RC Boat - Pump Jet Boat / 遥控船-喷射泵

thingiverse

I'm going to preface this by saying that I have no experience in boat design or RC in general. That being said, I am an engineer and I think I figured most of it out...if boat flips, porpoises, or otherwise does a barrel roll don't say I didn't warn you :P. If it does any of the above and you are a clever kitteh and have some tips based on real experience feel free to help me out. I couldn't decide if the boat wanted to be a trimaran, rigger boat, flat bottom, ground effect vehicle or stepped hull V so I mashed it all together in an unholy union of boat design sacrilege - your welcome. I modeled water displacement and Fusion 360 calculated the mass, because it is such a narrow hull it is critical to keep the print weight low and rely on the carbon fiber tubes for most of the strength. I am targeting a total weight around 700 grams for everything, motor and batteries included. Printing with low layer count (maybe even single) / low infill should reduce the mass of the boat to acceptable levels, using epoxy resin or wax on the finished boat should help with water tightness. I'm still experimenting with my prints, I'll let you know results and recommended print settings for each section once finished. The 2212 motor is a marvel to me, how is something so awesome so affordable? I wanted this boat to be affordable first, performance is secondary. While an outrunner is atypical for boats I think I've designed out the cooling issue via internal airflow and motor cooling impeller. Besides yeeting boats around at 200kph seems like a dangerous way to accidentally impale swimming googies...a project for another day. Scroll down to post processing pictures for detailed build instructions - currently updating. Certain components can only be installed before melding of the hull sections including the motor assembly - do not glue hull together before those components are fitted. Josh at Flite Test did a good vid about waterproofing RC electronics with Peter Sripol (Flite Test | Waterproofing Electronics) 11/6/2020 Performed a submersion test with sanded model - due to the way walls interface with fill there is capillary action between the individual lines of printed material at the base of the sponsons. This causes water to slowly seep into the model. I've read that brushing or submerging in melted candle wax (not too hot though - Paraffin starts melting at 37 deg C / 99 deg F) can seal a 3d print. If you do use wax keep fire safety in mind as some FDM plastics are flammable and wax has now made it even more flammable. Another method that I've seen RC airplane makers use for waterproofing is to coat model in a very very thin layer of gorilla glue - it doesn't bubble up too much if the layer is thin enough. On the more expensive / specialized handling side RC modelers will sometimes coat boats with polyester or epoxy resins - the fumes require special protective equipment. I tried System Three epoxy resin clear coat recently and it gives ok results, there was still some water leakage after a few hours in the tank but that may be an application error on my part - will try again and see if I can get a single coat water tight hull. I've also read that it is possible to get a print to be water tight straight off the printer but that requires larger nozzle diameters than this design will allow for (too heavy). If anyone in the RC community has any suggestions they would be welcome - maybe when you post your make you can let us know how you waterproofed your model :). I'll post pictures of my own process soon. With the latest updates to stator and nozzle the pump reliably runs at all thrust levels - sealing of the inner running surfaces of the pump inlet via wax or epoxy is recommended for reliable operation. 7/29/2021 Finally had a chance to test this in a pond! The old air intakes were positioned too low and risked water intake. I've already completed redesigning McBoatface and uploaded :). If you're still using the old model it seems to operate better at full throttle. Avoid sudden deceleration on the old model to prevent the front from dipping and flooding the air intakes. BOM: - Motor: 2212 Though, with some work, you can now use an alternate motor with the custom motor plate template step file (Dimensions center to center between the 3 mm mounting holes are 49 mm X 23.41 mm. The top surface of the mounting plate (measured from the large flat area not the post) is 18.738 mm below the center-line of the propeller axis). Keep weight and cooling considerations in mind if using an alternate. If using the 2212 just download the original 2212 Motor Mount. (I don't have enough experience to say what KV level is ideal, I am leaning towards lower KV motors as they deliver the same torque at lower amps with higher KT values. We like KTs - get it...KTs?) https://www.aliexpress.com/item/32672300112.html or https://www.aliexpress.com/item/1904754775.html - ESC: Water proof ESC - Prop: 32mm with 100 mm shaft (note get the 10 cm one with 3.18 motor as the 2212 shaft is 3.175 mm - I initially tried to model an axial flow impeller but I found this purchased item to be far more sturdy (*update as of 3/21/2021 there is an experimental two stage impeller uploaded). Advantages are it has built in high speed bearings and replacement props can be purchased from here as well. I also added provisions for a diy 3mm shaft using 623 bearings) https://www.aliexpress.com/item/32714017489.html https://www.aliexpress.com/item/32528668718.html - Servo: 9g servo (important you will want to waterproof or find an alternate waterproof servo) if anyone knows a reliable and waterproof 9g model please post in comments - Reinforcement: 6 mm OD 4 mm ID Carbon Fiber Rods https://www.aliexpress.com/item/32310771446.html - Power Source (btw if, like me, you are inexperienced with RC make sure to inform yourself about some of the precautions needed with lithium based batteries especially around charging and storage. Unlike your consumer electronics the LiPo cell is very much exposed in RC and special care must be taken. RC Review and FliteTest do some good youtube vids on the matter) : - Cat Grumpiness - 3S LiPo power pack (I was able to find a 1300 mAh one that fits well 3S 1300 mAh 60C LiPo 75 X 34 X 24 mm - The maximum extents of a battery able fit in the front of the boat are 82 X 36 X 26 mm if removal is not needed - the 75mm battery I purchased can just barely be inserted and removed through the hatch) - 18650 batteries are no longer being supported, battery compartments are left in but I've come to the conclusion that its just not a good idea..I originally thought it would be cheaper owing to relatively inexpensive cells and BMS modules but once you add a constant current power supply for charging, nickle strip welder, and discharge rate / weight / C.G considerations LiPo starts to make a lot more sense...never mind the potential effects of water on Li Ion cells... - Receiver and transmitter: Up to you (and your local regulations), I'm using a Flysky FS-i6X transmitter and FS-GR3E Waterproof receiver (note i6X will need to be in AFHDS mode to work with FS-GR3E) - M3 X 6 X 5.3 Embedded Nuts (old link no longer works - possible substitution M3 X D5 X L6) https://www.aliexpress.com/item/4000232858343.html - M3 Machine Screws (Button head and Flat head) https://www.aliexpress.com/item/32798146322.html or stainless https://www.aliexpress.com/item/32934186482.html - XT60 connectors, 3.5 mm bullet connectors, grease (marine), filament material compatible glue. The connectors are sold by the same storefront as much of the material above. - Sound system: Bongo Cats Tentative Print Settings (Cura has a setting under Travel called combing mode which helps reduce stringing in the inaccessible interior). Personally I'm printing in PETG but ABS or ASA might be a good options too if you are setup for that with enclosure and ventilated space (due to easier gluing / reduced density - if using ABS or ASA I would advise against acetone smoothing as thin walls will likely deform). I excluded PLA as it is not as temperature / weather / UV resistant as the above materials and will swell / deteriorate in water unless waterproofed. I print at between 30 mm/s (TPU) and 50 mm/s (PETG) but it might still print well at higher default speeds. Thrust vector nozzle, stator bowl, and 2212 cooling impeller will be printed in TPU for its reliable layer adhesion as these components will be subject to cyclic loading and high shear stress - I've tested the 2212 cooling impeller in PETG as well. STL files are already saved in as print orientation. *Note I've noticed that upon updating Cura that support generation varies between versions so you may need to tweak support overhang angles from posted values below - make sure to do a preview to check for proper support and alter as needed - values below updated for Cura 4.10.0 - PETG/ABS/ASA Parts - Boaty (PETG/ABS/ASA): 0.4mm nozzle, 0.16mm layer height, 1 walls, 5 bottom layers, 5 top layers, 16% grid infill, supports on overhangs over 70% touching buildplate (apx 116 grams) - McBoatface (PETG/ABS/ASA): 0.4mm nozzle, 0.16mm layer height, 1 walls, 5 bottom layers, 5 top layers, 16 % grid infill, supports on overhangs over 78% touching buildplate, brims are helpful for ensuring support adhesion (apx 218 grams) - Cover: (PETG/ABS/ASA) 0.4mm nozzle, 0.16mm layer height, 1 walls, 3 bottom layers, 3 top layers, 16% grid infill, supports on overhangs over 60% touching buildplate (apx 12 grams) - 2212 Motor Mount (PETG/ABS/ASA): 0.4mm nozzle, 0.16mm layer height, 5 walls, 3 bottom layers, 3 top layers, 25% grid infill, supports on overhangs over 70% touching buildplate (note the advanced option "Enable Support Interface" in cura 4.10 creates a fairly dense support structure that can be a challenge to remove - you may wish to disable for this part), brims are helpful for preventing warping (apx 30 grams) - Ride Plate: (PETG/ABS/ASA) 0.4mm nozzle, 0.16mm layer height, 5 walls, 3 bottom layers, 3 top layers, 25% grid infill, supports on overhangs over 70% touching buildplate (apx 6 grams) - Servo Hatch: (PETG/ABS/ASA) 0.4mm nozzle, 0.16mm layer height, 4 walls, 100% infill, supports on overhangs over 70% touching buildplate, brim (apx 2 grams) - Stage 1 / 2 Impeller / Mid Stator / (2212 Cooling Impeller): (PETG/(NYLON might be a good option too)) 0.4mm nozzle, 0.06mm layer height, 3 walls, 100% infill, supports on overhangs over 50% touching buildplate, brim - TPU parts - Thrust Vector Nozzle (Open): (TPU) 0.4mm nozzle, 0.16mm layer height, 3 walls, 3 bottom layers, 3 top layers, 25% cubic subdivision infill, supports on overhangs over 70% touching buildplate (apx 15 grams) - Stator Bowl (Axial): (TPU) 0.4mm nozzle, 0.16mm layer height, 3 walls, 3 bottom layers, 3 top layers, 25% cubic subdivision infill, supports on overhangs over 70% touching buildplate (apx 24 grams) - Stator Bowl Gasket: (TPU) Solid - 2212 Cooling Impeller: (TPU) 0.4mm nozzle, 0.16mm layer height, 2 walls, 4 bottom layers, 4 top layers, 100% concentric infill, (apx 2 grams). - Motor (50 grams), LiPo (129 grams) (18650s are out), ESC (28 grams), servo (9 grams), and receiver (6.5 grams) A note for early downloaders: I tend to iterate and reiterate a design multiple times before I'm happy with it. Designs subject to change...frequently... N7 Cat Says: "Dis my bigges and shiniest objec yet, 2020 u cray cray, I jus gonna yeet away from u in mah boat wif mah other kitteh fiends" Updated: 9/28/2020, Updated files Boaty and Cover, added provisions for gluing a 5mm X 2mm magnet on the cover and boat for magnetic latching Updated: 9/29/2020, fixed a problem with McBoatnose missing modeled tube for carbon fiber reinforcement. Updated: 10/2/2020, added a custom motor plate "McBoatface Custom Motor Plate.stl" for anyone wanting to use a motor other than the 2212. Intent is that you will be able to model your own motor housing (and share with others along with recommendations for ESC and battery :) for mounting onto the flat plate. 3 mm dia screws will secure the motor mount to the two sections of the boat, as the mount will be subject to vibration fastening hardware (embedded nuts or lock nuts) is recommended in your design. Like the original requires brims and supports. Updated 10/4/2020, added battery holder compartments along the length of the hull for an integrated 3S 2P 18650 setup. I'm planing on spot welding the batteries together with pure nickle strips and wiring to BMS. Note there are two battery caps (non symmetric). Added a baffle near the motor cooling impeller to improve airflow and cooling. Additionally minor improvements to printability around the servo mount and filleting of vortex outlet. For custom motor users new battery holders and air baffle might interfere with especially big motors but you can model / clip them out if needed. Updated 10/6/2020, added built in support for servo hatch opening, improved joints between the three boat hull sections by adding grooves and other features for the expanding polyurethane glue to latch onto. Fixed a printability issue with the fillet that supports the propeller shaft. Started first print of the boat section, may update if issues crop up after print finishes. Updated 10/7/2020, added a channel in the battery compartment to allow for wire to be routed through. Updated 10/8/2020, post printing improvements, increased diameter of battery compartment, prop shaft and carbon fiber shaft openings. Improved servo hatch support structure to make it easier to remove. Update 10/9/2020, Estimated static resting position of the boat in water and calculated rough target mass for finished boat. After a print failure of the protective grid on the intake as well as impact durability test failures I have replaced with a filament heat stake method as shown below in post processing pics. Insert 1.75 mm filaments into the slots and use a low temp soldering iron to melt the ends into the inlet - you could also use servo wire and glue in place...unless you like grinding baby fish into a pulp...u monster Updated 10/11/2020, with 18650s looking like less and less of a good idea and mass being more of a concern I trimmed the battery compartments down to two 0.4 mm walls. This reduces mass and print time dramatically. Update 10/11/2020, realized I derped and forgot to add an access port for servo wiring. I've created an access port on the battery compartment closest to the servo wire (port/left side). As there were some unbridgeable overhangs I made some built in supports that need to be pried or clipped out before use. Update 10/12/2020, created an alternate path for servo wire on the top of the baffle in case I decide to house all electronics in the cover. After fighting with carbon fiber tube on second print increased the diameter of the carbon tube channel and started printing again :P Ever had a carbon fiber sliver? I haven't nor do I ever want to...make sure to wrap the end of the tube that you are handling in masking tape or something to protect your hands Updated 10/14/2020, minor improvements to Boaty and McBoatnose to reduce support generation. Trying some new tricks to strengthen servo mount and embedded nut support by creating additional walls in those areas. Improved airflow on McBoatnose 2212. Updated the stator bowl to fix supports not properly generating Updated 10/15/2020, McBoatface 2212 motor mount was a bit of a tight fit, adjusted motor mount. Adjusted joints between sections Updated 10/16/2020, added receiver tray and antenna port to cover Update 10/18/2020, added mounting points for turn fin in case they are needed - I couldn't find any info on whether one would be needed for a pump jet nor dimensions for standard mounting points. Converted the fixed ride plate to a variable trim ride plate using 1.5 mm servo wire as the adjustable retention mechanism. Backup retention provided by an embedded nut and M3 X 10 screw + spring. Joint held by filament joint Updated 10/20/2020, some weird artifacts were cropping up in cooling impeller during slicing, increased to high resolution stl and fixed wall width. Encountered a problem with McBoatnose tips coming off due to single walls and how narrow it becomes. Blunted the tips. Also had to physically clip some of the joints between front and mid section in order to get it to fit, adjusted the model so this is not required. Carbon Fiber tubes now provide most of the support between the front and mid section. Created McBIGface for those of you with Z axis build volume above 280 mm so you can print mid and front sections as one piece #jealouscat (note that this will result in a weaker motor mount as you will still need to balance weight considerations with strength requirements - play around with slicer settings starting from McBoatnose as a baseline) Updated 10/21/2020, increased prop clearance in stator bowl Updated 10/22/2020, Major overhaul...N7 cat shredded all the toilet paper...McBoatface and McBoatnose now one part, motor mount now modular for both 2212 and custom motors, added a custom motor mount template, increased sponson size 25%, other minor improvements Update 10/24/2020, tried to reduce weight of McBoatface by hollowing out the sponsons, turns out a solid object with grid infill doesn't contribute much to weight...who knew...saved 5 grams! Update 10/26/2020, realized that the cover / McBoatface interface would allow water to seep in at the front seam, updated these two parts to improve water shedding in this area. Added a part called Compatibility Shim For Pre Oct 22 Update Boaty Prints.stl which will allow any prints of Boaty made prior to 10/22 update to be used with new McBoatface Update 10/27/2020, major iteration work completed - only minor changes from this point on - latest files re-uploaded to sync with my local files. Post processing instructions will be completed once prints are completed Update 10/28/2020, vortex was breaking during print, strengthened the part with additional walls Update 11/4/2020, updated servo hatch (missing cutout for gear) Update 11/5/2020, work in progress status lifted - no major issues encountered during assembly Updated 11/6/2020, after dropping my completed boat noticed one of the sponsons cracked ;.; strengthened the sponsons with option to add 200 mm carbon fiber tube reinforcement (adds ~35 grams to total weight inc carbon tubes or ~ 15 grams without carbon reinforcement). Thanks to a helpful tip from the community I have added a 5 degree pitch up to the thrust vector nozzle - if your boat has a nose down attitude and the ride plate adjustment is not enough this should hopefully correct that. There is a moment force applied by the thrust which pushes the nose down, this minor tweak will hopefully apply a slight corrective force which allows the boat to get on stable plane. Added a ride plate servo wire retention slot to stator bowl. Increased McBoatface carbon tube clearance. Added a 10 deg pitch nozzle in case 5 deg is not enough, as I'm still waiting on batteries to arrive I haven't had a chance to perform test to determine which will be best. Performed submersion test and posted results near the top of this summary Update 11/8/2020, moved the Boaty magnet catch to the top surface instead of underside to improve ease of assembly. Closed the opening in the baffle since servo wire was long enough to not need it. Moved the cover magnet closer to the Boat magnet to improve latching strength. Improved McBoatface by filleting the sponson to wing interface and creating water rejection channels near the vortex exits. Update 11/15/2020, added a version of cover with provisions for 3mm LED navigation lights Update 11/18/2020, first functional test of the waterjet, I found that even with TPU gasket air and water would spray out thus sealing the stator bowl to boat body with silicone or glue is necessary. I found an issue with air locking - I don't fully understand the cause but experimentally determined a way to resolve this via a pending modification to the thrust vector nozzle. Update uploaded (Thrust Vector Nozzle Variable Pitch) but untested - I'll hopefully get a chance to print and test by tmw. Testing confirms the air lock issue is resolved with this new nozzle - there is some loss of efficiency. I don't really know why this fixes the air lock but I suspect the stator bowl fins were not redirecting enough of the rotational energy generated by the prop resulting in an area of low pressure which would draw air in from behind the outlet...without high speed cameras or CFD I can't really say. N7 Cat says "wut he trying 2 say iz I putz mah paw ober da nozzle and dat fixes it" 11/21/2020 did some more study on propulsor units and updated the pitch and number of stator vanes - 5 vane stator uploaded - will test later and delete original if testing displays improvement. Test showed marked improvement in both trust and elimination of air lock using the 5 vane stator - original deleted. For anyone else designing or trouble shooting other pump jets on thingiverse the number of stator vanes needs to be a prime number and different than the number of prop blades otherwise a standing wave can develop. At some point in the future I might experiment with a properly designed axial flow prop and stator - salt sintering might be able to create a strong enough part for use as a prop. 11/25/2020 reverted thrust vector nozzle to neutral pitch 2/16/2021 Added a printable gasket 3/17/2021 modified the thrust vector nozzle to try and fix an intermittent airlock issue 3/19/2021 tweaked the thrust vector nozzle and modified the stator bowl by extending parallel fin surfaces in attempt to reduce rotational energy of water 3/21/2021 experimental two stage impeller / stator uploaded - my first try at a printed impeller - going to test TPU and PETG but am skeptical about the robustness of either. Note there are two versions of the second stage impeller 3/23/2021 fixed some clearance and mounting issues with the experimental impellers 3/25/2021 fixed some clearance issues with the mid stator causing it to bind to the second stage shaft 3/25/2021 had a chance to test the two stage impeller system and found flow was disrupted past 40% throttle - I suspect the pitch angle is too aggressive - removed files until I can remodel all the components 3/30/2021 tested stage 1 impeller and thrust feels like it has doubled :D moar powah! Will test the two stage setup soon. For now stage 1 works with existing parts and improves thrust 3/31/2021 tested two stage setup and it was overall a success, flow rate is higher with just the single stage but priming is much faster with the two stage setup. In fact device will prime even under full thrust, making for more reliable operation in choppy water. Two stage only functions with the new stator bowl axial, if you are only using the new stage one impeller it will still work with the old stator. Flow rates were getting so high that the nozzle started to impede flow, an open nozzle is now uploaded 6/22/2021 strengthened the servo mount and motor mount screw holes on part "Boaty" 6/25/2021 made the supports for the servo hatch on part "Boaty" easier to remove 6/30/2021 improved motor cooling impeller, larger fins, improved filleting, removed thin shroud, now secured by butting against cardan joint 7/29/2021 after testing on a small body of water discovered that the air intakes were positioned too low resulting in water ingress during rapid deceleration. Redesigned McBoatface repositioning air intake higher and adjusting wing profile. 7/31/2021 cleaned up stl files by removing some variants and the compatibility shim 8/1/2021 updated 2212 motor mount to allow wires to pass under 8/5/2021 improved motor cooling impeller profile 8/8/2021 uploaded an experimental rigger style front (McRigface :) - not sure if the profile is correct - will test it soon 8/10/2021 improved servo ease of installation 8/11/2021 still having issues with front of old hull dipping into water - old hull deleted and McRigface is the new McBoatface. Stage two impeller was shearing from the embedded nut - altered stage two impeller so that embedded nut enters from other end and added some more material to the embedded nut hole to hopefully prevent this 8/12/2021 wasn't happy with the thin walls on the 2nd stage update - moved the motor 10mm backwards and redesigned Boaty and McBoatface to accommodate. New 2nd stage impeller, Boaty and McBoatface loaded 8/14/2021 created internal baffle system in the sponsons 8/15/2021 unifying parts between boat project and pump jet. Updated water cooling stator bowl, uses embedded nut for more secure water nipple retention. Water cooling is useful for more powerful inrunner motors and associated water cooled ESCs. A new cover with holes for 3 X 5 water tubes is now uploaded 8/24/2021 added a direct vent version of boaty with a leaf blower inspired housing for the motor cooling impeller - increases cooling efficiency