Dragon Shark - a new start

[IdefixRC]

Hi guys,

first of all thanks for adding me to the forum.
I'm starting my 3rd journey into RC Subs after 2 fail attempts in the past.
- Attempt 1: was a scratch build German Sub - dynamic diver - around 25 years ago which I bought of a gentleman at a RC flea market. A nice kit. Unfortunately on it's 3rd voyage a seal must have bust which turned the dynamic diver into a static....one time.... driver. Lost at "sea" and could not be recovered (autumn and after the winter was over dives in the area did not yield any results...)
- Attempt 2: the first Dragon Shark which I ordered "ready to sail" from China. Took over a year and countless follow-ups but it arrived.....in pieces....guess that what happens if you pack a RC sub full of lead, wrap a layer of bubble wrap around it and throw it into a box and send it our from China. After sending the pieces back I got a kit back.....

And this is where we are today: A rather dated, first generation, Arkmodel Dragon Shark which, after many years in storage, has a few components (such as the dive bladder) turned into a powdery kind of mess.

What I have in mind:

• Replace all the resin parts (which are not of the best quality) with newly cut delrin pieces
• Throw the bladder dive system out and instead DIY a piston tank dive system - basically put a piston in the hull in place of the old bladder
• DIY a piston tank controller
• install a brushless motor, LiIon batteries and set sail

An ambitious list.....but what the hell ....it is my 3rd sub after all

First step: start drawing up the parts I need to make in CAD.... Stay tuned (and happy to receive comments and suggestions).

cheers,
Eric

[IdefixRC]

First drawings done.
- Center rib - needs to be glued in. Will hold the piston assembly (basically the plate on which the motor and the switches are mounted. The piston will move inside the tube behind it - open end on one side to draw water in)
- First shot at the end cap. Still needs a lot of work with motor assembly and pushrods etc...

Three question I would have for the seasoned Sub builders here:
1) what O'ring material is recommended for the endcap and the piston? The body tube is clear PC or Acrylic I think. The Piston will slide directly in it.
2) The body tube has a 110mm OD and 104mm ID. What O'ring size would you recommend?
3) I believe I will need to cut the endcaps a bit smaller - maybe with a 103.5mm diameter? How deep should the o'ring groves be? or how far should the o'ring be compressed?

Thanks !

[vasily]

Here's a link to the O Ring store, in the US. http://www.theoringstore.com/ It has good size charts and groove dimensions. You're looking for BUNA rings. If you're using a lathe to turn the endcaps (I hope so) aim for an "interference fit" which can be about -.008 below the ID. I find that making the last cleanup cuts with the crossslide set to about a 2 degree angle lets me ensure a good fit before cutting the grooves.
I'm a lousy amateur machinist, and I used much material to practice. I found acetal rod on ebay after diligent shopping.

[IdefixRC]

Thanks Vasily,
I will check the webpage out. And yes, I plan to turn them on a lathe.

If I understand you correctly I should have 0.008 inch (~0.2-0.3mm) clearance between the tube ID and the piston/cap OD.
On top of that I will add the groves based on the info provided on the webpage (looks like they recommend an oring squeeze of around 20-30% depending on whether it is a dynamic or static application....more research required :-) ).

[vasily]

You know, I just re-read your post and found that you want to use these as sliding pistons, not end caps. End caps should be fairly snug, but pistons need more clearance. Once the turned piston will fit snugly into the ID of of the tube, take another .005 to .010 to allow for a sliding fit. Using an o-ring with a larger cross-section (like US 3/16") and the shallower side of the o ring groove will keep t watertight.

[IdefixRC]

Thanks Vasily,

correct. I will have one double O-ring static endcap on one side and a dynamic double o-ring moving piston on the other.
I found this good page to calculate the right dimensions for both O-ring and the groove: https://oringcalculator.eriksgroup.com/

I also put some more time into the drawings.
One design flaw I found when adding the rear and front hull elements: it will be very difficult to get to the 2 rear nuts to open up the complete compartment.
While the rear hull piece has an hatch on the bottom from which the nuts could be reached, this is way to fiddly for my taste.
Only other 2 options I could think of:
1) make a bayonet type of mount on the front bulkhead - the one that will be glued and to which the piston motor assemble will be mounted. Issue: The complete assembly would need to be rotated inside the tube plus it will be hard to get this to sit accurately as the tolerances will be quite tight.
2) make 2 hatches in the rear hull aligned with the nuts so they can be accessed easier. I could make them spring loaded so that they close from the inside out after the nuts are in place. This will work but I have to think of how to place the seal over the M5 rod. Maybe I need to embed it into the rear cap.....

[Guest]

With regards to compression, you should aim for 10-15% o ring thickness for your cylinder seals, and no more than 10% for the piston tank, otherwise friction will be very high. The specs given on o-ring sites are designed for uses where the seals will experience much higher pressures.

[IdefixRC]

Thanks. I work with that.
When selecting the O-ring should I select one that fits exactly over the piston or make it a it undersized. e.g. Grove has a diameter of 96mm, O-Ring should have an internal diameter of 96mm, 95mm or even smaller? Guess it will reduce the cross section a bit when stretched. so maybe aim for one that fits snug - e.g. 95mm in the example above?

[Guest]

O-rings are specced to the internal diameter, plus two times the o-ring cross sectional thickness.

Therefore, a 96mm o-ring of 3mm cross-section would be 102mm overall diameter on the outside. In addition the o-ring should have a bit of 'snap' so it sits in the groove with a small amount of stretch, say 1-5% of the specified diameter, with about 2% being ideal.

So, for your 104mm tubing, a 3mm thick o-ring should be adequate for both the endcaps and the piston tank. For 10% squish on the endcap, that will need to push the o-ring in 0.3mm each side, 0.6mm overall. So we subtract 5.4mm from 104mm giving 98.6mm, in addition we want a small amount of stretch on the o-ring of between 1-5% as mentioned earlier, therefore a 96mm o-ring should be adequate, although rings from 94-97mm should be okay if you have trouble locating that size.

As plastic tubing tends to have a wide tolerance on the inside bore, it's wise to machine the endcaps to fit the pipe, then machine your groove the required depth, which in this case would be 2.7mm deep (to give the overall 5.4mm figure mentioned earlier).

Once the groove is machined, you can take a whisker off the endcap diameter so it slides in easily without risk of jamming.

If the tubing is a very poor finish, then you can increase squish to 15%, but I wouldn't go beyond that, otherwise you'll have difficulty removing the caps by hand.

You can go for larger o-ring cross section, but bear in mind this will also increase friction. Softer rubber compounds are available, but anything other than 70-80 shore will be difficult to find. Silicone, neoprene and nitrile (also known as Buna-n in the states) are all suitable compounds for use. Silicone seals are bit more resistant to o-zone degradation, and tend to be a bit more squishy, but you mustn't lubricate them with silicone grease as it makes them swell.

Couple of final notes, try and keep the piston tank volume to within 15% or less of the overall WTC displacement. This keeps internal pressure to sensible limits without loading up the motor too much and compromising your shaft seals. Try and keep the tank centrally located rather than offset at one end, otherwise trim will be a problem.

[IdefixRC]

Absolutely awesome !!! Thanks a million. The 15% volume hint is a good one. I'm currently above that which could cause trouble as you pointed out.
Will also see if I can do something about the central location although this might be an issue given the overall size of the hull and the fact that I can not route connectors/etc. through the outside as the WTC outer hull also forms the subs hull. Only option would be to install a smaller piston tank inside the WTC..... decisions decisions

Will update with the latest set of drawings soon (currently working on the motor assembly. For the shaft seal I have one of these combined with a ball bearing on the inside: https://www.engel-modellbau.eu/catal...jss475tvcachs2)

[Guest]

Wouldn't be a bad idea to make the piston tank a smaller diameter. The larger diameter will make it more difficult to effect a good seal, as the tolerances increase with an increase in diameter.

I wouldn't think a huge tank is required for the shark. An alternative to making one yourself would be a secondhand Engel tank, and trim it down to the required volume. Also Ark model do some piston tanks which look okay to me, but can't vouch from personal experience. Commercial pneumatic seals work very well with piston tanks, and reduce friction over o-rings e.g. Airzet, cup seals a la Engel etc.

The snag is they tend to be restricted to certain diameters, and the biggest challenge is finding tubing with a decent internal finish and size. The very best tubing I have found is spent caulking tubes used for silicone rtv, grab adhesives etc. These are made from polypropylene which is tough and highly water resistant, and they have an engineered finish internally- very accurate. As you can find them thrown out by builders, cost is zero (recycling?).

The snag is the diameter tends to be restricted to 46mm, okay for smaller boats or a trim tank, but a tank any larger than around 200ml is likely to get a bit unwieldy. I have on occasion found tubes with larger diameters in the 55-60mm range, but beyond that you have to look at other sources.

You could of course adopt a hybrid ballast system where you incorporate a main ballast tank which is vented, with a smaller piston tank for trim and going negative. That would enable a high waterline, without the inherent disadvantages of compressing air in a fixed space, and allow you to make a smaller piston tank which you may find a bit easier

[IdefixRC]

Hmmmm......some thinking required....back to the drawing board
One thing is for sure though: I will stay far away from Arkmodel stuff after all the issues I had with them in the past. Extremely unreliable dealer.

[Guest]

You could keep it simple by building it as a dynamic diver, but leave a space inside to fit a tank at a later point. The shark should work fine as a dynamic diver, as it has large control surfaces, and doesn't require a scale waterline. It'll also spread the cost. Ballast systems do add a layer or two of complexity and when you're just starting out, it's not really what you need.

[IdefixRC]

I will try my luck with the tank. After your comments around the piston tank diameter and the tolerances, I think it is better to go with the internal tank approach than trying to incorporate the piston directly into the hull. Plus it will help with the trim as you correctly pointed out.

Keeping the 15% rule in mind, I think a 80mm OD, 74mm ID acrylic tube (any other tube material other than acrylic or delrin is hard to get over here...) with a length of 140mm should work.
Taking the caps and seals into account it should end up at around 15% of the internal volume and can still be fitted rather close to the center (a bit of due to the piston rod movement and the fact that the complete internal hull has a length of 384mm only after installation of the 2 endcaps.

Increased the complexity a bit but at least solved some of the issues around cabling as well as the problem with the rear fasteners as the complete hull can now be screwed together through the nose cap.



P.S. has anyone tried to 3D print similar caps and pistons? I will give it a try and print them in ABS (FDM) once the O-rings have arrived.
While warp can be controlled, material shrinkage might be an issue and would need to be accounted for. Let's see how it turns out.

[Guest]

Okay that comes to about 600ml piston tank displacement, it'll be a bit less as the piston and tank caps will take up a bit of volume, so lets say 550ml. Do you need that much volume? I'd have thought 3-400ml would be ample for a boat that size.

3d printing is something I have no real experience of, but it is a very interesting technology. I would say a lathe and some stock material would be a preferable solution- cheaper, stronger and more accurate. Also you will still have to fabricate a spindle nut for the jack screw, and that isn't really within the realm of a 3d printer.....yet.

Matt Thor has been running a series of articles detailing his experiments with 3d printing parts for a WTC including a piston ballast tank. You may want to consider putting your hand in your pocket and joining the SC, as it's only $10 and it gives you access to all the back issues of the SCR. Enormous amount of information contained there.