Tesla Turbine v2.0 [3D Printed]

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
hi We Need to Talk I have something I need to tell you and I'm not sure what you'll think of it but maybe we can work this out there's no easy way of saying it I'm seeing another turbine you're giving me the silent treatment fair enough let me tell you all what happened it all started a while ago when I was shaving my my face yes that's what I was doing when suddenly I received a colleague in Texas of our build the tool binder sorry my mission was set to create a new wind original design for a 3d printable Tesla turbine an invention capable of converting pressurized fluid into electricity version 2.0 to succeed in this mission I first needed to understand how the Tesla turbine works and for that I studied the most trustworthy sources on the matter and this is what I got at the most basic level a Tesla turbine is constituted by two components the case II and the rotor the casing normally has a circular shape and the tangential Inlet to guarantee that when the pressurized fluid comes in it naturally creates a vortex that spirals into the center of the closed enclosure in the center the fluid is conventionally expelled by openings placed on the lateral part of the casing the rotor is a simple axis with several thin disks that is placed inside the casing right at the center of the vortex the air that travels in the vortex path is forced to go through the disks and because the air it's a fluid it has a certain stickiness to it what makes it stick to the walls of the disc and drag them along its circular vortex path thus making the axis rotate a few videos ago I designed and printed a very simple version of this turbine which was able to reach 25,000 rotations per minute was a pretty good result but I also wanted to test other parameters like electric power in the efficiency so I decided to design new one because in the first design I used a threaded rod as an axis and metal nuts to secure the disks which are both not really printed I decided to 3d print all the components this time except for the ball bearings because ball bearings are freaking hard to print so yeah I used regular ball bearings again for this design I had some stuff I wanted to try I wanted to exhaust the air through the center of the axis the more the exhaust is near the center the more surface area of the disc is touched by the air which means more torque I wanted to print thinner discs because thinner discs means less drag and more disk contact in the same space I wanted to incorporate an electric generator supporting to the design so I could attach a small DC motor to the x's and generate electricity I also wanted to use a flow regulator so I could gradually increase the pressure to get high speed with as little bag pressure as possible the complete model for the new turbine was formed by four main parts the casing the axis the lid and the disc the casing is a simple linear extrusion from a 2d profile that was prepared with five holes in a star formation to fix the lathe via screw connection to make the path of the air as smooth and continuous as possible I designed the carved cut into the inner surface of the casing that gradually disappeared into his curvature the leads match the profile of the casing but at the center were prepared with two ball bearings with a ten millimeter inner diameter the axis was designed to be simple to 3d print with no legs and no complex cross-sections a hollow interior and openings in the central hub allow the air to flow in natural stream lines with almost no resistance for the rotor I wanted to use all the 3d printed discs with a quarter of a millimeter in thickness and incorporated water of a millimeter spacers but because that would give me an insane amount of discs to print and some of you guys suggested that I use stuff like CDs and art drive discs in the rotor I decided to create a mix Roeder the thing is a lot of other youtubers already you see these and are drive dis in their design so I wanted to do something different to follow the theme of using digital storage units as power hubs in a tesla turbine I decided to use 3.5 inches floppy disks I had some laying around from a video I made on the art of cracking software and they fit a criteria they are thin and smooth disks with a hole in the center the initial plan was to create a rotor comprised of 3d printed disks with an integrated spacer floppy disks and floppy disk spacers the first step was to disassemble the floppy disk so I could get the disk inside the casing and the procedure is actually very simple you just need to take the floppy disk and smash that like button come on do it please I need it now in a more serious note it's time to talk about this video sponsor floppy disks are you tired of convenient in small storage units that offer you way too many gigabytes of storage room are you an iftar that loves inappropriate and absolutely technology then this is the product for you floppy disk by now for only $9.99 and get another floppy disk completely free after getting the 10 disks from the diskette I preached attendees and ten floppy spacers taking my sub device I decided to print the discs using only outline perimeters because the discs were only a quarter of a millimeter thick and formed by only two layers I took some time to level my bed after printing everything I painted the spacers with black and red markers so I could later align them with each other because beyond being spacers I also designed them to be paths that would guide the air into the openings of the central hub smart right I know don't forget to subscribe the next step I assembled all of the disks first a 3d printed disk then a floppy disk spacer and then a floppy disk creating an array in this order until I feel the entire hub to keep everything in place I printed a small red disk I glued to the structure using acrylic resin this is the final result with the rotor ready I could now print everything else to print the casing I chose my big girl Claudette with a big nozzle of 0.8 millimeters in diameter look at those lines dude with this Web girl I can print in half a millimeter layers in the end the surface quality wasn't the best but it was good enough for the job and the printing time was tiny just like your mama's hey what are you doing take your finger out of there this is a Family Channel dude anyway next came the lids first the simple one and then the one with a motor support once I had all the main parts I inserted the bearings in the lid and threaded the holes in the casing to connect the motor to the axis of the turbine I printed a customized coupler that compensates axis misalignment I also glued a six millimeter brass connector to the inlet so I could connect a flow regulator this is the final setup I was really anxious to test this version so I rushed to the compressor to see what speeds this baby could pull off after setting everything up I was ready to use my new digital tachometer to make history I slowly increased the air flow as I looked to the monitor of the tachometer and as you can see I failed the turbine was barely getting to five thousand rotations per minute I tried several times and nothing I had disgraced my family name what would myself think of me the shame was too big I had no other choice I had to terminate my YouTube account maybe try another medium like tumblr or Instagram or wait a second something is wrong here I'm pretty sure I had good ideas in that design well except for the floppy disk one that wasn't a great idea and the openings in the lids should be closer to the center also I shouldn't have printed a quarter of a millimeter in thickness disks they are not that rigid and the finish of the surface is not that great well back to the drawing board okay I needed to change some stuff but not all of it I kept the casing but completely trashed the axis and the discs the leads were not there but I needed to put the opening to the center so I printed new ones on the kree allottee to get better quality finish for better exhausting the air I redesigned the discs to still guide the air to the openings on the yellow axis but also have the conventional holes that allow the air to escape through the lids this time I printed 10 discs with point 75 millimeters in thickness and spacing so I could get a good stiffness and surface finish yes add stiffness don't be immature because I'm an efficient guy I took advantage of the openings on the discs and printed the new axis with built-in extruded guides to align the discs and also keep them from sliding looking at this footage I was also reminded that I needed to cut my nails so I also did that to complete the rotor I inserted all the DS and secure them in place without glue because a good recipe shouldn't be changed I also printed a small flywheel where I placed a reflective marker for the digital tachometer once again the turbine was ready to rumble but before the final test there was something I wanted to do to know if the rotor was synching with the speed of the air coming out of the compressors nozzle I first needed to find how fast the air was being expelled into the turbine so this was what I did I have a 100 litres tank attached to the compressor which is filled with air at 10 bars of maximum pressure as the the pressure lowers and at 6.5 bars a mechanical switch is triggered and the compressor starts filling up the tank to full capacity again to calculate the average speed of the air at the nozzle I needed to know how much air escapes between the ten to six point five bars interval and how long it would take to happen with a nozzle completely opened with these two values I could calculate the flow rate at the nozzle divide that by the nozzles cross section area and we get the linear speed of the air first I recorded a video of the barometer Boro meet present in the compressor as I discharge the tank with the nozzle completely open I used the video to time how long it took the tank to get from ten to six point five bars and that was precisely 64 seconds to get the amount of air that escaped in this time period I calculated the density of air at ten and six point five bars knowing the volume of the tank I calculated the difference in mass from one pressure to the other and converted that value into a volume of air at atmospheric pressure dividing that by the elapsed time I got the volume flow rate dividing that by the cross-section area of the nozzle I arrived at an average linear speed of 100 in 87 meters per second which is really impressive and the speed of the worldís fastest trained a record achieved by the Japanese maglev train in 2015 as impressive as this speed is for a train is not that impressive for air flow since there are many compressor nozzles that can make the air go sonic or higher we're talking about more than Mach one speeds higher than 343 m/s with 187 meters per second we are at fifty percent of what we could get with a better nozzle which tells me that the flow is being choked but because I don't have any other nozzles for now let that be a note for the future next I use the circumference of the disks present in the turbine to estimate the theoretical rotation speed if the turbine was able to sink to the speed of the Year the result was 43,000 rotations per minute my theoretical go was said it was time to test the turbine after regulating the flow to get the best speed possible I got a stable speed of 20,000 rotations per minute with a maximum speed of 24,000 rpm I know this doesn't sound very impressive since the version one was able to reach 25,000 rpms without any of these fancy design implementations but you need to understand that the version one at a metalic machine axis with much better tolerances and almost no vibration also version two as a bigger work volume which means it was designed for lower flow rates and higher torque and let me tell you this one is definitely more torque than the other one I know that because every time I ran a test I needed to stop the turbine to start again in most of the times I almost burned my fingers doing it unfortunately I don't have a device to measure torque but who knows maybe I'll build one in the future what I do have is a multimeter to measure the current and the voltage coming out of the motor I measured 12 volts with no load at maximum speed and 0.37 amp ere's of current with the internal load of the multimeter and the top speed over one 3,000 rpms once more add a lot of fun making this project and I learned stuff that will be useful for future designs for example now I know that I need to improve my nozzle so I can get the fastest air speed I can get from my compressor also I need to use non printed axes in discs and I need a better setup to measure the electrical power and mechanical power of the turbine what do you think do you agree with me tell me your thoughts in the comment section below officially this is the last video on the Tesla turbine series but of course I will try again later in the near future for now I have some project ideas I want to implement in the next video and I will share them with you on the community tab of my channel for now is everything I hope you enjoy the video until the next time [Music] [Music] you
Info
Channel: Integza
Views: 751,179
Rating: 4.9057007 out of 5
Keywords: Tesla, Nikola, Turbine, 3D, Printed, Printers, Modix, Big, 60, Creality, CR, 10, Printing, PLA, Thingiverse, Bladeless, Boundary, Layer, Fluid, Mechanics
Id: n5KmpFRSXiA
Channel Id: undefined
Length: 15min 27sec (927 seconds)
Published: Sat Feb 02 2019
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.