I 3D Printed an ELECTRIC MOTOR FAN
Вставка
- Опубліковано 2 бер 2023
- You and your company can try Onshape for Free at Onshape.pro/Integza
EDF:amzn.to/3QzoEIK
ESC:amzn.to/45jxQ8f
3D models
Simple motor:bit.ly/3ZDWCgi
Simple Fan:bit.ly/3IL2bD1
Drone Motor Fan:bit.ly/3SHxBid
EDF Motor Fan:bit.ly/3SMbdEd
Printer I used:www.prusa3d.com/product/origi...
Support me on Patreon: / integza
Paypal me:paypal.me/jgomes1992?locale.x...
Discord: / discord
Reddit: / integza
#3dprinting #fan #electronic - Наука та технологія
I'm guessing it has to do with the diameter of the intakes of the EDF and the 3d printed ones. It's the biggest difference I can see and would guess it to be very influential on both the flow rate and the static pressure delivered by the fans. But I don't know for sure, am not an engineer in this regard. Just know that I hate tomatoes too.
I believe you’re right. I am an aerospace engineering student and I believe there is some pressure distortion happening within the fan casing because it is not made precisely for the fan blade and the speed of the motor. When designing these type of casing MATLAB and Solidworks fluid flow models are used to determine the best design that matches with the casing. There are a lot of helpful guides online to discover more.
Your likely right. I think an added factor could be the weight of the fan relative to the original EDF. The printed fan may take more energy to rotate therefore leaving less energy for thrust. Might be completely wrong on this, but seems logical to me.
@@judahhawi3925 not sure if the weight of the fan could make a massive difference, as it's balanced, so the rate of acceleration in the spin is all that's affected, otherwise once it's up to speed the higher weight gives it more momentum like a flywheel. I could be wrong tho :)
I may not be an aerospace engineering student, but I'm reasonably fluent in physics. Ultimately, you are correct in your assessment. You may want to know how you are correct, and how it all correlates:
"There are a combination of reasons for this problem. All the physics that go into making a fan are torque, radius, radial balance, radial aerodynamic coeffecient, and mass of the impeller.
-Torque is a known and unchanging factor with the motor. We have no intention of changing this, so it becomes a foundational standard.
-Radius is variable. Radius referring to the radius of the impeller. This also correlates to the radius of the frame, or more specifically, how tight the tolerance is between the impeller and its ducting. Low tolerance allows back-pressure to directly 1:1 displace some of the original displacement from the impeller.
-Radial balance is a personal term I'm using to describe how well balanced the impeller is along its blades. Where is the center of mass along each blade including its respective portion of the hub? The further away from the point of rotation that the center of mass is, the longer it takes to spin up and the more of an impact aerodynamic drag, or back-pressure, will have on the impeller.
-Taking that same radial cross-section of an entire impeller blade and its respective portion of the hub, we then need to balance its aerodynamic coeffecient with at least two other factors. First is the velocity of the impeller blade at all points along its radius. Second is the range of operational flow of air into the impeller. So the aerodynamic coeffecient profile of every slice of the impeller cross-section from the hub to the tip needs to be balanced according to its intended operational velocity. If this doesn't happen, then you create unintended pockets of high/low pressure along the surface of the impeller blade. The blade does rely on a small high/low pressure imbalance along its surface to induce pressure displacement, but too much, little, or even inverting that balance will quickly ruin the efficacy of the impeller. You need to fluid-flow simulate this in CAD before production. Even just taking a Dyson impeller design isn't enough, because it was designed for a specific frame, intake, exhaust, RPM, torque, radius, and mass. All of these things work together. You're also working with a constant; the torque of the motor. Another facet to consider with radial aerodynamic coeffecient is how large your intake and exhaust openings are. The standard ducted EDF blower doesn't try to pressurize the air any more than needed to maximize its CFM potential. Meanwhile your centrifugal impeller is also trying to mechanically compress the air and is thus encountering resistance. The impeller design needs to be further augmented for that factor. From there, the CFM potential for the intake and exhaust each need to be considered and calculated based on the expected pressure displacement of the impeller and its frame. If the CFM potential of either the intake or exhaust is too low, then you get back-pressure again ruining the displacement balance you're trying to create.
-Mass of the impeller goes without saying. However there are some subtle nuances to this that most either forget or fail to recognize. Rotational mass in a frictionless environment will only slow down how quickly a specific torque will get up to maximum rotational speed. We're not working with a frictionless environment though. You're combating the balance of ambient air pressure. You are trying to use a spinning screw of blades to dig into the air and pull it in a specific direction creating a pressure displacement. Keeping the mass of the "screw" as low as possible will allow more of the motor's torque to be applied to the air instead of the impeller's mass."
I suspect that the small impeller would allow the motor to spin it at a higher RPM under the same power load, negating the intake diameter as the issue. I wonder if instead the losses can't be attributed to back pressure when the air is compressed causing the RPM to get reduced under the same power load.
In other words, either the motor gets fed more current or the impeller/casing gets scaled up with the same intake diameter.
I'm also not an engineer and I personally get bloated when eating tomatoes.
Integza make a Turbo rollerblade using its self-built engines
Agreed but make it a Rollerblade suit also congrats on the win or at least what I can see
@@marksman754 integza stop using chinese tomato bearings, and you need enough rpm to make paris hilton spin
@@patrickwatkins7572 why is this in reply to someone else’s comment? 😂
@@marksman754 🔴 What Is Islam?
🔴 Islam is not just another religion.
🔵 It is the same message preached by Moses, Jesus and Abraham.
🔴 Islam literally means ‘submission to God’ and it teaches us to have a direct relationship with God.
🔵 It reminds us that since God created us, no one should be worshipped except God alone.
🔴 It also teaches that God is nothing like a human being or like anything that we can imagine.
🌍 The concept of God is summarized in the Quran as:
📖 { “Say, He is God, the One. God, the Absolute. He does not give birth, nor was He born, and there is nothing like Him.”} (Quran 112:1-4) 📚
🔴 Becoming a Muslim is not turning your back to Jesus.
🔵 Rather it’s going back to the original teachings of Jesus and obeying him.
More .....👇
🔴 THE RETURN OF JESUS
@@1islam1 Dont care 🗿
Hi Integza! Another cool projekt :) The close view of your impeller at 4:03 shows why… well, why it sucks at sucking anything ^^
It has the wrong, almost "reverse" geometry for such a tiny inlet (same for the second model at the end of the video). Moreover, one cannot simply "open the intake wider" as suggested elsewhere in the comments. Otherwise, the volume of the shroud would become so small that the blades would no longer compress the fluid.
The solution here is to twist and orientate the blades the opposite way:
1. At first close to the the shaft (the "hub") your impeller needs an important geometrical part that is missing in your current models: the "inducer". The inducer is shaped and acts almost like a regular PC fan or propeller. Its blades are wide, with a high incidence angle (say, about 45°). Front facing without the shroud, they should look like almost those of a colored windmill toy. However in your current model, in this place your blades are shallow and oriented normally to the incoming flow, so they can barely pump anything. The inducer once there induces a high velocity to the fluid in order to suck it inside the engine (with not so much pressure on the other hand): that is to say, this missing part acts as a PUMP.
2. Then, as the blades of the inducer expand radially outward, thinner and thinner, they really become the impeller. They straighten up, and end up at their orientation at the periphery is almost radial and completely upright (i.e. the tip at the trailing edge should be normal to the shaft and to the external flow - the one outside of the engine) a bit like the blades of a water mill this time. In this second trailing part, the impeller acts this time as a COMPRESSOR i.e. better at creating pressure than accelerating the air (but it doesn't matter then, because the fluid velocity will be already high at this point, thanks to the inducer!).
So a centrifugal compressor impeller indeed acts as you explain at 4:38 except it imparts velocity to the fluid first, then compresses it, in this order :)
Too bad he didn't notice your comment, it's brilliant. I hope he sees it seeing as you spent so much time typing it up so well and explained everything great. How do we tag him so he sees it?
@@PaulK390S90V I dunno 🤷♂ Let's try @Integza
@Integza ! Let's hope this works!
@Integza
@Integza
My guess would be that it has to do with turbulence caused by the impeller, use a smoke machine (or burn some tomatoes) to see how the air forms around both the EDM fan and your impeller fan. Also, since your impeller is more structure than fan, the output thrust should be less than the EDM fan. Which is more fan than structure.
How would he ever get the burnt tomato smell out of his workshop? He could use the fan! :)
I'm no engineer, but from just looking at how significant the difference is between intake diameters (EDF vs 3D print), my guess would be that you're restricting your flow rate on the 3D print. Might be able to compensate with higher RPM, but I bet that would more than likely exceed material capabilities; probably easier just to scale size, and maybe even reduce RPM?
My thoughts exactly.
That’s what I thought also, a possible diagnostic for that would be to try running it under water since fluids don’t compress, if there’s a ton of cavitation I *think* that would point to the intake being insufficient
But like yourself I’m far from an expert
Also the shape of the edf intake being curved helps with air intake as fluids tend to stick to curved surfaces. This allows the edf to pull in a much larger volume of air around it compared to the 3d printed fan
Also the housing on the one is flared outward to allow for more airflow in. The 3d printed one is tapered to almost a cone shape. Try wet sanding the 3d printed model. If the surface isnt smooth like a tomato it will cause drag and slow down the air being forced out. Think of it like your plumbing lines. If your water lines are bumpy the flow of water will slow down dramatically. If the pipes are smooth the liquid flows at a higher rate.
A higher rpm means more drag (friction is velocity^2) , lost as heat in the air, and less energy going to push air.
Larger propellers are inherently more efficient because they can move more air slower and get same thrust as a smaller prop.
And air flows through a large hole easier than a small hole.
It is all about efficiency and that means keeping the velocity low, which means a larger intake/exit area and impeller.
What i can recall from fluid mechanics classes is that by compresing a fluid you decrease its speed , so the centrifugal fan that you designed basically outputs a slightly compressed air but at a lower speed, in jet engines this is desired because you need a high density of air with a low moving speed and by igniting fuel you generate thrust .
Yes and no, the centrifugal compressor is easy to build (tolerances can be looser) and good at increasing the pressure of the fluid (air), it's not that good at creating mass flow. Which is the reason why modern jet engines use axial compressors, as they can flow high volume/mass and compress the air enough for combustion.
After the compressor you add fuel and ignite the mixture to increase the pressure again and finally you expand the high pressure fluid through a nozzle to increase the speed of the fluid and create thrust.
@@JainZar1 Bah! I forgot about mass flow. You're totally right. Thats probably the main factor. Even if you recover the pressure, you're wasting energy pressurizing gas instead of pushing mass.
Yeah I was thinking that turbofans only work if you are actually injecting fuel. Without fuel, you are just building a more complicated fan with more steps and more inefficiency.
So centrifugal is for pressure applications and normal fan for high airflow applications
@@MeesDeppe_Official In general, yes. You can get some pretty insane pressures with multi-stage axial compressors, like in modern jet engines, but you need really tight tolerances to not have spillage back to the low pressure side.
To my knowledge an EDF is biased to pushing a large mass of air at a high speed but at low pressure (this is comparable to horsepower). Whereas a ducted fan is biased to creating less airflow but much more pressure/higher PSI etc (this is comparable to torque). It's irritating when people make a motor jet engine on other youtube channels using an EDF and a combustion chamber, as they should be using a ducted fan instead or even better an electrically driven turbo charger impeller wheel (as these are compressors rather than just fans). Thank you for your videos. I feel they are actually getting even better.
I produced a similar shape of rotor to yours a while ago. It was part of a small 3D printed supercharger. The rotor is actually hybrid being half way between an axial and centrifugal compressor stage. I've since moved back to a 3 stage axial compressor which is a bit more predictable. In theory each axial stage should offer a 1.2 pressure rise while a centrifugal compressor will offer something around 3.0. You exist somewhere between these values. The tip clearance for the hybrid rotor may affect the performance.
I miss the mustache! F*** tomatoes.
I believe that the thrust difference is down to mass flow, the EDF fan is optimized for flow rate while the Dyson impeller is optimized for pressure. It also seems to have much less blade area requiring higher RPM for the same mass flow.
A higher pressure will provide a higher thrust provided the same nozzle area but will also require more power to drive the fan and may or may not be realized depending on the rpm and the throat and nozzle shape.
exactly.
What Eru says... ^
That’s what I wanted to say! Static pressure vs. flow rate
I would note that the Dyson isn't optimized for pressure, it would be more of a true centrifugal design if it was. As a mixed flow device it is optimized for a design point that is a balance of moderate pressure and moderate flow rate.
@@videoviewer2008 Of course, I should have said "more optimized for pressure" of "optimized for pressure in comparison". Depending on how far you go down the route of optimizing for pressure, it isn't even a fan anymore.
Is it possible that your fan is stalling? The idea there is that the static pressure behind the impeller rises to a point of being higher than what the impeller can sustain. The way out there would be to try a different outlet/nozzle geometry, maybe with a larger aperture. Or to change the impeller geometry to maximize static pressure. That may be a good topic for a future video btw 😅
At any rate, search terms to learn more: fan stalling, impedance mismatch.
I was thinking the same thing, essentionally causing the air pressure to overcome the airflow and move "backwards".
The quality of the print can also make a huge difference in pressure. If the walls aren't butter smooth , you're creating unnecessary drag and losing a lot of air speed .
Very nice and interesting remark! There is a difference in fans between a propeller, turbine, compressor etc. Different air flows for different purposes.
@@asakashigure yes, that's a good point. Added resistance from boundary layer vorticity will increase impedance. This is also true for the max static pressure the impeller can generate in the first place.
Position, attack angle, slope, distance between the impeller vanes also influence this and need to be fine tuned to the pressure and volume flow expected. All this needs to match up such that the system is close to or at it's efficiency maximum at the desired rpm.
That also means the design shown may produce better results at lower rpm.
One thing you could do integza to not only demonstrate your hatred of tomatoes while also using your love of creating your own stuff with a 3d printer. Use tomatoes as a crash test dummy… create a high thrust engine and mount it on something that has low friction bearings. The tomato being the cushion so as not to damage anything around it. 👍🏻 you can even put this on trial and use numerous different engines to see which of your creations do more damage to the tomato
I didn’t go thru any of the comments, but I feel sure there is a lot of good advice. But I can tell you this for sure. First fan centrifugal fan blade open air in and out. Now using the same motor is not going to work the same with what your next design became. You basically turned into a centrifugal pump. Also the reason with the difference in flow with just the new design is this. You had two different impellers on the new design, an open impeller and a closed impeller. Meaning closed impeller not because of the outer housing it sits in. A lot of consideration goes into setting what is called your impeller clearance. Temp, metal, cold, heat etc. Closed impeller set clearance from the housing and open from what in front casing etc. This is important, a jet engine may be centrifugal but it’s what is called a turbine. Totally different dynamics than a fan or pump. Hope this helps.
That's a really neat and simple way of making a BLDC motor! I'll have to try it mmm
Wow, Maker's Muse comment with just 4 likes! Rare!
26 now
200 now
but dont forget about iron core if it should be good
Although I am sure Angus doesn’t want your printer. He lacks space and time not 3D printers.
A centrifugal fan is more useful for static pressure into a chamber especially at those smaller sizes whereas the edf is more for linear air flow. The centrifugal fan will work better at higher rpm (dysons spin at a crazy speeds ) to create high static pressures but not huge flow
thats what i was about to say, its like comparing static pressure pc fans with high cfm pc fans
which is why they're used in combustion engines to cram as much oxygen in as possible. I work more with pumps than fans, but the principles are the same (ish). pure axial flow with always give you more flow (thrust) while radial will give pressure, and mixed, well mixed results... in jet engines the thrust comes from the combustion gases, not the compressor - in a turbo fan you have the jet engine turning an axial flow fan to generate thrust (as well as with the exhaust gases) (yes, over simplification, i know).
this is where my head was at as well
this is 100% what i wanted to say, its why centrifugal fans are often used for 3d printers since they can reliably push air to cool parts, whereas axial fans struggle much more with this.
Air compressor is for stuck in air molecules together while a fan is for pushing air
molecules
ᴄᴏɴɢʀᴀᴛᴜʟᴀᴛɪᴏɴꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴛʜᴇ ʟᴜᴄᴋʏ ᴡɪɴɴᴇʀ ꜰᴏʀ the 🤶 ɢɪᴠᴇᴀᴡᴀʏ ᴘᴀᴄᴋᴀɢᴇ ʜɪᴛ ᴜᴘ✉️ ᴛᴏ ᴄʟᴀɪᴍ ʏᴏᴜʀ ᴘʀɪᴢᴇ 🎁
A) Your pump style is adding a compression stage which is a energy loss (and the entire structure is adding extra mass so now conservation of energy is involved).
B) A wing style fan works exactly like a wing, which would be playing around with pressure differentials basicly (very efficient).
C) What exactly are you trying to do here because there's a lot of engineering and considerations that goes into aerodynamics. What media are you trying to move because each media has its own most efficient blade profile and even the temperature can interfere. Fan blades for like a house fan use a general profile, while the blades manufactured for a jet would have much finer tolerances on their profile callouts on the blueprints. Surface finish callouts would even be a factor.
Idea : If you want to make a more efficient fan, make one where blade profile can change. Like maybe have an inner metal structure with a rubber bladder over it that's inflatable. Play around with the thickness of the rubber bladder to help control the profile. (Add some stiffer rubber patches or fiberglass or something so it doesn't just turn into a big balloon.) Then have all sorts of sensors that measure the atmospheric conditions with a computer to do all complicated calculations to determine the most efficient profile. Most of logic motors have speeds their most efficient at as well so thet can also be part of the calculations.
ᴄᴏɴɢʀᴀᴛꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴏᴜʀ ꜱʜᴏʀᴛʟɪꜱᴛᴇᴅ ᴡɪɴɴᴇʀꜱ ꜰᴏʀ ᴛʜᴇ ᴘᴀᴄᴋᴀɢᴇ ᴄᴏɴᴛᴀᴄᴛ ᴛʜᴇ ᴛᴇʟᴇɢʀᴀᴍ ᴀʙᴏᴠᴇ
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^🎁
Some thoughts:
- Keep in mind for your deign: Modern jet engines try do move the air as slow as possible through the engine, but a lot of them. The actual fan in an jet engine has low rpm when compared to the other rotating parts in it.
- Match the overall inlet area and the ratio of free area to closed area for comparison. (Comparing apples with bananas gives reasonable results only until a certain point)
- Change the angle of attack of the leading edge of our compressor blades (There is an impedance mismatch at the very first edge - You chop of the air at the inlet with you compressor blades and rely on centrifugal forces. The speed difference is too high. Better "cut" into the air at the first edge.) The higher the velocity difference the "flatter" the first edge should be. Have a look how the slowly rotating blades of a jet engine are shaped.
- change the angle of attack of you compressor blades with radius. At the axle it should look more like an I or L, at the outer radius more like an S. This depends on the type of "fan" you try to build. Radial or axial accelleration.
- use an stator with thinner blades (impedance mismatch)
- the stator should look more or less like an inversed version of your compressor. The air flow directly after the compressor is spiraling, this reduces thrust, because the side-wise portion does not contribute much to the trust.
- reduce the distance between your rotating parts and the inner side of the hull. For example: At the red lines in your design, there is air "leaking" over from one section to another section due to the rotation of the compressor. At the back side of the blade, vortices occur which break air flow and create friction.
Since the compressor is 3d printed, and he's relying on centrifugal force anyway, he could probably connect the duct to the blades and have them co-rotate
True there are multiple planetary step downs.
Sounds like someone is indeed well read with aerodynamics.
What he said
yeah, one look and the shape felt wrong, he had a dome shaped cowling that closely followed the impeller.
the reason he wasn't getting much air flow was because the intake was too small vs the rest of it, also, the low air flow was simply rotating within the cowling and not going anywhere.
putting a cover attached to the rotor from the intake to the impeller exhaust would massively reduce his losses, leaving air only one path to go, outward, radially, this can be captured and redirected with an outer cowling that redirects the airflow back from radially to axially.
with stator blades to also redirect the spiralling airflow back to straight
This Friday just got better!
Fr tho
Fr
Yes my friend
Agreed
One of the reasons it might not be working is because the type of motor you have is for speed and not torque. There are some motors that use the same type of system and are more powerful just what is needed for the project.
These are used in larger edf’s such as in x-fly motors which I used in a version of your afterburner design.
Zepplins or Blimps using impellers or jets for directional force would be an awesome video!
With all of these balloons and such floating over us recently I'd love to see how you can make things nerdy and fun!
The look of that squirrel cage has me wondering if the pitch of those blades isn't far too shallow.
ᴄᴏɴɢʀᴀᴛꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴏᴜʀ ꜱʜᴏʀᴛʟɪꜱᴛᴇᴅ ᴡɪɴɴᴇʀꜱ ꜰᴏʀ ᴛʜᴇ ᴘᴀᴄᴋᴀɢᴇ ᴄᴏɴᴛᴀᴄᴛ ᴛʜᴇ ᴛᴇʟᴇɢʀᴀᴍ ᴀʙᴏᴠᴇ
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^🎁
Finally, something that it's perfect for the the theory I'm learning in flight lessons!
Fair warning, part of this is going to seem counterintuitive, because to move the most air possible with a propeller we actually reduce the angle of attack of the propeller so we can increase the speed of the propeller. We want the propeller to move as much air as possible in the least amount of time, cause the end of that runway is approaching pretty quick during takeoff.
We can either move a decent amount of air per rotation but rotate much slower (high propeller angle of attack), or move less air per rotation but rotate much faster (low propeller angle of attack). That last one gives us the most air moved per unit time, which is what we want during takeoff because we need to move as much air as possible as quickly as possible.
There are actually two types of propellers used in aviation, fixed pitch and variable pitch. Fixed pitch propellers are not able to have their Angle of Attack changed during flight, so we use a design that is a compromise between moving as much air as possible during takeoff and efficiently moving just enough air during cruise. Variable pitch propellers are able to change their Angle of Attack during flight, so we can get the best of both worlds. To get the best performance out of a propeller during takeoff, we have to move as much air as possible with the propeller. To do this we actually use a pretty shallow angle of attack, but as high an rpm as we can get out of the engine. We move less air per rotation of the propeller, but rotate much faster to more than make up the difference.
The exact numbers vary based on the plane engine and propeller, but these numbers are taken from the Pilots Operation Handbook for a Bonanza A36. Lowest officially recommended cruise setting is 2100 rpm during cruise vs 2700 rpm during takeoff. So a 28% increase in RPM during takeoff. Even with a small reduction in air moved per rotation, the increase in rotations per minute more than makes up for it.
By the way, I'd love it if you made some type of plane that might be used to test the idea of a gas turbine electrical generator making power for electric motors powering fans or propellers of some variety. I think that's probably where aviation will end up. Right now batteries just have too many weight problems and lack of range problems for practical, widespread use in aviation. Plus there are a lot of benefits from being able to reduce the overall weight of a plane during flight as well as changing the center of gravity while in flight. Changing the center of gravity alone can drastically increase the efficiency of aircraft because it lets us reduce induced drag from the elevators. Can't really change the center of gravity or total weight in flight with batteries, but we can with liquid fuels.
PPL trainee here. Very good explanation!
Also very good point about battery powered aircraft being heavier. Avgas is 720g/l, whereas a battery pack for, say, a Tesla Model 3, weighs around 500kg. And at about 250 watt-hour per kg (870kJ) for the batteries vs 1,240 wH per litre for avgas... You can see the clear winner for energy density here. And considering how massive weight is as a factor in aviation, this would be a dealbreaker when it comes to electric planes unless they find a better means of energy storage.
(1240 wh/l is roughly equal to 1660 wh/kg)
@@iamtheoneandonly_ it's just a guess, but I'd expect the issue being in the efficiencies - the generators, converters and motors have some losses, however small that may be. Also there's the weight of the infrastructure which would have to be present on the aircraft. I highly doubt it'd be more efficient than using the gas straight for propulsion.
Perhaps a bit of an equivalent might be to check how is the efficiency with those 'hybrid' cars with small batteries and internal gas generator to provide power when the battery drains.
Also, good luck with your PPL!
Haha noise goes BRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR
Measuring power consumption should give you a better idea. If the EDF consumes more than your fan, it has to do with your impeller pitch either not being steep enough, if you make it steeper and it still doesn't increase velocity you need to make the fan bigger to increase mass flow.
Exactly my thoughts. The limit seems to be the maximum rpm of the motor right now, but te load is much smaller. So the motor has less physical resistance to push against, which is just wasted potential. It's comparable to driving in 1st gear with your car: the maximum speed won't be limited by your engine power, but by your rev limiter, while the entire motor will produce a lot less power than if you were using the correct gear.
My idea: print a bigger impellar
@@lucamagni5098 Fans/Pumps produce pressure and flow, and in doing that they absorb work. You should try measuring the current (ideally the power but as esc tend to produce constant voltage so current is a good approximation) used by both models of fans, and compare.
1. If the models use different power/current, you need to scale/modify your design until they do, and then you'll probably have to look at (2) below.
1. If both models are using the same power/current but producing different flow then you have an impedance matching problem, or an efficiency problem (wasted energy). The impedance matching problem is ether the inlet or outlet of your Centrifugal fan doesn't ideally suit the pressure that it is operating at.
Rule of thumb is that Centrifugal fans/pump produce higher pressures at low volumes than axial/ducted fans. I think you have an impedance matching problem.Your Centrifugal fan is producing higher pressure but at lower flow air than the axial fan, thus when you measure wind speed (flow rate) you see it as lower.
This is why rockets have nozzles, high pressure gas at low flows isn't useful for propulsion, nozzles convert high pressure/low flow to lower pressure / higher flows.
The fact you called it an aerospike turbo fan, I wonder if you already know this because an aerospike is a form of open nozzle, which is a impedance matching device.
Integza
I have an idea!
You can try to make jet pack by making two jet engine and build a small jet pack
It will be remote controlled jet pack
ᴄᴏɴɢʀᴀᴛꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴏᴜʀ ꜱʜᴏʀᴛʟɪꜱᴛᴇᴅ ᴡɪɴɴᴇʀꜱ ꜰᴏʀ ᴛʜᴇ ᴘᴀᴄᴋᴀɢᴇ ᴄᴏɴᴛᴀᴄᴛ ᴛʜᴇ ᴛᴇʟᴇɢʀᴀᴍ ᴀʙᴏᴠᴇ
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^🎁
Hey
One very simple way to "tune" propellers, turbines pumps etc. is to measure the flow and preasure (like you would measure Amps and Volts on an electrical system).
Basically measure the ammount of air moved. A very crude way of doing so would be to measure the static preasure of the built fan by encasing the output in a sealed container and measure the preasure when its running. This will let you calculate the ratio of the needed cross sectional area (not absolute area but rather the ratio between input und output).
For example if the centrifugal fan hits a peak preasure of 1,3 athmosphere youd need roughly 77% of the input area on the output end to not hinder airflow too much.
This is why for example airliner engines narrow down in the compressor stage and expande in the burn chaimbers.
Measuring the flow is pretty straight forward like you already did with the Anemometer.
if you calculate the surface area of your Anemometer (in sqaure meters since this simplifys further calculationes) and multiply it with the air speed in meters per second (since the output is basically at athmospheric preasure) you get the rough volume of air moved per second or rather the air flow.
Since (if for example the centrifugal fan from the preveous example is used) the air on the outputside doesnt like staying at that 1,3 times athmospheric preasure at all for no good reason it accelerates rapidly and thats basically a turbofan or even rocket engine (bellhousing are basically just expansion chaimbers to convert preasure to speed) but if the engine chokes because of to little input area or backlogs due to too little output area the throughput is greatly hindered.
So basically you want to have the eqivalent area for the input flow as the output flow scaling by the speed the air is moved through the system.
Of course there is also friction and inertia turbolence and so on at play but thats too much for a simple "lets see what happens if i print this"
But at the same time this is also the reason why these cheap EDFs are so damn efficient : no struggle with preasure ratios very little turbolances due to the duct and very few surfaces hindering airflow.
Make a drone with similar propellers that you used in this video.
I think he would make a very original and interesting one :)
That would be insaneeeee
The Dyson edf uses a friction fit with the duct it sits in. The body of the duct is smooth and I believe the materials used for the duct and fan are self lubricating. This allows for compression to build in the veins instead of it slipping in the turbulent zone causing drag.
Don't be fooled by the imitator, that's a fake Integza. Those "telegram" labeled accounts are a common fraud pattern.
When all else fails, more lube.
wow, I actually was right. My guess was excessive clearance between fan and duct.
That makes perfect sense! Although approaching this level of seal might be hard to achieve with 3D printers.
Maybe Integza could use a slightly flexible material on the edges of the impeller? Perhaps even a bit of electrical tape taped to both sides of the blade with about 0.5 a mm overlap with the duct and some lube in the form of graphite powder would work?
nice explanation ! and nice exemple !
Hi Integza, first comment ever if I remember well, but as it is my job’s field, I thought I could answer your question !
The problem is that fundamentally, a compressor doesn’t produce optimal thrust. To achieve the best result, you have two choices :
- accelerate the flow but low mass flow (often military)
- high mass flow but low outlet speed (civilian turbofans)
The outlet pressure is often considered more or less equal to the ambient pressure, as a difference between them causes huge losses.
You then design the turbine and the to accelerate the flow, converting the potential energy given by the chamber to kinetic energy.
The centrifugal fan (more often named centrifugal « compressor » has the benefit of having a higher pressure ratio than an axial compressor compared in size, and giving the ability to case a compressor where axial blades would be too short (helicopter engines as example). It does increase pressure, but not really speed.
Therefore, because your losses increase due to the pressure difference, you don’t retrieve the same thrust.
And with that, I’m not even mentioning that building a fan requires what we call « velocity triangle », giving you a relation between angular speed of the fan and the inlet and outlet angles.
The magic behind the Dyson fans is that they are amazingly optimised. It’s mesmerising to see the shark teeth aspect of the trailing edge of the Dyson fan in your video. On some aspects, they are even further in technology than jet engines.
Hope it helps !
Video Idea: cry for help while someone throws tomatoes at you
I know you've done it before but since you have access to stronger material now I'd love to see you try your hand at a tesla turbine again. And as always, great and very educational video!
A couple things I noticed that might make up the difference in the thrust could be a weight difference between the fan vs the impeller, the blade size difference, and the restriction of airflow into the impeller
You definitely dont have enough air flow through your fan, if you have a bigger intake and expansion chamber then you will make more thrust. Video idea, build a mini dragster with a tiny rocket engine😂
I was just recently working on an impeller design for a pressure based system and not flow. I also have a background working with a million different motors, including designing my own BLDC ESCs.
If you're using the same motor, make sure the controller you use is the same as well. Long story short, different timing and/or control will result in different outcomes.
Measure *everything* I'm sure for the sake of brevity, you've left out a lot of this. But, if you're going for the same type of performance, make sure your impellers, nozzles, intakes, chambers, etc. All have the same geometry. You want to get the curve right, the thickness right, the angles right, the sizes right, just about everything. Reduce the amount of leniency for "good enough"
Shoot for the smoothest and most consistent surface finish you can get on your printed parts.
You might gain some mileage by combining your best printed parts with the stock parts as well. A printed impeller with the Dyson chamber, for example, and vice versa. You then get a comparison of what is and isnt working from your design and you can isolate where the problem truly lies.
Hi Integza! I am an aerospace engineering student at the University of Illinois U-C. I am working on an electric thruster project through the AIAA that has many similarities. There are many potential points of optimization, but the biggest problem has to do with Bernoulli's principal and subsequent equation relating pressure and velocity of flow to cross-sectional area. In other words, the impeller produces high pressure flow, and the exhaust nozzle geometry converts that high pressure flow into higher velocity flow. By decreasing the area of the nozzle, you can further increase exit velocity by converting the high pressure air from the impeller to lower pressure and higher velocity. There will be inherent limits with the power specifications of the motor and impellers (among other complex inefficiencies); so, I'd say the easiest means of optimizing your thruster would be producing an impeller you like and just playing around with different nozzle exhaust areas. Thanks for the great content!
Keep in mind an impeller and a normal fan moved different amounts of air.
Also, with the impeller, you kinda are changing the direction of the air a lot, that can slow down the air.
Dyson uses one because they want to do their bladeless fan thing, the fan itself does not have to be that efficient with power since the 'air multiplier' method already moves air much more efficiently than by using a normal bladed fan. Its main purpose was really to smoothen air flow from the fan.
Impellers are mainly good for increasing pressure, like in a turbocharger, increasing the amount of air in a fixed volume.
A turbine jet engine does not really use an impeller anyway, it's a more typical fan and the compression is advantageous in that scenario because you want to cause combustion as well.
It is mainly that process that propels a normal jet engine, modern airliners use high bypass turbofan engines, they are more efficient because the large fan itself provides propulsion, the air coming from it goes around the compressor area so it's more like an EDF or PEM but instead of a motor, you use a turbine engine to spin it. The impellers in jet engines are really for compression to promote combustion.
Essentially, you're moving the same amount of air, just at a much lower velocity, therefore the thrust is lower.
Also the tolerances and stuff, air could be bouncing back out the front.
However, if you turn the impeller sideways and add another one on the other side, you might get better results.
We did that one time to give an rc plane more thrust, there wasn't much space for an EDF with the size of the craft unless we wanted to mount it on the wings so instead, we swapped the fuselage for one that had an impeller on both sides like a double sided leaf blower, the EDF was still in the back pulling air from below and out the back but now we had another thing providing thrust right above the EDF exhaust.
Also made the plane more stable since the cg was further back. The guy who made it didn't really know what he was doing and initially put all the electronics in the front. But the centrifugal module had the electronics mount in the middle.
Took forever to print that on the school's cheap 3d printers lol.
@-integza1 lol
Here's the 3 things that I reckon need to be optimised to get a similar thrust, likely will always make a little less thrust though because of the extra energy losses.
- Impeller blade needs to be optimised for the motor, so the blade angle needs to match the motors optimal torque/speed which would be low toque and high speed. So an impeller with very shallow blade angles.
- Making sure that the outlet cone's reduction in diameter is designed so that the all the extra pressure that the centrifugal fan creates is converted back to velocity (thrust). Pressure and velocity (thrust) are interchangeable. So you need to know how much static pressure the centrifugal fan creates to then design the outlet cone. Any extra pressure in the exhaust above atmospheric is wasted energy and any pressure less than atmospheric creates an unwanted suction effect.
- Gap between impeller blade and case needs to be real tight, high precision here.
Changing the air flow directions in the engine and adding pressure, then reclaiming it through the outlet cone to thrust will sap a fair bit of energy out of the motor which is why the thrust won't be the same as the ducted fan which is the most efficent fan for creating thrust.
Hope this helps, liking the videos!
This is the reply I wanted to pile on to. You can probably run a much more aggressive pitch than the dyson due to wide tolerances around the case area.
Always great stuff 👍 👏 on your channel
Your channel is awesome I just signed up as a patron I'm 42 year old auto technician but I lost my legs and took 3 years to heal and 2 years to get up outta bed and I tried by starting to watch UA-cam and this awesome stuff I've seen gave me motivation again maybe there still something I can do anyway thx being so cool
The way that you explain these not so easy to grasp concepts from simple steps to the whole idea is amazing.
I think thrust is made up of mass and speed and a edf make a lot of air go fast but a centrifugal fan doesnt move a lot of air at the same size but it can push it harder so you can decrease the nozzle size to increase the speed so you lower mass but higher speed but that is less efficient but need for jet engine because combustion chambers make a lot if back pressue while mixing fuel
YES!!!! THATS RIGHT!!!
@@anthonyb5279 thanks bro
You know whats better than a awesome fan? 4 awesome fans! You should really build a drone with Jet engines🔥
i mean if u wanna destroy a drone and four engines you might as well cover it in kerosene and throw your wallet on it too
I've got to say. I've been enjoying the ASMR and watching to process of building it.
ᴄᴏɴɢʀᴀᴛᴜʟᴀᴛɪᴏɴꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴛʜᴇ ʟᴜᴄᴋʏ ᴡɪɴɴᴇʀ ꜰᴏʀ the 🤶 ɢɪᴠᴇᴀᴡᴀʏ ᴘᴀᴄᴋᴀɢᴇ ʜɪᴛ ᴜᴘ✉️ ᴛᴏ ᴄʟᴀɪᴍ ʏᴏᴜʀ ᴘʀɪᴢᴇ 🎁
It would be super cool if you built a Tesla coil and used electrical discharges as a propulsion
The problem is "impedance matching" - the energy required to drive the ducted fan vs the impeller is different. You need to design your impeller to match the output characteristics of the motor - I can't help much there, but I do know there are significant changes in aerodynamic properties when you switch between axial and centrifugal flow, and it looks like you are not extracting enough power from the motor (in your impeller) so it reaches speed and doesn't do much
from.... the motor?
@garetclaborn364 yes - the motor is producing rotational energy (and doing it continuously, so "power") and the fan is consuming that
Really enjoying these videos,
Video idea what about the overall process of creating an idea(with a probelm/solution) and walk through the steps in an efficient way for beginners and pros alike.!
I think far to often people get brilliant ideas but struggle bringing them to fruition when some of the starting steps feel daunting.
But I also know how incredible it is to find a problem, design a solution, print it, trouble shoot and ultimately fix said problem and how good this can feel!
Best Alex.
Hey Integza, this is a straight thermodynamics problem. We know that Kinetic energy= 1/2 M V^2. Because of the velocity squared, it means that increasing just a bit of velocity takes a lot of kinetic energy. The most efficient way to make thrust is to accelerate a huge mass of air to just above the speed of the aircraft.
This is what the EDF is doing (high mass flow rate). What you are doing is accelerating a tiny bit of air to high speed. The only advantage with doing this is that you can make an aircraft with a higher top speed compared to an EDF.
Well it could be that the nozzle could be too small which could be bottlenecking the fan and since it’s not an explosive amount of force like a jet engine it doesn’t end up making the fan explode resulting a small amount of air leaving the fan but that’s about all I can think of
Back pressure on the fan. I agree
it would just over speed if that were the case
@@anthonyb5279 I know this is a fan that we’re talking about but it’s easier to explain it if we pretend it’s a jet engine yes you’d be right the smaller The nozzle the more thrust is produced but if there isn’t a sufficient amount of energy pushing all the hot gases through the nozzle then it causes bottlenecking and in which case instead of having A increase in thrust we have an increase in pressure in the combustion chamber that will result in an explosion this is what’s happening to the fan But since compression is not in the mix when you’re talking about a fan it relies on the rotation of the propeller and I highly doubt that small engine would be able to reach high enough RPMs to create that extreme amount of force to increase the airflow so instead The fans Air pressure raises to certain extent which is not enough pressure to make the fan explode so even though the engine may have enough power to create a lot of thrust it’s possible that the 3-D printed parts it’s holding it back specifically the nozzle because it’s too small and the small nozzle is causing restriction stopping the air from going out the back into the fan and as we discussed the raise in pressure is causing all the air to go out the inlet of The Fan instead of the outlet
my guess is that the surface area of the blades on the impeller is much smaller than on the EDF, so the impeller moves less air. a good way to test this would be to measure amp draw for the two designs. you could also run them at full speed for a few mins and then check the temperatures of stuff.
^ this. Measuring the current flow and voltage gives you input power, and then if you measure both thrust and exhaust velocity you can measure the jet power.
Divide one power number by the other and you have efficiency, which feels useful to know.
You can also measure both exhaust velocity and thrust for both fans (the EDF and your centrifugal jet) to get the power of both. I would expect your centrifugal fan would have a higher exhaust velocity and much lower mass flow at any given motor power level, which will explain where all the thrust is going. I mentioned in another comment already but for anyone who hasn’t seen that:
Power = (exhaust velocity ^2) * mass flow
Thrust = exhaust velocity * mass flow
So trading away velocity to add mass flow increases thrust, even when power stays the same.
@@ChrisDRimmerVery nice explanation
The reson for having high velocity is being able to still produce thust while movibg quickly, but comes at the expense of power
Hey Integza :) Here are some ideas:
First of all, if you have something that generates smoke, use it to check the airflow in your Fan. Your connection of the two pieces sits right at the point of maximum compression. You could lose alot of pressure that way without knowing it. An O-Ring might help there. It would be in general intersting to print it again in a clear resin to then put smoke through the design to check the flow. maybe you have high boundry losses :D
the next obvious idea would be to check your intake. Maybe it is too small and you choke the fan (which would also increase heating). Try diffrent sizes or claculate the size you need if you are sure of the values you need/aim for.
Next, you need very thight tolerances between the impeller and the housing *at operating temprature* otherwise, you again leave alot of pressure on the table. Check your operating temps so you know what thermal expansion you need to take into account.
Good luck with the fan :)
I recommend using a burning tomato as a smoke source!
Hey make a 3d printed turbocharger
@@happyjack3497a stand-alone turbocharger is just a turbojet engine
First off this video is fascinating, just casually making a brushless motor, no biggie.
I think this fan under performed because of 2 reasons. The fan you build is meant to be used in an enclosed system, the Dyson vacuum is not open to the ambient air it has the advantage of creating a pressure difference on either side. Which doesn’t necessarily translate to air speed.
The EDF really has an advantage with measuring just air speed out of it. I’m assuming it is just able to create a much tighter column of high speed air, that is really easy to measure. Where as the Aerospike isn’t able to do that.
A thrust measurement might be a more comparable measure of their performance.
This video was awesome and gave me a good think about fans, jet engines, and impellers.
You should make a jet engine with a turbo charger from a car as a side quest for this project!
Edit: realizing that you’ve done similar things with the compressor side of the turbo charger, this idea is to also use the exhaust assembly as well to spin the compressor after a combustion stage
From what others are saying about static pressure vs flow, this might be a good way to show how static pressure can be used with combustion to generate lots of thrust since turbo chargers use similar fans/compressors to the Dyson impeller you’re using
A turbo jet by definition does exactly what you want but better. Remember that air flow in a straight path moves faster and can transfer a sufficient amount of energy without obstructing further airflow (important). The turbo gets spun by the exhaust which creates suction.
Cars don't generate thrust the same way as jets. Jets use their exhaust to create thrust cars have the thrust generated and dispose of waste through the exhaust. So jets need that speed of the exhaust while cars don't.
@@phasepanther4423yep agreed! But at the same time you see these DIY
Turbo Jets with turbo chargers that generate some sort of thrust (definitely not as efficient as a straight line jet engine)
Anyways, that’s why this would be more of a side quest to understand a bit more how these kinds of compressors/impellers/fans work and maybe some cool conclusion can be made from it.
@@Pscribbled
I see.
An informational video like this would help.
As there are key differences in purpose of certain motors and of certain fans.
For example, a motor designed for higher rpm won't be able to handle too much torque (within a power consumption range). And generating pressure does require some force, not just speed.
Remember the basic electrical formula such as number of poles to rpm to get an idea of your torque curves.
This can be seen with pumps as well (technically the design he is using is closer to an impeller as it generates flow through pressure, so a higher torque motor might suit it better). You get different head heights. The max height a pump can pump to. Separate from the flow rate.
@@Pscribbled
Also I completely forgot to mention something in my last comment but can't be bothered to edit it.
The other reason for the lack of thrust is that Dyson fans are made for flow not thrust. A good comparison would be rpm vs torque vs horsepower.
With horsepower being your thrust and rpm being your flow. Pressure being your torque.
This is just a basic concept and not a formula. Since you seem to know cars.
LETS GOOO NEW VIDEO!!
Dear friend, as an old modeller, we have always known that EDFs perform less than open propellers.
The problem is in losses due to turbulence due to the space between the EDF propellant and the walls of the EDF chamber.
If you manage to leave that distance to a minimum then you will not have turbulence and you will not have losses.
sorry for my bad English,
a greetings.
Hi, as a Steamfitter the answer to your question is 2 fold. First a centrifugal pump works best with a perpendicular flow pattern that uses the "Centrifugal" force to release the fluid in the desired direction. Your casing requires the air to attempt to move sideways, restricts it, and forces it to change direction. I'm pretty sure a physicist could come up with an experiment that would show the losses as heat, in the casing, the air, etc. The reason a Dyson works is the very efficient turbine they use powers a very powerful force, the venturi. Because the air from the dyson comes out at a high speed thru a very small opening, and this is the important part, all around the inside opening of the ring, the air inside the ring goes along for the ride so to speak. The dyson doesn't restrict the flow like your design.
Thrust is directly related to the momentum (and thus the speed) of the air you shove through.
The motor should (al things being equal) dump the same energy into the flow.
With the EDF, that energy is in the form of kinetic energy, speed, and thrust.
With the impeller the work being done is instead meant to pressurise the gas, adding to the pressure energy instead.
A well designed nozzle at the rear should allow you to work with this and speed it back up, (though there will be losses and it needs calculating) and a good jet design requires the pressure as well
In essence, you should not be expecting the same thrust from the two of them, but if you take a look at the stagnation pressure it should be similar
I will make some suggestions by looking at your impeller design and the one of the Dyson. Look at the airfoil geometry, the attack angle is everything when it comes to sucking air, so I would test different angles and check their dependency on the intake diameter and the top impeller diameter. I would also check the tolerance between the impeller and the casing because it's responsible for a stalling phenomenon or for not compressing the fluid enough. Then for the outlet, how about creating a variable-diameter nozzle to test the right diameter to produce the same trust (like the ones combat planes have).
My best guess is that the EDF is utilizing Bernoulli's Principle to move much more air than it would otherwise (due to the viscosity of air), using the energy that is lost on friction between the air and the inside of the casing in the 3D printed fan to generate thrust.
Further explanation in case I wasn't properly clear: The air accelerated by the EDF fan rubs against the air around it and pulls it along, producing more thrust. The air accelerated by the 3D printed fan rubs against the 3D printed housing and produces heat. My hypothesis is that this causes the difference in thrust.
To increase the efficiency of the radial flow you have to have an axial flow stage in front of it. Radial flow compressors are inefficient at lower rpm and when sized down the “lower” rpm can be higher than you could safely achieve. Try an axial pre compressor
i would love to see you try the flat turbine engine concept, it basically uses a fan that has a specific profile that allows for the region closer to the center of the fan to act as a normal forward thrust fan, and the region towards the tips of the blades to push the air outwards ( like a centrifugal turbine) to copress air in a tight duct then cumbust it, this in theory should allow for a very compact turbine/turbofan engine.
Video Idea - show the difference of high Air Flow fans and Static Pressure fans. Their design is different in terms of blades and space between them. Greetings from Brazil! 🙂
Great idea!
Best idea!
@integza could you please confirm this with your youtube main account? Thanks!
@@metschnikowia likewise with mine i have no idea if the comment i got was a scam
@@antonventura6364 It is a scam. I wrote to integza. And also report to UA-cam.
🇪🇬I am from Egypt, I did not understand anything, but your style of speaking is amazing and beautiful. A new follower from Egypt continued
Try the scoops that are used in water wheels. It's a fancy blade that captures the most energy of a force. If it runs in reverse perhaps it can deliver the most force by using even less energy.
Hello, mechanical/mechatronic engineer here. A few things to note about the problems you are having.
1. Lower efficiency. The main reason is that (as you mentioned) there are extra steps. The process of compressing a gas is usually adiabatic, meaning that the temperature of said gas will increase as pressure increases. Likewise, the temperature also decreases as it expands (hence why propane tanks get cold). Basically, what often happens in systems like this is that the hot compressed gas will conduct heat into any object they are touching (in this case the printed fan duct), which is lost energy in the system and results in the gasses not being able to re-expand as much. This shouldn't cause a huge difference in efficiency, but enough to matter in some specific applications.
2. Centrifugal fans also spin air along the inside of the inlet (between the blades and the front cover) for some time as it is expelled outwards, which loses energy due to drag and turbulence. Unfortunately, 3D printed surfaces tend to be somewhat rough, which increases turbulence and therefore drag/resistance. 3D printed parts (especially ones that flex like you mentioned) also can't have as close of tolerances to the cover, so there is often a decent sized gap that air can settle in without being pushed down/out. The easiest way I know of improving both of these is by enclosing the top of the impeller, as the "lid" at that point is also spinning with the air and there are no gaps for air to escape through.
3. Cross sectional area. Another likely cause of the lack of thrust is the cross sectional area of the inlet. Remember that the larger inlet area will be able to move more air for each given rotation of a motor, so if the EDF inlet is twice the diameter (4 times the area) of the inlet to the centrifugal compressor, you will get a far smaller amount of airflow for the same rotational speed. This is partially why centrifugal compressors spin so fast (sometimes past 30k RPM), that way they can both displace more fluid and increase pressure differentials to more useable levels.
4. Fluid medium vs. speed. Finally, the pressure generated by a centrifugal compressor is heavily correlated to the density of the fluid that is being pumped. Since centrifugal compressors rely on the inertia of the pumping fluid to generate pressure, lower fluid densities will naturally produce lower pressures. This lower pressure will result in increased flow resistance inside the fan duct, therefore lower exhaust speeds. This is partially why centrifugal compressors spin so fast (sometimes past 30k RPM), that way they can both displace more fluid and increase pressure differentials to more useable levels.
If you would like some help designing an improved fan (I have designed a few centrifugal compressors for high pressure gas systems before) feel free to send me a message! I also have a number of high performance SLS/SLM 3D printers that are able to produce some significantly stiffer parts that would handle higher rotation speeds without warping or expanding.
Bump.
Also I didn't consider the inlet size, that's a pretty big deal.
And I believe there are lots of (lesser used) methods of smoothing out printed parts.
Aecetone, sanding, buffing, painting, etc. Painting is a particularly fun one because of surface tension xP
Damn this comment was very insightful, I learned 3 new things, that I hadn't even thought about before, even though I know the theory behind all those 3 things. I didn't know that the fluid's density mattered that much in centrifugal fans, but it does make sense, cause if you think about it in terms of steady flow energy equation, the centrifugal fan is exerting essentially a contact force on the air, and hence, the heavier the air, the slower it has the chance to lose inertia as it knocks on the insides of the fan.
Next is having a large enough inlet, having a large enough inlet makes sense to me in a sense of how venturi tunnels work, there needs to be a decent enough flowrate to start with, and I think that's probably the biggest difference in an EDF, where the fan has no blockage on it's front end. I can imagine centrifugal fans do a better job if air was forced into it, like in a jet engine scenario, but if air has a hard time getting into it, I guess the flow rate just won't be big enough.
I also knew that a fluid being compressed heats up, any thermodynamics module at a university covers this with how a refrigerator works, but it didn't cross my mind, that it would apply with centrifugal fans as well. But then again, I doubt that the work that is put into the air mostly goes into heating it up. It is probably something you should consider, but I doubt it affects the max pressure reached by such a large margin.
@@movin3148 Oh fluid density is super important,
desnsity is related to pressure remember, which is also related to velocity.
Pressure "flows" from high to low but in all directions, which also means backflow... To spin a fan means to compress a region of air, which means to create a low pressure region in front of the fan and an *even lower* region of pressure behind.
Which reminds me, you can also get backflow aft of the fan exhaust. Which is... interesting...
I love this guy's videos! I also did some research on fans for selecting the most optimal off-the-shelf fan for cooling a particular battery system. What I learnt was that each fan has an important characteristic, which is it's PdV curve; volumetric flow versus pressure difference between intake and outlet. The output power is the product of these (check the units tho, if you want Watts, you have need to consider the density of air). Axial fans have a very different curve compared to radial fans. The former can produce a lot of speed, but the latter can make more pressure and maximum power. Every fan has a point of maximum power output, where P*dV/dt is the highest. This may change depending on multiple factors, such as inlet and outlet area, number of blades and pitch. You want to match the motor's optimal speed for maximum power output to the fan's characteristics to get the most power out of the fan. I hope that someone finds this helpful :)
When making a thruster, you should order the metal parts from a 3d printing service.
Also try to make a rod spin at 1 million rpm using thrust
Double check your calculations on the magnetic cct. Are you slipping (coils are turning on faster than the permanent magnets, and keep up). With the air density and friction of the shaft, there is a bit of mechanical load there. So might need stronger magnets/electro magnets to create a stronger magnetic force to stop slippage. Feel free to reach out and I can explain more in detail.
ᴄᴏɴɢʀᴀᴛꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴏᴜʀ ꜱʜᴏʀᴛʟɪꜱᴛᴇᴅ ᴡɪɴɴᴇʀꜱ ꜰᴏʀ ᴛʜᴇ ᴘᴀᴄᴋᴀɢᴇ ᴄᴏɴᴛᴀᴄᴛ ᴛʜᴇ ᴛᴇʟᴇɢʀᴀᴍ ᴀʙᴏᴠᴇ
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^🎁
I feel this has something in common with the Rocket Nozzles you love.
My theory goes like this : when impellers increase the pressure of airflow, and the exhaust has the same area of cross section, more air flows radially outward from the exhaust due to pressure differences. The same way rocket engines use bell nozzles to reduce the pressure and increase the velocity of the exhaust, if you tried something to equalize the exhaust pressure with atmospheric pressure, you'd technically achieve the same amount of thrust.
Basically, you'll be converting the energy in the form of air pressure back to velocity to achieve the same thrust as the EDF. And considering EDF's don't mess with the pressure at all, it makes sense they're able to reach higher exhaust velocities than impellers at atmospheric pressure.
All the best with your project tho, looking forward to a cool video with whatever you're making...
Exactilly
Also impalers are heavier soo
Hello. First of all im a huge fan of your channel and sorry for the imperfect English.
I think I have the answer to your problem , in fact I even made the solution for it.
The problem is that radial fans are great for producing pressure while being Very compact. Meanwhile they are absolutely terrible at moving large volumes of air compared to their axial brother's.
Now , the axial fan does the exact opposite, its great for moving large amounts of air while being compact , but it's not great at creating pressure. Now, both can do what the other can as long as you make them "larger". For the radial fan/compressor, you basically wanne increase the inlet while not really making the base of the fan larger to prevent it compressing the air. Now you can only get so far with that without making it so large where there is no point in using a radial one over a axial fan. To solve that problem, you wanne give it a double sided intake, basically two radial fans with the base attached to each other. Pratt and withney did this with their J42 in 1948.
I recently made a 3D printed double sided radial compressor and I would like to share the files with you if you want them (so you get a better idea of the concept). In this video you can see it in a early stage, but its fully operational now and producing more thrust then my axial "jet".
ua-cam.com/video/glBI2g7bvzU/v-deo.html
The "and that's a very cute fan" bit genuinely brightened my day. Thank you
I would guess a couple things.
1) the EDF will direct air liniarly through the duct to produce a given thrust (which we will consider baseline) proportional to the rpm the motor can spin the fan, the pitch of the fan blades, and the drag the blades make.
2) the PEM will have more drag for the blades as it is not accelerating the air through the fan in a single direction (air has to go out to the sides brushing against the flat plate of the fan as well as the entire legth of the blade assuming it enters from the centre). More drag- lower than baseline thrust.
3) the PEM will have more mass than that of an EDF due to the way it functions, reducing the rpm of the fan blade. Lower blade rpm- lower than baseline thrust.
4) When it comes to centrifugal fans, tip clearance is difinitive to efficiency. The tighter the tip clearance, the less air you are losing to inefficient pathways through the fan... but i leads us to 5.
5) Too tight a tip clearance will cause the blades to clash with the housing of the centrifugal fan, creating more drag. There are two fixes, losen the tip tolerances and lose efficiency, or redesign the fan blade to be less forgiving to the centrifugal forces by adding support material, making the fan blades thicker, or making them out of a more appropriate material. Either way you risk lower than baseline thrust.
None of that however, takes into account the geometries effect on air pressure, as an EDF is not as dependant on the effects of air pressure like a PEM. So we need to increase pressure and use it effectively. Despite the lower than baseline thrust properties I expressed a PEM would have, it make up in increasing pressure, so we need to keep it.
So, a suggested to do list:
Focus on increasing air presssure generated by the fan while trying to keep airflow from the front of the fan through to the back of the fan as free ass possible.
ANY mating surfaces, ANY gaps will benifit from restricting or sealing off airflow providing its viable. You want the pathway you tell the air to go to be the easiest root for it to take. Air hates congestion. It would happily take the equivalent of longer country lanes if the motorways are congested, even if the motorway is quicker.
After that, focus on the exit geometries to utilise the pressure you have generated effectively.
P.S. dont hold me to any of this, its jusst educated guess work.
This man never fails to impress me
He's creativity amd ingenuity as a person
And that's without him eating tomatoes. Shows how tomatoes are overrated 😂
The big one: Tolerances values of the turbine vanes, to the shroud.
Other things to check: Intake, and exhaust port diameters.
Agree they are higher pressure and get this through tighter tolerances than other fans if you got gaps it will dramatically hinder performance.
I'm guessing that Integza already knows why the results are different, could be a number of possibilities from weight, aerodynamic shape, size, or direction of rotation. It also could come down to which end the electric motor is attached because the motor will obviously be more efficient when self rotating as opposed to sitting still and rotating the propeller/ fan 😁
❤ ᴄᴏɴɢʀᴀᴛᴜʟᴀᴛɪᴏɴꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴛʜᴇ ʟᴜᴄᴋʏ ᴡɪɴɴᴇʀ ꜰᴏʀ the 🤶 ɢɪᴠᴇᴀᴡᴀʏ ᴘᴀᴄᴋᴀɢᴇ ʜɪᴛ ᴜᴘ✉️ ᴛᴏ ᴄʟᴀɪᴍ ʏᴏᴜʀ ᴘʀɪᴢᴇ 🎁
You blinded me with that introduction, I’m just trying to sleep and this really bright light blinds me 😭😭😭
ᴄᴏɴɢʀᴀᴛᴜʟᴀᴛɪᴏɴꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴛʜᴇ ʟᴜᴄᴋʏ ᴡɪɴɴᴇʀ ꜰᴏʀ the 🤶 ɢɪᴠᴇᴀᴡᴀʏ ᴘᴀᴄᴋᴀɢᴇ ʜɪᴛ ᴜᴘ✉️ ᴛᴏ ᴄʟᴀɪᴍ ʏᴏᴜʀ ᴘʀɪᴢᴇ 🎁🎉
Thrust is mass flow rate multiplied by exhaust velocity. So either your intake is not large enough, or you're not fully converting the pressure generated by the centrifugal fan into air velocity (by having the nozzle be too small for example). The latter could also cause your "compressor" (which it basically is) to stall, reversing the flow through it.
this is what I'm thinking, I've remembered on seeing a video about jet engine saying that centrifugal turbine engine is better at building pressure then increase the air flow or something like that. Although to properly compare it, one has to make sure two motor are using same amount of power.
Damn, I was hoping I’d get to be the big hero who came along to say this. Still, my guess is what was described above (which also describes the difference between the huge thrust of a modern geared turbofan and the paltry thrust of a very high exhaust velocity turbojet) plus aerodynamic losses inside the airflow path inside the ducts.
I think the term you’re looking for to fix the mass flow problem is “entrainment”, which is how you draw nearby stationary air into your high speed jet stream to add mass flow at the cost of exhaust velocity, to gain thrust.
You could probably do this with - as already mentioned - a larger intake. Or you can use the method found in industrial machinery and Dyson fans, and do the entrainment after the fan, by using a duct structure in the exhaust that allows nearby air to enter the jet stream. This is probably easier, in terms of “it’s just a new exhaust” but might be harder in terms of sizing the ports and whatnot. But basically, exhaust gas goes into a tube, tube has ports that let air into it, tube expands to allow for the increased air flow, thrust goes up, exhaust velocity goes down.
The equations you care about here are:
power = (exhaust velocity ^ 2) * mass flow
And
Thrust = exhaust velocity * mass flow
And that quadratic scaling on the exhaust velocity to power is why cutting velocity to add mass ends up adding thrust.
The reason I've heard to as why centrifugal fans are worse is that you change the direction of the air as you push it out to the side then back instead of an axial fan which just pushes it back
@it's Morbin Time i hate You
From what I remember from my on and off again studies of aeronautical and turbine engineering , axial flow turbines or fan blades generate more thrust and more efficiently through airfoil lift like a propeller (hence a shift to high bypass turbofan engines for commercial airliners ) and generally move more volume of air , whereas centrifugal turbo chargers or compressors generate higher pressures and velocities but move less volume of air and have less low end thrust and are generally less efficient for thrust , especially because they have to redirect the airflow 90 degrees tangential then 90 degrees axial. this causes backpressure and friction and turbulences and other losses . the reasons why one is used over the other in specific applications depends greatly, but the general differences in cost , durability, technical manufacturing requirements , and pressure/ volume / efficiency / bulk / weight tradeoffs are usually the biggest factors.
I have had many ideas for new designs for piston engines and hybrid turbines that takes things quite a bit further .
I think you are on the right track , taking the drive turbines out of the back of the engine , as they are a huge expense and limiting factor of military and commercial turbines, as they have to build super expensive and complex single crystal alloy blades with internal channels directing cooling air through and out around them to be able to withstand the heat after the combustion chamber. the higher the pressure of the inflowing air into the combustion chamber and the higher heat of combustion ,the more efficiently thrust is made and the more thrust you get out of an equal size and weight of engine . and this is all limited by the drive turbines to withstand the heat and flow turbulences destructive effects. also , especially in high bypass turbines they are putting gears in , to change the drive turbine rpms relative to compressor turbine blades and bypass fan blades, so that each can operate at their respective individual highest efficiency at any specific airspeed , and altitude at any particular throttle setting . somewhat like when in cars , ignition and valve actuation was taken from the mechanical link of the crank and cams then were rebuilt to be computer controlled to be independently electronically actuated. it was more expensive and technically more difficult , but gave better performance and efficiency.
so the hybrid design changes to a turbine must make sense financially or performance wise or it is a waste of time no matter how fun it may be.
the centrifugal compressor makes sense in a turbine if the reduced cost of production , service , etc. outweigh loss in overall performance or efficiency , or added weight and bulk . also , since something has to generate the electricity to power the centrifugal electric motors , you have to factor that into the equation, what is the bulk , and weight and cost and efficiency losses of having a generator make electricity , and transmit it to the electric motors in the centrifugal compressors (losses in generating , transmitting and motors turning electricity to motion) , versus the turbine compressors and drive turbines being connected through a shaft with a lot less losses (friction of bearings etc.) but yet not being able to independently adapt to altitude , humidity , airspeed , throttle variables to produce optimal performance and/or efficiency in any circumstance.
good luck, it is something i have been pondering a long time, along with elements of rotary pulse detonation and passive venturi like flow inclusion of ambient air for added cooling and extra / more efficient low end thrust.
Are you trying to take the place of Moe, Larry, Shemp and Curly?
ᴄᴏɴɢʀᴀᴛᴜʟᴀᴛɪᴏɴꜱ ʏᴏᴜ ʜᴀᴠᴇ ʙᴇᴇɴ ꜱᴇʟᴇᴄᴛᴇᴅ ᴀᴍᴏɴɢ ᴛʜᴇ ʟᴜᴄᴋʏ ᴡɪɴɴᴇʀ ꜰᴏʀ the 🤶 ɢɪᴠᴇᴀᴡᴀʏ ᴘᴀᴄᴋᴀɢᴇ ʜɪᴛ ᴜᴘ✉️ ᴛᴏ ᴄʟᴀɪᴍ ʏᴏᴜʀ ᴘʀɪᴢᴇ 🎁🎉
Hi Integza. Typically, axial fans create ‘high’ flow at low pressure ratios. Impellers/centrifugal fans create ‘low’ flow at high pressure ratios. For impellers, this pressure ratio is a function of rim speed (and a bunch of other things).
Also, I forgot to add that your throat areas look significantly reduced moving from the edf to the impeller/ centrifugal fan.
Yep, was going to mention that. The dyson one he finished with is a mixed flow fan, kind of a medium pressure, medium flow type fan. Regarding the outlet area, since he has additional pressure available from this type of fan he could probably have a smaller outlet area, but not that much smaller. One thing I learned when doing some CFD work on a mixed flow fan is that you want the outlet area to gradually increase as you progress to the exhaust. We found that sudden increases in area would generate a lot of turbulence and reduce the fan flow. Also stationary vanes that "catch" the outlet flow from the fan and straighten it out also helped the performance, like the stator blades of a jet engine
Not to mention the impeller copied from a vacuum will probably be better suited to draw a vacuum than to move a lot of air....
The resistance is mainly created by swirling the air inside the motor. As it was written above, the diameters of the inlet and outlet should be larger. The angle of attack of the blades should also be changed, but it is best to make the impeller not made of plastic (not so weak to bend).
Hey Integza! My suggestion to fix the expansion of your impeller and increase RPM on the PEM is to add at least 2 more copper coils and decrease the distance in the way you coil the wire itself (it will probably create a weird hex design) I hope this helps solve your problem/idea! I'm also just now realizing I haven't subscribed or liked the video so I'm going to do both after I post this.
ps; I'm in college (ERAU) and this 3-D printer would change my life. I hope the internet helps me with this one, and Integza I hope you see this because I have so many ideas I want to share and create and I think this might be the start of making it all happen.
My theme for you next video is to mix this type of engine with combustion to increase the force it emits :)
Or print your logo with every printer you have and seeing clarity and definition differences only to destroy all the tomatoes in the end lol
As other people say, think about the relative inlet and outlet apertures. The fan is creating a pressure differential, the flow is just a secondary effect.
Also the aperture for the blower fan is a tubular surface infront of the blades. Not the circular transverse opening in front. The inlet could be choking the flow of air to the blades in the first place. Try a wider diameter ring of blades on the blower.
Also look for losses in the system (from recirculation and work done to travel through the casing to the exhaust). Normal axial fans have a little recirc around the blade tips but the housing is very open, there is no resistance to the exhaust.
Video idea - comparing the thrust from combining multiple pulse jets together vs a single scaled up pulse jet? Interested in the impulse profile differences!
It seems to me that the problem may be due to the shape of the impeller, perhaps if you look at the fan you mentioned you can notice the difference, because although the vacuum cleaner may work similarly, it is not the same as what you want
Would be cool to see a fan made like the dyson except multiple motors hidden around the diameter of it and used as a powerful compact leaf blower. Great theme imo for another video 😊
You are on the right track looking at your impeller design. You, however, looked in the wrong dumpster. Vacuums like the Dyson are high-pressure, low volume fans. The EDF is a low-pressure, high volume fan like the ones used in dust collectors. Try looking at woodshop dust collectors. Then, narrow down the exhaust port to increase the speed of the air, leaving the fan.
You should look at some fan showdowns. They went through a large number of designs and could help isolate the areas your combo may be having trouble in.
you could possibly use a centrifugal compressor design instead of an impellor, similar to Sir Frank Whittle's original jet engine or what is found in modern turbo chargers.
You could, also like modern Turbo chargers and Jet turbines, incorporate a type of "air by-pass" to add a bit of slower moving air "mass" (via ventricle shear) to increase the usable air volume at the ejection point (Exhaust).
Awesome channel, keep up the great content
the fan thing was pretty clever. nicely done sir
I think making an aero-spike water gun or hose would most visually demonstrate the advantages of this the aero-spike concept and could do well in the algorithm.
Velocity out the back matters.
Also the flow rate: Vel x Area.
Centrifugal drops velocity to raise pressure but like you said there's a lot of unnecessary steps => turbulent losses. Also lower velocities getting thrown out the back per unit area.
Best answer so far
I love all the things you do, super inspiring!! I'm definitely a fan ;) You also are the most in depth video maker about fans/engines etc on this platform, so here my video idea: Investigate and try out the toroidal propeller/fan that the MIT has developped (which has been in th news lately) and see what all it can do
Parabéns pelo teu canal, é incrível! Já tinha passado por ele algumas vezes mas não tinha percebido ainda que eras português :D Força com tudo ;) Já está subscrito!
Great video. 😊
Hey integza its more than a year that i'd been watching your videos they are just incredible and that rocket series is outstanding and i really love your affection with the tomato 😂. Why shouldn't you try making cryogenic rocket engines this is missing.... Lots of support and keep on making the videos 👍👍
I would love to see you make a see-through casing of a two-spool jet engine, with a high pressure and low pressure section for both the compressor and turbine. You can add a fan to the front to make it a turbofan as well if you choose. I think that the best way to understand the operation of something is to be able to see it, and with multi-spool engines (like every commercial aircraft engine being either 2 spool or 3 spool), most people have a large misunderstanding of exactly how they work. I also see a classic turbojet engine as the basis of understanding of all your other turbomachinery shenanigans which, as a "fan" of your channel, would help me appreciate your shenanigans even more!
Integza youer idea's are perfect and you try any thing to make it worke think you
My understanding is that ducted fans produce greater airflow in a low pressure situation, but centrifugal blowers are better creating static pressure. So fans are better for vehicles and blowers are better for inflating things.
Hi Integza!
My video idea: Could you make a video where you test different types of fans for your fans? Let me explain... I would be very interested in knowing the pros and cons of different designs, and you always do quick explanations in your videos and I enjoy that very much. So maybe you could make a compilation with different designs and what their applications might be? Also featuring YOUR all time favourite design in the end.
Thank you for your videos!
I think a nice video would be about a fan design for Major Hardware's Fan Showdown. I think you are able to create a very good fan with your knowledge and tools. Maybe you two can make a collab video togheter. 👍
This
I would watch this
Further observation adding a horn or bell shaped structure before the opening to the inlet would increase efficiency smoothing the flow hitting the compressor.
You could also try adding a tube on the front of the impeller. If you think about a car it has a specific air intake duct that restricts the air to that specific spot