah you're right. The cartridges I had in mind aren't for SodaStream but for CO2 pistols they have similar size and go up to 800psi/60bar. But adding two valves and a cartridge is a lot anyway, so the oil variant is probably the easier solution.

[@BenKDesigns]

@@visionofmalkav What about both? There's definitely some improvements that can be made for the end caps to prevent them de-laminating, including increasing the nozzle temp and printing the caps at an angle to vary the layer lines. The mineral oil is a good idea for the inner cylinder, but I'd imagine it would have an impact on the optics of the lens.

[@AlexJoneses]

Less of an issue with later adhesion more of an issue of 3d printing, fdm specifically being a bad option for manufacturing. The part is too unpredictable strength wise so you never know what will happen at pressure

[@user-du2tq1zx6d]

Your vision will become clear only when you look into your heart. Who looks outside, dreams. Who looks inside, awakens.

[@BraydonBlanchette]

Can we appreciate that really sick transition he did at 8:23 he combined the time lapse footage, button overlay of real time, reaching for the 3-D printed part

[@seabeepirate]

On the end caps if you want to stick with 3D printing, try using exposed infill so the resin can bond more thoroughly. Super cool project guys!

[@dominikzz6667]

Na prawdę mam nadzieję że dostaniecie się do światowego mainstreama DIY youtube. Bardzo fajnie się ogląda, jesteście naturalni i ciekawi. Powodzenia :D

[@CPSdrone]

Dzięki wielkie :)

[@RADIUM108]

Im watching this from India so i think its pretty global already. Hope more people get to know this 😊

[@larsolaerts9078]

15:34 contrlolling a submarine whit a playstation controller is giving flashbacks

[@folderx2000]

This is the maker community that I want!! Feedback after feedback!! Glad I followed these makers!!

[@joetom9429]

They are scam, Not community mate, they want money for the 3d files.. what kind of community is that?

[@azertyQ]

in my experience with silicone degassing, you can reduce the amount you need to degas at subsequent stages by degassing before mixing (especially useful with short cure time materials, but mixing will add gas back in, there's techniques to reduce that), then degassing after mixing, then a final degas once its in the mold. otherwise, be careful with those vacuum chambers with the inlet threaded into the cover, they are prone to cracking.

[@en2oh]

the above is good advice. When you are pouring and liquid, you can reduce air entrainment by filling with a very thin stream of liquid. One question I had. Would using a more flexible rubber compound be better than rigid epoxy?

[@jakervamduf1523]

Robisz dobrą robotę!

[@FMMProductions]

There's a famous dive spot in a small town called Polcenigo in Pordenone, Italy which is where my mum lives. The water there is incredible and so crystal clear. I'd love to see you guys tackle that at some point.

[@HarveyGuitarBoy]

Oh man I'm going to start saving to buy the full build course on this. What an awesome idea, I've always wanted to build one of these but haven't the technical knowledge - without a guide!

[@smoxesk]

learn !!!!!!!!!!!!!

[@TheAnoniemo]

Have you thought about printing the endcaps at an angle (30 - 45degrees) instead of flat on the build plate, so the layer lines don't create a weak spot around the o-ring?

[@benstone217]

Or just fill the tube with oil so it can’t be compressed

[@TheAnoniemo]

@@benstone217 yes, but that will make upgrades and changes a lot more messy and difficult. Pressurising the tube could also work.

[@hippopotamus86]

Time to invest in a CNC I say.

[@TheAnoniemo]

@@hippopotamus86 but then it's no longer a 3D printer submarine

[@duckypolice]

@@benstone217 SAME IDEA BRO!

[@VRinRL]

But you didn't use a logitech controller...

[@gator1984atcomcast]

Good job guys!! Gives me confidence in your generation of engineers.

[@diwaperkasa2351]

First time i know about ROV is when Erick Stackpole came into my country and show his DIY ROV which is Open ROV. I feel amazing until right now when someone build they amazing DIY ROV. Congrats for your team 🤜🤜

[@MartinBogomolni]

I suggest using ALUMINUM CNC milled endcaps, O-rings, and steel bolts to hold the assembly shut end-to-end. Also pressurizing the endcap to ~25 bar or so to compensate for depth. OR alternately using non-conductive mineral oil which is not compressible but fully transparent. This will require switching from 3D printing, but the techniques are still available for anyone to do using a simple 3D printer and simple CNC.

[@thombaz]

Well you can machine an endcap like this in minutes on a lathe, but fitting a lathe or a "simple CNC" in a flat is not possible, but everyone can have a 3d printer.

[@thombaz]

Milling would be really a waste as its a perfect cycle.

[@thewalrus1968]

Perfect timing for this video ... Well done

[@ultimateearrapechannel31]

⛴️☠️☠️

[@bekofan]

Love the music right here right now from the hotpoint washing machines comercial

[@rafael_13]

A 360 camera would be so cool on this submarine...

[@JoeNagle-Bagle]

It looked like the end caps both failed directly between layers, which makes sense as layer adhesion is one of the main weak points of 3D printed objects. You could potentially print at a 45 degree angle or just some sort of angle so that some of the force is transferred to the material. This would prevent the layer adhesion from being the sole force that is preventing the caps from failing.

[@Gainn]

That, and they also need a better printer. Someone else did pressure testing on FDM and they found an angle of about 12 degrees worked the best.

[@CeesMeijer]

May seem a good idea for increasing the stregth, but the layer lines will then also be 45 degrees on the section here O-Ring is. This will make is almost impossible for the O-ring to function without leaking.

[@sobertillnoon]

What timing for this video

[@ahmethakanarslan3893]

Bro controlled with ps dualshock like titan submarine

[@jalugamer3515]

i loved that the camera still worked after the water came in for a few seconds. you can see the water rushing in and then see some dirt come in too before the camera dies

[@jstadude9810]

Maybe if you added 550 cord (paracord) to your umbilical it would add enough strength to confidently pull up the cable without damaging the fiber optic. It may even add enough bouyancy to compensate for the weight of the fiber optic.

[@lsswappedcessna]

It's no secret why this is in everyone's recommended, submersibles and ROVs (like this one) are on everyone's mind right now. That being said, this (remote controlled underwater vehicles) is a really cool hobby that I wish I had the time and knowledge to do myself.

[@theboyinthedark6521]

This video is literally just 12 days old

[@lahtawsinwanawlahtaw7410]

Thank You, Sir.

[@MusicalBoarder]

Imagine this guy putting more effort into making sure his little robot toy is safer than the titan submersible.

[@deaconthomas7930]

Hey, you made it deeper than the titan

[@MarcoSkate1]

powodzenia w przyszłości chlopaki ❤

[@kamiljaje4123]

Cos pięknego, dopinguje. Sam chce zacząć budować drony a wy mnie do tego motywujecie 😁

[@Krecioreq]

w 22:10 sekundzie jakoś chłop mówi po polsku

[@Chewligan1]

Thanks for uploading this very exciting project.

[@fazzie1916]

Fajnie widzieć, że ktoś w polsce robi taki content na yt :)

[@tireballastserviceofflorid7771]

Awesome build my friend. Another tip for your epoxy. Look at the additives you can get. Such as milled fibers. You could double or triple the resin strength with minimal viscosity increase. Also products such as west marine G-FLEX epoxy is a nano reinforced resin. It is incredibly stron. It comes in liquid or pre-thickened into a gel. Also things like glass sphears fumed silica as well as metal powders. Use stainless powder in the resin for wear resistances and extra weight. Use your vacuum pot with all of the additives or you will get bubbles. You also should not apply curing heat until the resin is atleast to late stage gel. Curing rapidly especially in a humid environment leads to lots of amine blush. Not a big problem but will limit rebonding. Looking forward to some other projects.

[@Jerebo_mec_extreme]

The timing 💀

[@lipaldi]

thank you, my cat is very happy with your video, he can't wait for more videos

[@WACkZerden]

excellent video thank You for sharing

[@marcinkiewicz2020]

Chłopaki kozacka robota, cieszy mnie to, że mamy w Polsce takie głowy i nagrywacie po angielsku dla większych zasięgów.

[@ProjectHub1]

POLSKAA GORAAA

[@carnonPL]

osobiście nie spodziewałem się usłyszeć Polskiego głosu w filmie. Miłe zaskoczenie

[@archaycxstreamer1270]

What about printing the end caps at a different orientation? If the layer lines are vertical or diagonal to the direction of force, it should support more force before separation.

[@CapApollo]

perfect timing

[@elitemedium]

Your PS5 controller for your drone is amazing 😃😃😁👍👍........diving so deep not bad at all

[@gqinc1202]

You may also try printing the end caps on a resin 3D printer, they'll be much stronger and won't separate across the layer lines, that combined with filling it with non-conductive oil should make it super strong.

[@loki_fit]

Better then ocean gate

[@JQ3B94]

Well erm this is conveniently timed

[@123evilwolf]

Projects like these are so interesting to watch tbh.

[@marciofernandes7091]

As a professional freediver, I have to say this is ultra impressive. I would love to exchange ideas/expiremtns.

[@TheMcpele]

Miałem pytać gdzie to dopóki nie usłyszałem polskiego 😅 jestem z was dumny Panowie życzę sukcesów 👍

[@theengineeringmonkey407]

I’m not sure if it’s possible but try to print the end caps at a different angle that way it can’t separate the layers and has a lot more strength against the pressure. You could try printing it vertically or at an angle just with a little extra support

[@flaviomonteiro1414]

Up, this may help!

[@examplerkey]

If you want to strain test the fibre optic cable, do it on a separate piece or make sure you cut that strained bit out. Unlike a normal wire that you could solder, you need a splicer to join the fibre.

[@carmatic]

oh wow, another submarine with a cylindrical pressure hull and controlled by a gamepad not only that, but you also had a delamination failure mode too! except it was in the 3d printed endcap and (more importantly) during a test

[@Milkywayboy]

You should build one and charge 250k for passengers

[@adamlisovsky5233]

remeber to charge the controller!!!

[@brucebadeau9906]

Watching from Houston Texas --- really great work -- very interesting experiment -- SMART GUYS...

[@YkBigMak]

Oceangate needs this guy

[@drxym]

Looks like the end cap failed on the layer lines of the print. It might be an idea to anneal the prints in heat & salt to make them more solid. i.e. surround them with salt to hold their shape and then put them in an oven to bond the layers more securely

[@tomclanys]

This, it's a great idea, but needs 100% infill otherwise it will deform

[@Fischertek]

Nice job guy's! 😀 Is that kevlar in the fiberoptics cable? So it should be able to withstand som pulling force! (And some abrasion). Have you considered metal endcaps? They would be easy to make from aluminium. Except the cable passage, but that could be made in another shape. Also other casting resins would be better than epoxy ( Remember to wear nitril gloves!!) Maybe you should even have floaters on the cable!? If it's much heavier than water, you might get into trouble somehow. Especially if you enter a corridor in the mine. I'm looking forward to your next adventure. And also a film from the Open Sauce! Good luck there!

[@CPSdrone]

I think it was Kevlar in the cable - it made the strength very impressive. We'll definitely consider aluminium endcaps for the next time we'll go into the mine. I think for the next time we'll also buy an ROV grade fiberoptic cable, which will be neutrally buoyant by default, we'll see. Thanks so much for all the support Peter! We really appreciate it :)

[@FLIGHTORY]

Great channel, just discovered you guys. Powodzenia :)

[@mschr2880]

11:33 You are using radial compression for the seal. The water pressure compresses the o-ring out of its way into the notch or pushes it in axial direction, where it has little support. A better design would be to compress the o-ring axially between the cross section of the main pressure cylinder and the end cap (just take a look at your vacuum chamber for reference). The o-ring only needs some support from an inner cylinder attached to the end cap, which also aligns the end cap with the inner diameter of the pressure cylinder. However, there are two challenges with this: The cross section where the pressure cylinder was cut off needs a good surface finish for the o-ring to rest on. But this usually can be accomplished by polishing, fine sanding, heat treatment or chemically, depending on the exact material. More difficult is the fact that the density of the o-ring must match the ratio between axial and radial pressure that the water will apply via the end cap and directly on the o-ring. Therefore, you should not try to overcome the surface finish problem by using a soft o-ring that cannot withstand the axial load and will be squished out of its notch. An additional safety feature (that should not be relied on for the design) could be an external support ring, that is put over the o-ring after the end cap has been closed and the ring is axially compressed, so that the o-ring cannot escape from the notch in radial direction.

[@jordanflorance4419]

At least you’re using a PlayStation controller for this one

[@josephrsalexander4589]

Could you fill the tube with a non conductive fluid that does not compress?

[@Ninguem77777]

Do de o-ring seat at the policarbonate cilinder. In the way you did, the 3D printed cap is taking tensile loads from o-ring at the channel wall.

[@georged7627]

With that raspberry pi you would be able to play some Mario Kart under water 🤘🏾yeah duuuude

[@SarahKchannel]

A tip given by somebody else already, increase the distance of lights away from the camera lens. The further away the less 'blinding' you get from sediments in the water. Angle them away from the camera axis, so that particles get lit rather from the side than head on. I know your drone is narrow, so there are limits. In professional underwater camera gear we have arms reaching 50-100cm out from the camera.

[@Zeta_9mole_rats]

you wouldn’t happen to use a logitech controller would you

[@stillridgeway]

Maybe adding an Logitech F710 for steering would be quite innovative!

[@ucjakozicka978]

wow thats so cool guys! 👏👏 cant wait for the next videos 😁

[@spika5887]

Bardzo fajnie się was ogląda

[@kozaks5727]

They got a better controller than..

[@bremsshadow8722]

They didn't have billionairs inside, that's why that thing broke so fast.

[@OAikoT]

Hello fellow drone pilot Pole on YT! Good quality content; keep it up mate!

[@bobua2137]

First time watching you and I only realized you are Polish when you used Allegro, your English sounds really good!

[@weird_dude124]

better than titan

[@DemonPig666]

Well that didn’t age well

[@raziel3726]

Amazing Craftsmanship 👍🏻👍🏽👍

[@itskleb5641]

They used the Sherlock Chronicals Music 😍😍

[@injury11]

This is way better than a implodable submarine controlled by a logitech videogame controller

[@lmaobox4068]

*submersible

[@lmaobox4068]

*submersible