All engineers should be schooled like this one- to experience the disappointment and frustration of mechanics that repair and have to service the equipment they design.
I forget the exact make & model but my uncle had a US brand car with a Mitsubishi motor which I was going to buy off him. Turned out that I would need to pull the motor just to change the spark plugs…lol.
@@tomchelle1That would almost certainly be a Dodge Stealth. They were basically the same car as the Mitsubishi GTO/3000GT, and had a transversally mounted V6 whose rear bank was so far into the engine bay/close to the firewall that the plugs were essentially impossible to remove without pulling the motor out of the car.
Most engineers already do this stuff during school. You have to do a capstone project your senior year. Engineering students now have to cover more material in less time. School is just meant to be a foundation and you learn the rest on the job.
This reminds me of Pinewood Derby. Someone will nail it by studying the test vehicle and getting the number right for the prop by optimizing for the test vehicle.
We did this with the launcher mechanism in pinewood derby when I was younger. There was a single dowel rod in the middle of the track the would fold down and release the cars. We figured if your car passed all the length width and weight regulations but had a nose that was high up and just barely touched the tip of the dowel, your car would start moving immediately while all the other cars would only move when the dowel was all the way retracted. Optimized this design in following years with a similar high nose design but with a channel cut out of the length of the car that allowed the car to gain almost an entire car length head start on the other cars as soon as the dowel was retracted a tiny amount.
@@brianwelch1579 cry more because you didn’t think of it. We designed for the track. That’s what every racing team does in real life. Real life lessons learned during cub scouts make me more clever and smarter person today.
@@brianwelch1579 Racing is based on finding gaps in the rules to gain an advantage. If it's not in the rules, it's not cheating, and even then, it's almost encouraged to push the rules to their absolute limit to gain an advantage. Every professional racing series is like this.
Back in the early 70's I had a 13' Boston Whaler with a 40hp Johnson. In those days you had a limited choice of aluminum props and pitches. With the never ending search for performance and speed, this company (Columbian) came out with this "variable pitch" plastic prop. They were available in different colors, I remember having some red and blue. They were also advertised to be unbreakable (the company should have contacted me for testing first). They were constructed out of Lexan with a bronze hub. The pitch would "flatten out" under acceleration to help get on plane. This helped when towing skiers or to get going with like 4 people on board. Once on plane, you could back off on the throttle to more of a cruise speed and save gas (even though gas was cheap back then). The problem was that after some time, the Lexan would fatigue from all the flexing and a blade would snap off. Of course you would have to limp back home real slow because of the vibration. After running and breaking like 3 of these lexan props, I went back to the aluminum style. There's actually a NOS red one for sale on ebay now. This concept for a boat propeller has some advantages, finding the proper material is going to be the challenge
idea for a flexible thin nitinol version of that lexan prop! It would be super resistant to fatigue effects and flexible and able to flex in a similar way.
If you want to run numerical analysis but don't have acces to cfd and a large amount of compute, an alternate approach is using blade element momentum theory. I used a python implementation, PyBEMT to optimize a propeller for my BSc. Thesis and it worked really nicely.
Thrust (in Newtons) * speed (in m/s) is a measure of the power delivered by the prop to the boat. So if you have thrust, and speed through the water, you can calculate prop power that way. Prop efficiency is (prop power) / (shaft power). I think 50 percent efficiency means you are doing pretty well. You can estimate the motor power output using torque (as estimated by VESC) and rotational speed.
How do you get the thrust in newtons of a propellor, is this by doing prior testing and already having an understanding of the boats drag and knowing how much force is required at a given speed?
@@-B_S- Been a while since I watched this video so I don't remember what I was getting at with my comment. But, to answer your question, if you have a real propeller and boat, you can potentially measure the thrust with a load cell on the prop shaft. You can also measure the thrust by tying off the boat to a dock and measuring the tension on the line with a load cell. That may not be the same as the thrust while underway, though, so you have to be careful with that. That is called "bollard pull." If you are designing the boat, you can make predictions about how much power it needs based on rules of thumb or detailed study using computational fluid dynamics (cfd). If you have a hull, you can tow it and measure the tension in the tow line using a load cell. The rules of thumb are based on displacement of the boat and desired speed you want to achieve. So rctestflight could maybe just tow the hull at a few different speeds to establish a thrust vs speed curve, then for each prop he could estimate the actual thrust it is delivering based on speed. That will tell you how much thrust power has been delivered to the boat. Compare that with shaft power (torque x shaft speed) and you can get the prop efficiency. Or at least an estimate of it.
Good stuff! But you really want to think of pitch the same way you think of thread pitch on a bolt. The pitch is the distance the propeller would advance through the fluid (with no "slip"). And this gives you the nominal blade twist automatically. You can then refine the twist with small changes to optimize the span loading. You typically want to unload the tips somewhat.
@@terryboehler5752 I do. Jack sent me a copy of it when I was designing and building a wind powered vehicle that goes directly downwind faster than the wind. I was eager to make the most efficient propeller I could, and hoped he had some good secrets. My recollection is that the book had some good concepts and explanations, but overall it was kind of a mess. We ended up designing the propeller using JavaProp.
Nice video (and competition). I was wondering about the drag on the surfaces themselves, which also adds to a loss in efficiency. What would happen if you would change the surface structure of the blades and cone to very smooth instead of the rough surfaces that come out of the 3D printer? Or even better: make the smooth and very hydrophobic with an additional coating? I'm aware of the fact that this no guarantee for lower drag, because for example golf balls are purposely dented to achieve better aerodynamics. But maybe this is an aspect that you (ar a contestant) would find interesting to look into.
This is going to be a fun one to watch. For anyone else who feels the same you might also enjoy the fan showdown. Similar concept different channel. Can’t wait to see this.
This stuff has always interested me ever since i got my own boat to maximise fuel efficiency. The real key for efficiency is prop pitch a bit of the design but most importantly is the cross section profile of the blade itself like an airplane wing to generate lift but on a prop you are trying to minimise friction as most of your thrust and economy loss comes from vortices generated by the tips of the blades as heat
You need to revisit the diameter, after discovering how-wrong the pitch was, on your 1st batch.. This is part of the principle of looping back *far enough* in the design-cycle, for what one's project is actually doing.. ( Wysocki's book on project management, Extreme & Emertxe projects loop back to "should this project continue?", Agile projects loop back to "What is the next deficit that we remove?", Traditional projects loop back to "what is the next item already-decided that we do?" Research is extreme/emertxe category. )
Did something like this with an ultralight propeller, drew the tip circle as a vertical line, set the rpm (IC motor with 2.36 reduction giving 2650 max rpm at propeller) and worked out helical path at desired 60mph cruise. This gave the ideal tip angle, discovered reducing diameter allowed positive pitch angle of attack, as diameter increased it began to go negative. Research showed aero profile still give thrust, even if aircraft velocity and prop rpm resulted in negative, up to about 2 degrees. Your work is brilliant.
Hi, i dont know if it has already been taken into account, but for a better propeller i think you should take into account also the weight of the vessel, the "hydrodynamic drag" of what you are trying to push/pull, and the result will also change for the cleanness of the waterflow, the best is the pulling propeller layout, if i'm not wrong. Don't know if someone already pointed them out, hope this is useful.
Daniel, often times, the power efficiency in an electric set-up is dependent on how well paired a load is with a motor, and where you are in the efficiency curve of the motor. Have you considered measuring watts as Torque * RPM, rather than V*I going to the motor, in order to exclude any efficiency variation in the motor?
You also should calculate the optimal hull speed for your catamaran, since it works in displacement mode. And after that test the props for that speed.
What brilliant idea for fun competition, there way smart people than me that hopefully come up with some great designs ? I like ask few questions ? 1) all our designs limited to 3 blade like yours were ? 2) What about surface drive props that may be better at high speed ? can you change the depth they run at ? 3) Could someone make variable pitch prop ? not sure if you could design passive system at this scale Am really Looking forward to the results video in few months love jenifer xxxx
If you circulate water in a loop you can get dynamic thrust data indoors 🙌 and I bet you could tune the loop to match drag from a boat..... although actually please keep lake testing it's more fun to watch 😉
Would recommend trying out configurations in onshape for this sort of thing. They work really well for switching between models while changing a bunch of dimensions and are much easier to set up than I expected.
"The 1930 Robertson Waterplane ground-effect vehicle powered by an outboard motor". There are pictures and videos, but I don't think I can post a pic in the comments. I saw this the other day and immediately thought this might be something you'd build or at least be interested in knowing existed, if you don't already. In other news... it's getting to be grass cutting season. I hope you are working on your next autonomous mower contraption. I'm a long time subscriber to the channel and grateful for your willingness to share your talents with the world. Keep up the great work! - Stanford
it's less fun because you'll actually find the sensitivity of each parameter pretty quickly, but taguchi's method will let you use the data you already have to explore real impact of the changes. turns out the one variable people doesn't always have a big impact. wouldn't it be nice to know if an incidental change unrelated to the control had a huge impact?
Another thing to consider for efficiency is rudder location. Offsetting it would reduce drag since the prop wash is effectively moving faster than the surrounding water. Then again it would potentially reduce it's ability to turn the boat for the same reason. Better yet you could make the motor pylon itself rotate. Outboards and tug thrusters are obvious examples.
I took an almost identical approach trying to design a propellor for my electric outboard motor for my sailing dinghy. It's super interesting seeing how sensitive the diameter to pitch relationship is. I was also trying to find an optimal propellor for recharging my battery when under sail.
i would HIGHLY recommend looking into multivariate experimental design! there is a video by nighthawkinlight that explains the basics. as destin also mentioned in his comment about DOE this would be a great situation to do some really cool experimental design and data analysis!
Varying only one variable may not be what you want to do. It may happen that the pitch and diameter are both need to be changed. Vary rarely the geometric variables are uncorrelated in fluid dynamics. I would suggest to print multiple variants where all variables are changed. Diameter, pitch, thickness, the size of trailing edge serrations. Then you can use something like linear or logistic regression to fit the efficiency data using those variables, then you can use the fitted function to find the global minimum.
In addition, the boat hull should not distort the experiment. You use a waterdisplacement hull. A planing hull can change the water resistance extremely. This can be achieved, for example, using sliding steps or smaller areas with water contact. Ultimately, with a wing you have more glide and very little waterdisplacement. so its hard to compare
I was a first responder in the United States Coast Guard and drove several boats. HANDS DOWN - my 18-foot jet boat was far and away fastest. We used it to intercept "cigarette" boats that violated President Nixon's 1 square mile off-limits area on Key Biscayne Bay. 102 mph on flat water with slight (1" or less) ripple. Try designing an "impeller and watch your little boat FLY!
I saw a video by smarter everyday where he went to the coast guard and got to ride their new unique jet boat. It had multiple independently controlled movable jet engines which allowed them to conduct extremely unique movements in order to help with rescues and other stuff. It was the coolest most maneuverable boat I've ever seen
@@WhatDadIsUpTo it's really cool to hear a little bit about your past experiences. I think you will enjoy the video and seeing what modern stuff they now use. I'm curious how drastically different things will be in certain areas. Mostly related to engineering capabilities
In a recent online ADC test, the lowly ATMEGA 328 on the Arduino Nano was shown to be considerably more accurate and full-ranged than the ESP32. Three cheers for Microchip IP.
You need to match the propeller to the motor uniqe torque curve and efficency curve, togeter with the drag VS speed of the boat. Without that, it is only shoots in the dark. Get all the curves and see where they match.
I feel that a healthy part of a monohulls benefit is that is disperses weeds and sea clutter which really does help with these smalle scale props and preventa them getting clogged up.. just a consideration
Unless water behaves completely different from air, center of pressure for a symmetric profile is close to 25%. (In air it jumps to 50% when you go supersonic)
After watching this video (and the linked conversation) I got an ad on another social media platform that showed a close up image of the market’s leading toroidal propeller. Note: It doesn’t have a tapered cone on the trailing edge. It has an open cup with small fins inside. I wonder if the open cup on the trailing edge of their propeller adds efficiency by giving the bits of cavitation on the prop surface a centralized negative pressure space to propagate them into an organized column that can allow for more efficient flow? I haven’t spent any time using computer modeling so someone else will have to test the theory. I hope you find one that suits your needs
Wow, excellent video! Maybe a good way to measure current would be with an induction amp meter, used commonly in a lot of multi-meter style tools? Excellent to see Airshaper in the video, who I highly recommend to anybody looking for efficient CFD studies. Their website is full of interesting studies, detailed in extensive PDFs. In case anyone is wondering about the Onshape propeller template, it is also meant to work with the book of HAVF airfoil vectors I published. So, the onshape model is capable of swapping airfoil very easy, from the currently used ones, to any of 1500+ other popular airfoils. That's part of the reason Daniel is able to change the trailing edge so easily. Airfoil vectors are the future, ditch those DAT coordinate files!!! Now.... to start on my submission.
A current transformer is even less accurate at low currents, a shunt is the best method here. Just need to choose a slightly higher shunt value or different amplifier for better data at low current. He could just replace the shunt resistor on the board he has with a better value, the math is literally just ohm's law.
Sending a wish out into the ether for a prop in the 800 to 1500w range that could power a mini catamaran powered by solar. I truly think something like that paired with modular catamaran hulls would be a commercial success for various applications. Love your videos.
You're testing efficiency on a displacement hull with a specific waterline length, which will not be transferrable to other types of vessels; planing, vings or displacement hulls with other waterline lengths
7:23 seems there are some cavitation behind the propeller, maybe the propeller cone could have another shape also for better efficiency. And change out the wood stick with a thin flat piece of steel plate for less drag.
As for you SQUARE way point mission. In aviation we use both FLY OVER and FLY BY waypoints. Fly over looks to be what you are using in your missions. This causes the craft to float over the point, then turn, then adjust course to get back on course. A fly by uses an algorithm to start the turn prior to the point and intercept the out bound course. Would programming. The mission with a a 4 fly by waypoints reduce the curves in you measurement graphs? I ask this as when a home built aircraft builder is figuring out its actual speeds for a a newly build plane they fly a square box similar to what you have made, but using fly by way points helps to eliminate many errors.
#1 I love the little rotating flag on the servo. It looks like a mini radar. It must signal something. #2. I really like you have made a modular joint for all the props to align with. Next time, include a molded latch into the prop for easy connection and disconnection. #3. I see you have a toroidal prop, what about a tri-toroidal prop? Something like a double stack toroidal. Have you done any testing using a KAPPEL propeller design?
Subscribed to Fan Showdown for a year now, and as RC hobbies guy I am much more interested in competitions like that o) And that is where I might even try to submit something despite being a software developer o)
There’s one more key design criterion you should assess these on! It’s how dangerous the propellers designs are to marine wildlife. Propellers are a huge hazard to aquatic animals, and in Brisbane where I’m from we often find dead turtles, killed by outboard motor props. It’d be great to move towards propellers that are efficient and also safe.
because onshape is giving me a migrane, mine's gonna be exported to Blender, made, and then re-compiled as an STL. (fun fact, the squares in Blender are exactly 1x1m, so I can make sure it's within ruleset while also using software that I can just click and drag with)
Think about a propulsin method like a 'twin screw gear' vise. A central prop turns a cog to surrounding sides. with 3D printing, you could potentially have a global toroidal prop. I wonder if you could also have toroidal impellers inside a tesla valve.
I really enjoy your videos. This is a good one and a very ambitious project. I am concerned that you are doing this the hard way, though. A bit more analysis at the front end can save a lot of work on the back end. It is not necessary to use a complex CFD code to design propellers. Consider learning the two codes released by MIT professor Mark Drela. These are "QMIL" and "QPROP". They can be downloaded for free, along with "concise" instructions. QMIL provides a minimum induced loss propeller design from specified inputs, for a single design point. This code can be used to get in the ballpark in terms of diameter, solidity, RPM, and power. This initial design can be refined with QPROP for off-design conditions. QPROP can include electric motor behavior so that combined motor-prop performance can be estimated over a range of speed and input power. This is a command line code and is not for the faint of heart. In the right hands it is very powerful and can give very good results. A few hours designing can save many, many building and testing hours! I hope this is useful. Thanks again for your very interesting videos.
Try a 3 blade surface prop. So one or two blades are above water line. On your props try adding cup at the edge of the top of the blades. Make the blades more rounded.
Love your videos, and if I may give something. How about having a "baseline" prop that you test each day, so you can get some day to day conversation. Like if day one it does 1watt/gram or what ever and day 2 it does 1.5watt/gram you have a ratio you can normalize your other props with. Just a suggestion. Thank you for your great videos.
Very curious to see if a generative model could take your results and inputs and provide some recommendations for your prop. Then compare against the community designs.
You mean the golf ball dimples? They're actually a kind of 'turbulator' in other words a device that forcibly transitions flow from laminar to turbulent boundary layer conditions. Doing this can be helpful if your design is experiencing laminar flow separation, which can be an issue if (for example) your object is fat compared to its length and the speed is low. Outside those conditions they just add extra drag :(
@nerd1000ify pretty sure he's talking about tubercles. They're the opposite of a dimple, but serve the same purpose. If you want to see them in nature, check out the larger whales.
Looks like you need a bigger/heavier boat. If you can't push the motor to 100%, without the boat planning you can't see what the full effectiveness of a prop design for that hull. Now if you want to make a prop to see how fast you can get on plane, that's a different design criteria.
This is absolutely AWESOME. This is Ripe for a DOE (Design of Experiments) Analysis
Wish you guys could do another collab. You both have wonderful minds. I had to dig to find the video you guys did.
Do you have a video explaining DOE? I’m struggling to understand how to set one up and execute to it.
@@garybenninger1544whats the video? Had no idea they collabed
@@nateevink7237 nighthawkinlight has an incredible video on multivariate experimental design! (basically DOE i think)
使ったことはないけどタグチメソッドの可能性を感じますね
All engineers should be schooled like this one- to experience the disappointment and frustration of mechanics that repair and have to service the equipment they design.
I forget the exact make & model but my uncle had a US brand car with a Mitsubishi motor which I was going to buy off him. Turned out that I would need to pull the motor just to change the spark plugs…lol.
@@tomchelle1That would almost certainly be a Dodge Stealth. They were basically the same car as the Mitsubishi GTO/3000GT, and had a transversally mounted V6 whose rear bank was so far into the engine bay/close to the firewall that the plugs were essentially impossible to remove without pulling the motor out of the car.
@@KaminKevCrew yep! Nailed it
Most engineers already do this stuff during school. You have to do a capstone project your senior year. Engineering students now have to cover more material in less time. School is just meant to be a foundation and you learn the rest on the job.
Actually thats how you get fired, a good engineer wouldnt have to waste so much with trial and error.. understand, design, verify....
running a laptop on a rc boat is crazy
I don't know who prop-osed it, boat i'm a fan.
😂you must be new here!? He's a Savage!
Yeah, that took a pair 😂 It wasn't smooth water either!
He wouldn't risk it if he wasn't confident.
Should have used Minnie by Voyagesystems to save his laptop 😁
This reminds me of Pinewood Derby. Someone will nail it by studying the test vehicle and getting the number right for the prop by optimizing for the test vehicle.
We did this with the launcher mechanism in pinewood derby when I was younger. There was a single dowel rod in the middle of the track the would fold down and release the cars. We figured if your car passed all the length width and weight regulations but had a nose that was high up and just barely touched the tip of the dowel, your car would start moving immediately while all the other cars would only move when the dowel was all the way retracted. Optimized this design in following years with a similar high nose design but with a channel cut out of the length of the car that allowed the car to gain almost an entire car length head start on the other cars as soon as the dowel was retracted a tiny amount.
@@tmappe I guess that's clever, but in the lamest and most shameful way possible. Did you work for VW in emissions?
@@brianwelch1579 cry more because you didn’t think of it. We designed for the track. That’s what every racing team does in real life. Real life lessons learned during cub scouts make me more clever and smarter person today.
@@tmappe perioddddd. you slayed 😂😂
@@brianwelch1579 Racing is based on finding gaps in the rules to gain an advantage. If it's not in the rules, it's not cheating, and even then, it's almost encouraged to push the rules to their absolute limit to gain an advantage. Every professional racing series is like this.
Back in the early 70's I had a 13' Boston Whaler with a 40hp Johnson. In those days you had a limited choice of aluminum props and pitches. With the never ending search for performance and speed, this company (Columbian) came out with this "variable pitch" plastic prop. They were available in different colors, I remember having some red and blue. They were also advertised to be unbreakable (the company should have contacted me for testing first). They were constructed out of Lexan with a bronze hub. The pitch would "flatten out" under acceleration to help get on plane. This helped when towing skiers or to get going with like 4 people on board. Once on plane, you could back off on the throttle to more of a cruise speed and save gas (even though gas was cheap back then). The problem was that after some time, the Lexan would fatigue from all the flexing and a blade would snap off. Of course you would have to limp back home real slow because of the vibration. After running and breaking like 3 of these lexan props, I went back to the aluminum style. There's actually a NOS red one for sale on ebay now. This concept for a boat propeller has some advantages, finding the proper material is going to be the challenge
So TPU might be fun to play with is what I'm hearing.
idea for a flexible thin nitinol version of that lexan prop! It would be super resistant to fatigue effects and flexible and able to flex in a similar way.
Imma need an option to orient the motor at 90 degrees so I can submit a cyclorotor
He already literal paddle prop adapter XD
My man is about to print 300 propellers
more than 300😂😂😂
Factorial experiments not random OFAT. 😂
yeah way more than 300
he better get sponsored by matterhackers or something 😂😂
Maybe thats why PCB way is sponsoring, 300 prints for them is nothing
you really should name this red catamaran 'Mystique', in honor of all the faces it had since it was made.
Every week or so I get this weird urge to watch an rc TestFlight video, then I go to his profile and boom new video
If you want to run numerical analysis but don't have acces to cfd and a large amount of compute, an alternate approach is using blade element momentum theory. I used a python implementation, PyBEMT to optimize a propeller for my BSc. Thesis and it worked really nicely.
Thrust (in Newtons) * speed (in m/s) is a measure of the power delivered by the prop to the boat. So if you have thrust, and speed through the water, you can calculate prop power that way. Prop efficiency is (prop power) / (shaft power). I think 50 percent efficiency means you are doing pretty well. You can estimate the motor power output using torque (as estimated by VESC) and rotational speed.
How do you get the thrust in newtons of a propellor, is this by doing prior testing and already having an understanding of the boats drag and knowing how much force is required at a given speed?
@@-B_S- Been a while since I watched this video so I don't remember what I was getting at with my comment. But, to answer your question, if you have a real propeller and boat, you can potentially measure the thrust with a load cell on the prop shaft. You can also measure the thrust by tying off the boat to a dock and measuring the tension on the line with a load cell. That may not be the same as the thrust while underway, though, so you have to be careful with that. That is called "bollard pull."
If you are designing the boat, you can make predictions about how much power it needs based on rules of thumb or detailed study using computational fluid dynamics (cfd). If you have a hull, you can tow it and measure the tension in the tow line using a load cell. The rules of thumb are based on displacement of the boat and desired speed you want to achieve.
So rctestflight could maybe just tow the hull at a few different speeds to establish a thrust vs speed curve, then for each prop he could estimate the actual thrust it is delivering based on speed. That will tell you how much thrust power has been delivered to the boat. Compare that with shaft power (torque x shaft speed) and you can get the prop efficiency. Or at least an estimate of it.
I was stunned when I saw a laptop on that boat! Amazing work mad, I love it!
I'm not interested in propellers at all but I've been watching for 20 minutes, I like your videos. Just the right level of nerdiness😂❤
That laptop on the boat was terrifying hahaha
Good stuff! But you really want to think of pitch the same way you think of thread pitch on a bolt. The pitch is the distance the propeller would advance through the fluid (with no "slip"). And this gives you the nominal blade twist automatically. You can then refine the twist with small changes to optimize the span loading. You typically want to unload the tips somewhat.
if you so smart and knowledge glutted why you don't submit your own "improved" design and let the numbers speak? Too afraid? ;)
@@guillermomarturetfendt9037 I hold 4 world records in aero. Yes - I'm to afraid.
@@Rick_Cavallarohave you read jack Norris's book on propellers? It's hard to come by and is usually the expensive
@@terryboehler5752 I do. Jack sent me a copy of it when I was designing and building a wind powered vehicle that goes directly downwind faster than the wind. I was eager to make the most efficient propeller I could, and hoped he had some good secrets. My recollection is that the book had some good concepts and explanations, but overall it was kind of a mess. We ended up designing the propeller using JavaProp.
This is awesome! Time to dust off my Ansys and do some simulations...
Would be fun to see a variable pitch prop like you see on aircraft
That would be cool, but it probably would not be printable (or atleast work well when printed)
Controllable pitch is being used to achieve higher power regeneration when sailing. This is where propellers need innovating.
I was thinking the same thing. Maybe even some kind of Auto feathering system that sets pitch automatically according to speed.
Nice video (and competition). I was wondering about the drag on the surfaces themselves, which also adds to a loss in efficiency. What would happen if you would change the surface structure of the blades and cone to very smooth instead of the rough surfaces that come out of the 3D printer? Or even better: make the smooth and very hydrophobic with an additional coating? I'm aware of the fact that this no guarantee for lower drag, because for example golf balls are purposely dented to achieve better aerodynamics. But maybe this is an aspect that you (ar a contestant) would find interesting to look into.
This is going to be a fun one to watch. For anyone else who feels the same you might also enjoy the fan showdown. Similar concept different channel. Can’t wait to see this.
This stuff has always interested me ever since i got my own boat to maximise fuel efficiency. The real key for efficiency is prop pitch a bit of the design but most importantly is the cross section profile of the blade itself like an airplane wing to generate lift but on a prop you are trying to minimise friction as most of your thrust and economy loss comes from vortices generated by the tips of the blades as heat
0:58 yeah it’s sweet using multi material. I use PLA as support for PETG and the undersides are Lush.
It is amazing to see how your projects just grow and grow. You sir, are an inspiration to us all.
you could have just shifted the voltage up slightly with a voltage divider -- just two resistors.
dunno if you trolling but voltage dividers only go down not up, to go up you need an amp
You need to revisit the diameter, after discovering how-wrong the pitch was, on your 1st batch..
This is part of the principle of looping back *far enough* in the design-cycle, for what one's project is actually doing..
( Wysocki's book on project management, Extreme & Emertxe projects loop back to "should this project continue?",
Agile projects loop back to "What is the next deficit that we remove?",
Traditional projects loop back to "what is the next item already-decided that we do?"
Research is extreme/emertxe category. )
Now I want to see also drone propellers competition as well if we have PC fans and boat propellers already :)
This is such a good idea!
Did something like this with an ultralight propeller, drew the tip circle as a vertical line, set the rpm (IC motor with 2.36 reduction giving 2650 max rpm at propeller) and worked out helical path at desired 60mph cruise. This gave the ideal tip angle, discovered reducing diameter allowed positive pitch angle of attack, as diameter increased it began to go negative. Research showed aero profile still give thrust, even if aircraft velocity and prop rpm resulted in negative, up to about 2 degrees. Your work is brilliant.
I just wanted to say i love these watercraft videos. Especially the manned full size expeditions!
Thanks!
so glad to see you now using ELRS!!!
just looked it up - wow! very interesting!
This series is going to be absolutely amazing, hopefully it can be visited a few times a year.
Hi, i dont know if it has already been taken into account, but for a better propeller i think you should take into account also the weight of the vessel, the "hydrodynamic drag" of what you are trying to push/pull, and the result will also change for the cleanness of the waterflow, the best is the pulling propeller layout, if i'm not wrong. Don't know if someone already pointed them out, hope this is useful.
Thanks, man. Better yet, Amazon should thank you. Your vids inspired me to spend a few hund-o's on RC stuff... so far.
Daniel, often times, the power efficiency in an electric set-up is dependent on how well paired a load is with a motor, and where you are in the efficiency curve of the motor.
Have you considered measuring watts as Torque * RPM, rather than V*I going to the motor, in order to exclude any efficiency variation in the motor?
This will be a good competition, hope you get some great entries.
So cool! 👍💪✌
You also should calculate the optimal hull speed for your catamaran, since it works in displacement mode. And after that test the props for that speed.
What brilliant idea for fun competition, there way smart people than me that hopefully come up with some great designs ?
I like ask few questions ?
1) all our designs limited to 3 blade like yours were ?
2) What about surface drive props that may be better at high speed ? can you change the depth they run at ?
3) Could someone make variable pitch prop ? not sure if you could design passive system at this scale
Am really Looking forward to the results video in few months love jenifer xxxx
If you circulate water in a loop you can get dynamic thrust data indoors 🙌 and I bet you could tune the loop to match drag from a boat..... although actually please keep lake testing it's more fun to watch 😉
Would recommend trying out configurations in onshape for this sort of thing. They work really well for switching between models while changing a bunch of dimensions and are much easier to set up than I expected.
"The 1930 Robertson Waterplane ground-effect vehicle powered by an outboard motor". There are pictures and videos, but I don't think I can post a pic in the comments. I saw this the other day and immediately thought this might be something you'd build or at least be interested in knowing existed, if you don't already. In other news... it's getting to be grass cutting season. I hope you are working on your next autonomous mower contraption. I'm a long time subscriber to the channel and grateful for your willingness to share your talents with the world. Keep up the great work! - Stanford
it's less fun because you'll actually find the sensitivity of each parameter pretty quickly, but taguchi's method will let you use the data you already have to explore real impact of the changes. turns out the one variable people doesn't always have a big impact. wouldn't it be nice to know if an incidental change unrelated to the control had a huge impact?
the best use of the prusa XL I seen so far
Another thing to consider for efficiency is rudder location. Offsetting it would reduce drag since the prop wash is effectively moving faster than the surrounding water. Then again it would potentially reduce it's ability to turn the boat for the same reason.
Better yet you could make the motor pylon itself rotate. Outboards and tug thrusters are obvious examples.
I took an almost identical approach trying to design a propellor for my electric outboard motor for my sailing dinghy. It's super interesting seeing how sensitive the diameter to pitch relationship is.
I was also trying to find an optimal propellor for recharging my battery when under sail.
i would HIGHLY recommend looking into multivariate experimental design! there is a video by nighthawkinlight that explains the basics. as destin also mentioned in his comment about DOE this would be a great situation to do some really cool experimental design and data analysis!
Awesome! Gives me pc fan showdown vibes, cant wait to see how the series progresses
Looking forward to see more of that radar prototype thing in the future. The small/mid scale USV market desperately needs that.
GOD YES ive been waiting for someone to do this sort of video
Varying only one variable may not be what you want to do. It may happen that the pitch and diameter are both need to be changed. Vary rarely the geometric variables are uncorrelated in fluid dynamics.
I would suggest to print multiple variants where all variables are changed. Diameter, pitch, thickness, the size of trailing edge serrations. Then you can use something like linear or logistic regression to fit the efficiency data using those variables, then you can use the fitted function to find the global minimum.
In addition, the boat hull should not distort the experiment. You use a waterdisplacement hull. A planing hull can change the water resistance extremely. This can be achieved, for example, using sliding steps or smaller areas with water contact.
Ultimately, with a wing you have more glide and very little waterdisplacement. so its hard to compare
I was a first responder in the United States Coast Guard and drove several boats.
HANDS DOWN - my 18-foot jet boat was far and away fastest.
We used it to intercept "cigarette" boats that violated President Nixon's 1 square mile off-limits area on Key Biscayne Bay.
102 mph on flat water with slight (1" or less) ripple.
Try designing an "impeller and watch your little boat FLY!
I saw a video by smarter everyday where he went to the coast guard and got to ride their new unique jet boat. It had multiple independently controlled movable jet engines which allowed them to conduct extremely unique movements in order to help with rescues and other stuff. It was the coolest most maneuverable boat I've ever seen
@@benmcreynolds8581
I'll have to watch it, thanks. My jet boat experience was 1972-ish. I'm 75 now.
@@WhatDadIsUpTo it's really cool to hear a little bit about your past experiences. I think you will enjoy the video and seeing what modern stuff they now use. I'm curious how drastically different things will be in certain areas. Mostly related to engineering capabilities
In a recent online ADC test, the lowly ATMEGA 328 on the Arduino Nano was shown to be considerably more accurate and full-ranged than the ESP32. Three cheers for Microchip IP.
You need to match the propeller to the motor uniqe torque curve and efficency curve, togeter with the drag VS speed of the boat. Without that, it is only shoots in the dark. Get all the curves and see where they match.
I feel that a healthy part of a monohulls benefit is that is disperses weeds and sea clutter which really does help with these smalle scale props and preventa them getting clogged up.. just a consideration
Props (proper respect) for an excellent video.
Hey, this reminds me of the fan showdown series. Neat!
Looking forward to seeing the next part of this contest.
Unless water behaves completely different from air, center of pressure for a symmetric profile is close to 25%.
(In air it jumps to 50% when you go supersonic)
After watching this video (and the linked conversation) I got an ad on another social media platform that showed a close up image of the market’s leading toroidal propeller. Note: It doesn’t have a tapered cone on the trailing edge. It has an open cup with small fins inside. I wonder if the open cup on the trailing edge of their propeller adds efficiency by giving the bits of cavitation on the prop surface a centralized negative pressure space to propagate them into an organized column that can allow for more efficient flow?
I haven’t spent any time using computer modeling so someone else will have to test the theory. I hope you find one that suits your needs
In your Onshape CAD, the intersections are faulty. Thanks for leaving it public. 👍
Awesome way to control the parameters in your experiment! Very impressive!!
Wow, excellent video! Maybe a good way to measure current would be with an induction amp meter, used commonly in a lot of multi-meter style tools? Excellent to see Airshaper in the video, who I highly recommend to anybody looking for efficient CFD studies. Their website is full of interesting studies, detailed in extensive PDFs. In case anyone is wondering about the Onshape propeller template, it is also meant to work with the book of HAVF airfoil vectors I published. So, the onshape model is capable of swapping airfoil very easy, from the currently used ones, to any of 1500+ other popular airfoils. That's part of the reason Daniel is able to change the trailing edge so easily. Airfoil vectors are the future, ditch those DAT coordinate files!!! Now.... to start on my submission.
A current transformer is even less accurate at low currents, a shunt is the best method here. Just need to choose a slightly higher shunt value or different amplifier for better data at low current. He could just replace the shunt resistor on the board he has with a better value, the math is literally just ohm's law.
That weird patch of water was probably some kind of thermal inversion, or there's an outflow pipe dumping storm water out there.
Sending a wish out into the ether for a prop in the 800 to 1500w range that could power a mini catamaran powered by solar. I truly think something like that paired with modular catamaran hulls would be a commercial success for various applications. Love your videos.
Oh what a difference the pitch makes. Twenty four tiny degrees
Looks like you equipped your boat with an SPS-48 radar set. Nice touch!
You're testing efficiency on a displacement hull with a specific waterline length, which will not be transferrable to other types of vessels; planing, vings or displacement hulls with other waterline lengths
If I had a girlfriend I would wake her to watch this video (it's almost midnight here when this video is uploaded)
Berlin?
Yes
@@TravlieGerman gang!
Me in Berlin rn...
Dieser Kommentar ist absofort deutsches Staatsgebiet
7:23 seems there are some cavitation behind the propeller, maybe the propeller cone could have another shape also for better efficiency.
And change out the wood stick with a thin flat piece of steel plate for less drag.
Didn't know that about petg and pla not sticking to each other as tediously. Interesting!
Hello, love your channel! Question: do the modifications you made after tests skew the data or are you measuring unrelated things?
As for you SQUARE way point mission. In aviation we use both FLY OVER and FLY BY waypoints. Fly over looks to be what you are using in your missions. This causes the craft to float over the point, then turn, then adjust course to get back on course. A fly by uses an algorithm to start the turn prior to the point and intercept the out bound course.
Would programming. The mission with a a 4 fly by waypoints reduce the curves in you measurement graphs?
I ask this as when a home built aircraft builder is figuring out its actual speeds for a a newly build plane they fly a square box similar to what you have made, but using fly by way points helps to eliminate many errors.
#1 I love the little rotating flag on the servo. It looks like a mini radar. It must signal something.
#2. I really like you have made a modular joint for all the props to align with. Next time, include a molded latch into the prop for easy connection and disconnection.
#3. I see you have a toroidal prop, what about a tri-toroidal prop? Something like a double stack toroidal.
Have you done any testing using a KAPPEL propeller design?
Subscribed to Fan Showdown for a year now, and as RC hobbies guy I am much more interested in competitions like that o) And that is where I might even try to submit something despite being a software developer o)
if you design it to wear slowly you can continue to be a software developer.. badum tsss!
how easy those supports come off is so satisfying
There’s one more key design criterion you should assess these on! It’s how dangerous the propellers designs are to marine wildlife. Propellers are a huge hazard to aquatic animals, and in Brisbane where I’m from we often find dead turtles, killed by outboard motor props. It’d be great to move towards propellers that are efficient and also safe.
because onshape is giving me a migrane, mine's gonna be exported to Blender, made, and then re-compiled as an STL.
(fun fact, the squares in Blender are exactly 1x1m, so I can make sure it's within ruleset while also using software that I can just click and drag with)
This channel rules so much. Awesome content!
Excited to see the results!
Yeah! See you at Open Sauce!
This is really cool! Looking forward to see what horrors people submit!
thanks for this great video! im so inspired, and its also VERY great for PCB way, they can produce very nice components!!
What's the rotating red and white thing? Looks like a cute little radar!
In full-scale boating, the diameter and pitch of a propeller is the equivalent of a transmission's gearing in a car. Absolutely critical
Think about a propulsin method like a 'twin screw gear' vise.
A central prop turns a cog to surrounding sides.
with 3D printing, you could potentially have a global toroidal prop.
I wonder if you could also have toroidal impellers inside a tesla valve.
I really enjoy your videos. This is a good one and a very ambitious project. I am concerned that you are doing this the hard way, though. A bit more analysis at the front end can save a lot of work on the back end. It is not necessary to use a complex CFD code to design propellers. Consider learning the two codes released by MIT professor Mark Drela. These are "QMIL" and "QPROP". They can be downloaded for free, along with "concise" instructions. QMIL provides a minimum induced loss propeller design from specified inputs, for a single design point. This code can be used to get in the ballpark in terms of diameter, solidity, RPM, and power. This initial design can be refined with QPROP for off-design conditions. QPROP can include electric motor behavior so that combined motor-prop performance can be estimated over a range of speed and input power. This is a command line code and is not for the faint of heart. In the right hands it is very powerful and can give very good results. A few hours designing can save many, many building and testing hours! I hope this is useful. Thanks again for your very interesting videos.
Try a 3 blade surface prop. So one or two blades are above water line. On your props try adding cup at the edge of the top of the blades. Make the blades more rounded.
computer fans are 120mm
but you can also do 140mm and they have better cfm. may want to check 140mm to see what efficiency you get with the gains
This is similar to fan showdown where you can design 120mm fans. Love it
Love your videos, and if I may give something. How about having a "baseline" prop that you test each day, so you can get some day to day conversation. Like if day one it does 1watt/gram or what ever and day 2 it does 1.5watt/gram you have a ratio you can normalize your other props with. Just a suggestion. Thank you for your great videos.
Solid suggestion!
props for your efforts!
This is a good way to promote technological advancements in propeller design.
Very curious to see if a generative model could take your results and inputs and provide some recommendations for your prop. Then compare against the community designs.
Is that a rotating ~60 GHz(? or something high) radar at 11:05 ?
Awesome video, but I wish the RCTestFlight channel would get back to planes some day :)
I wonder how much of a difference smoothing/polishing these would make.
What about dimpling the props surface in a way that mimics fast swimming ocean fish or mammals? Could that make a prop more efficient?
You mean the golf ball dimples? They're actually a kind of 'turbulator' in other words a device that forcibly transitions flow from laminar to turbulent boundary layer conditions. Doing this can be helpful if your design is experiencing laminar flow separation, which can be an issue if (for example) your object is fat compared to its length and the speed is low. Outside those conditions they just add extra drag :(
@@nerd1000ify Ok, interesting... Thanks for the reply!
@nerd1000ify pretty sure he's talking about tubercles. They're the opposite of a dimple, but serve the same purpose. If you want to see them in nature, check out the larger whales.
This is awesome! idk, maybe i'll give this a shot. Also, whats the rotating radar looking device you see on the boat in some shots?
It's interesting how different this process is compared to gasoline boat engines.
Looks like you need a bigger/heavier boat. If you can't push the motor to 100%, without the boat planning you can't see what the full effectiveness of a prop design for that hull.
Now if you want to make a prop to see how fast you can get on plane, that's a different design criteria.