I think you're close - just need a planetary gearbox instead of a regular one - more gears per stage, less likely to strip teeth. Maybe a nema 11 size one? You can get them in 27:1 ratio.
@@mealex303 yeah, but a drill gearbox is probably better for torque than speed. Also anything that would be economical to strip for parts would have ABS gears most likely.
Hey Johnny, before you totally scrap this venture, maybe try some of the suggestions in this section. I didn't go through all of the comments, so sorry if I'm repeating the suggestions of others' but I would recommend; 1. Looking at impellor design to include stub fins inbetween the main ones to make double the number of fins, as seen on real turbo/S.C. impellors. 2. Stronger gears, either by using substantially scaled up gears with chunkier teeth or full planetary. 3. Changing pulley sizes to get more mechanical advantage (in this case, speed). You did so well making these components, it would be a crying shame not to continue tinkering with it some more. I can understand the frustration, & look forward to seeing the positive displacement type being built, but would love to see a revisit to this project. Keep up the fascinating work, love those skills! 😉👍
he needs to add a stator to build up pressure, same as the centrifugal compressors on jet engines. You try and research them, they serve the same purpose as stators in axial compressors. agentjayz has videos on his channel that have explained the significance of stators.
@clicking on trees Dude stfu like you would even try to do this shit and figure out everything. Hes trying and its not a tutorial sp go take you stinky comments to someother video
@clicking on trees By watching his design & build skills, one can clearly see this guy has a lot of knowledge about materials & engineering. I don't think he started this thinking, "Brass is SO STRONG - I can overload it, and it will be fine!" I think it was more along the lines of, "How could I make this work - even with brass?" Also, there are other material properties of brass that makes it a good material to use, even in high speed cases. Things like friction and bearings come to mind. I think that He was just trying to make it work, and maybe the brass was there to be soft enough to "deform around the other gear teeth," maybe to compensate for any imperfections at this tiny size? What do you think? I don't think he would go to this much trouble for click bait. So many other things he has built have worked really well. Have a good one.👍
WOW ,, i'm surprised it ran at all ,,you'd need titanium hardened gears to run that kind of RPMs for any length of time ,,, brass just wasn't meant to run 156K RPM
My thoughts exactly. While I’m certainly no expert, such high RPMs no doubt require near perfect balance. Something you’ll never achieve with a Dremel tool.
Well, you can, but you must balance it. And you can balance with dremel and proper tools. Moreover those bearings are not rated for these rpms and ceramic would work much better. No cooling made to the gearbox as well
aWhen Turbochargers and centrifugal superchargers are made, they have those smaller fins behind the main fins. The idea is to break the turbulence and stop the eddy current air that can keep you from Gaining boost. Before you completely scrap the idea, you might want to look at impeller design. I would certainly love to watch you build a positive displacement supercharger though... there are several different designs to choose from, watching build a Roots blower would be fun. A screw-type might be louder though.
I just want to add, I use speech to text a lot, I didn't proofread what I wrote, so I apologize for looking like much more of a moron than I really am. LOL
The issue is with the scale, as centrifugal requires having air *mass* moving for it to do its thing. Really small turbines only seem to work ok with liquids or gases that are already compressed to a high level to begin with.
The other problem is the low tip speed of the blades due to how small the impeller is. If you cant get enough velocity at the outer portions of the impeller, it will never make any real boost.
dont worry, when you got the interest.. thats all to it.. i was about the same where you are now a few years ago. i dont own a cnc.. but i do have my own milling machine and a metal lathe. after getting used to milling and lathe work i feel confident that i can do this 2, t i realize that my limitation is just not having a cnc mill and nothing els..
If he goes automotive style downscaled it would be very hard to do but there are some super basic designed positive displacement blowers in the industrial world that are extremely basic.
I second this. Material composition is also a primary concern once you start pushing tolerances including temperature. And five to six figure RPM speeds is nothing to slouch at. With that speed, gears would start hitting speeds where material composition would become a huge factor as gears rotate and heat up themselves along with surrounding pieces towards points of expansion or breakdown/torsion max and then breaks happen along stress points like the gear teeth. Brass is a rather flimsy material to use if you want to tolerate heat because it expands at rather lower temperatures than other metals which is one of the reasons why it's used for casing of ammunition. Steel would've been a better material choice, titanium if he has it on hand or a hard aluminum like 7170.
You need to put a stator after the impeller, or some sort of analogue to that. The air must be slowed down to increase in pressure (and thus give boost). Otherwise you just have a fancy centrifugal blower. If you look Up centrifugal compressor I am sure you will find the info you need. Agentjayz has videos on his channel about stators, and why they are needed.
The problem is that it isn't compressing the air, a lot of airflow is good, but it's pointless if it's not compressed, the denser air is where you make the power
You can look up the rotrex centrifugal superchargers. The use a panetary "gear"train that uses hydraulic friction instead of gearteeth. It may give you some inspiration to build a version of your own. I'd say give it a go
Even tho there is no contact betwen parts using fluid viscosity as a mean to transfer force the gear box can still be destroyed if run on extreme conditions or not be efective at all in some conditions
@@bernardoalbano1816 yes, that is true, everything has a breaking point, however the use of this specific type of transmissions in this specific scenario has been proven efficient and effective. So I concur to your point in theory but I fail to see how it applies here
Amazing to watch and gives great satisfaction. "Turbo charger" "supercharger" who cares I would like to see the ones who are immensely rude like to do anything remotely interesting unless they are creating space rockets in their shed and spare time. Which I seriously doubt it.
@@mobbdeep615 No, turbochargers use the exhaust of the engine to shove more air into the intake and superchargers use the power of the engine and are driven by shaft or belt to shove more air into the intake.
Amazing skills as always. 2 gears in housing 60 primary 10 secondary on impeller shaft , big gears off a 1/8 rc car driveshaft / motor. Also inlet diameter plays a small roll also on vacuum pressure you go to big for impeller you loose pressure, Hope you finish it .. 3D print some different size inlet restrictors 0.5, 0.75, 1mm smaller than you have to plug and test. Can't wait to see it go and the blower build but this will work with your craftsmanship. GOOD LUCK
Your problem is sealing. If you want to attach this to a glow engine, you'll run into another problem; the carbs do NOT work worth a damn if they're downstream of a compressor. Do one more test with a centrifugal compressor. Put the carb upstream of the compressor. The castor oil in the fuel should seal the impellor and allow boost to build. You could probably get away with a 1:5 or 1:10 gearbox if you had extra sealing and I think itt'l get plenty if it's pulling A/F in through itself. This is how they're set up on actual hotrods that have carbs, too. The carbs sit on top of the blower.
Draw though setups (carb pre charger) are easier to get running, but are not very safe in full size applications as the fuel also gets compressed the whole journey to the engine, which turns your entire intake system a potential bomb if a backfire occurs, unless a pressure release fail safe is used, & as such, they aren't good to use with intercoolers, although I have seen aftercoolers used with this setup. Blow through setups (carb post charger) are tricky in that you need to pressurise the float bowl (or fuel metering block) to equalise the boost pressure thus to allow fuel delivery. The benefits are that only air is compressed through the intake tract, making intercooling more viable & safer with the fuel added closer to the engine. Model engine carbs generally don't use float bowls so the fuel tank itself would need more pressure than the usual exhaust pressure used to push the fuel through the carb jet, like a boost bleed line to the tank. EFI solved these problems in full size, but we're probably some way off seeing that in RC applications, although probably not that far off!
@asshat Jackson That's a valid point you made, perhaps the tiny bearings he used on the pocket turbojet would do the trick. Sorry I don't know how to put links in comments, but you can search his channel. It was from 7 years ago so will be well down the list of vids.
If a 6-71 was good enough for the last of V8 Interceptors, I hope it's good enough here. Very cool project. Easy for me to say especially given the title but after you said "we need 5xs that," and revved it, I though "it's hurt."
I feel like if the impeller shift was press fit to the bearings in the housing it may not have failed as soon. I understand brass is soft, so if the impeller was to shift and cause friction between the bearings and the shift it may put too much load in the brass. Great jog with the videos. This is so cool to watch. I am a diesel mechanic and to see the mini version of an air charger like this amazes me. Keep up the unbelievable work!!! 👏 👍
I think you need to improve compressors wheel design, to build boost. For gears, oil or grease, when they heat up, they lose viscosity, so continious oil delivery (circulation) would be good thing to do. Like a mini pump up top, and a small container to colect it on the bottom
Also mate, Try a different oil rather than glow fuel. A low viscosity group 3 synthetic. Oh and yeah, don't give up. We're all watching and hoping the best comes of this project. Keep going. Oh and nice work with using a bit of Moly grease, I love me a bit of Moly!
When he says glow fuel I'm assuming he means RC nitromethane mixtures. For those fuels the incorporated lubricant is only mineral oil, which has no film strength and would not handle high pressures and soft metals of the small surface area gears.
No chance, a drill gearbox is made to turn 20-25k rpms from a motor into usable 500-2500 rpms at the chuck, and its big and bulky to take 70-80nm torque. you want the exact opposite, small, light, and it doesn't need to take a lot of torque but it needs to be able to turn that same 20-25k rpm into 150+k rpm, and a drill gearbox just isn't made for that speed, infact I don't know any existing gearbox that is, at this scale...
@@stevend4544 I don't know if .5-1 psi of boost would make any more difference on this than a car engine. It has a much much smaller cylinder volume so boosing to .5 a psi would only add like 1/15th more volume of air meaning you could generally only use 1/15th more fuel than before. Theoretically it would make only 1/15th more power than it would before. For example it isn't very unheard of to have atmospheric pressure of around 15.2-.3 psi at sea level in certain conditions.
Hey, It has been so interesting to watch videos of this project. Please, don't give up on this supercharger yet! As many youtubers do, ask for some online help. I am sure someone is able to run a CFD simulation on your impeller design and tweak it to be more efficient. Another person would be able to help out with the gearbox. This project seems to be so popular that it should not be only your project. It is ours too! :)
You built a centrifugal supercharger. The issue of your lack of boost is the outer diameter of the impeller/compressor. You need a larger diameter expeller area compared to the induction cone area. you are working with such a small piece a few MM larger diameter is a big increase in velocity. That outer velocity is what will increase pressure and volume of air moved. On the hybrid compressor for my k04-023 turbo the inner housing was machined 10mm larger. A 42/66 compressor wheel fitted vs the stock 40/56. big increase in cfm at the same psi. Im thinking a scale of 2.5:1 or 3:1 expeller to induction cone area giving the compressor more area to compress. On your small scale and induction cone of 5mm and an outer diameter of the expeller blade of 15mm would be much more efficient. You saw it pumping air previously to fill the balloon. At 100k rpm a 10mm diameter compressor expeller is only moving 52.3 m/s. 15mm=78.5 m/s. 20mm 104.7 m/s. Outer diameter of a centrifugal supercharger is important. I agree with your idea of moving to a roots style positive displacement. Just make sure you are calculating the positive displacement compared to the overall displacement of the engine you are installing it on. Too big or too small you create inefficiency.
Dude. You have to stop screwing around with brass gears. Brass is not a strong enough metal for your application. Consider how much force the impeller must exert on the air to compress it, and then consider that the gear connected to the electric motor has to withstand _17.2 times as much force._ You require steel gears -- preferably in a planetary arrangement to distribute the load across more gear teeth. Those cheap no-name servo gearboxes are a complete waste of time. Alternately, skip the supercharger concept entirely and build a turbocharger instead. Your rotary engine is a 4-stroke engine, so it can still work if the exhaust has backpressure from driving a turbocharger. Then you won't need any gears at all.
Number one, it's not about practicality or hardly even feasibility, it's about learning and experimenting. Secondly, it's incredibly hard to measure boost in that small of a supercharger.
Not being snobby but how is a rotary to be measured on the strokes? I feel like that's a difficult thing to determine. I don't think that rotary's are two stroke or four stroke, correct me if I'm wrong
@@hadleygirty8449: Wankel engines are four-stroke engines because they have distinct 1)intake, 2)compression, 3)ignition/combustion/expansion, and 4)exhaust phases. Two-stroke engines don't have those distinct phases; the intake and exhaust phases in a two-stroke engine occur simultaneously, which does not describe how Wankel engines behave. It only _looks_ like a Wankel engine might not be a four-stroke because the design allows three separate combustion chambers to share the same hardware, whereas in a piston engine each combustion chamber requires separate hardware. A single-rotor Wankel engine is most comparable to a flat-2 piston engine with a boxer crankshaft (basically half of a Subaru WRX engine), so each full rotation of the crankshaft experiences a single ignition event. It would be more technically correct to call Wankel engines Otto-Cycle engines or four-phase engines instead of four-stroke engines, since a rotor doesn't reverse direction at the end of each phase of the combustion cycle (i.e. a "stroke") like a piston does, but the commonly-understood term is "four-stroke" so that's the term I used.
I think we can all agree that despite the failure it was a rad disassembly. This machine is beautiful. I want one just to hold in my hand. (I've been watching all these videos and I still don't know what this device is used for. An RC car/airplane? I wasn't paying attention) I feel like this could be used as a mattress blower upper that stays attached to the mattress rather than having to carry a separate compressor. Side note: You could drill a small hole in the gearbox, cover it with plexi and use that micro lens with the long tube on it with a high speed camera like the one MKBHD and the SloMo Guys made a video about a month or two ago. See where the gears are catching. Cross over vid? I'm probably full of it - I don't know anything about anything but I love these videos.
I am unsure of your claim that it's NOT making boost at it's scale size, I just think you need a more sensitive gauge. Atmospheric pressure is somewhere are 14 psi and the fact you were able to blow up a balloon with it shows that you were able to get something out of it. 1 psi of flow out of that is probably extremely significant to it's application.
For years I've been watching you progress and each time I return I'm never disappointed. If you're not producing and selling these or a real supercharger in the future I'll probably feel sad.
Getting enthralled in your videos. Fascinating. As far as this supercharger goes, I imagine the centrifugal force on the impeller will want to throw the vanes outward. If they are very thin they may warp slightly and if your tolerances are very tight I guess the vanes could rub/bind on the housing? Fair play to you though for giving it a good go.
The supercharger is actually making boost since it can inflate a baloon. To meassure boost you need a restriction. The reason why you can meassure boost in a car that is supercharged is because the supercharger is outputting more air then the engine can consume at that rpm.
I think the biggest reason the gearboxes keep failing is torque multiplication... The impeller requires x amount of torque to overcome the friction of it and the air it's moving (and compress it). In your case given the scale, might seem negligible, but for every time you're doubling the speed it also has to double the torque to achieve that. There's also a limit to how much air can be moved/compressed in the given space of your impeller, after that there is a huge increase in resistance. Fluid dynamics aside, think about it like a servo gearbox in reverse, the motor is easy to turn with your finger when it's not connected to the gears, but try turning it from the output shaft and it's a lot more difficult. If the impeller was the input shaft, you would get loads of torque on the other end (your current input) and the gears on that end would need to be able to handle that torque. Unfortunately that very same torque is what your issue is, except it's on the input side, and you're driving that input with a motor that far exceeds the torque handling of the input gears in your design. This is an issue you are going to have in every iteration and supercharger type: getting the necessary RPM to compress (which adds a significant load), vs the space you have, vs the friction and consequent input torque to drive the gearbox, vs the amount of loss. If I had no other choice but to do what you are trying to, I would work backwards, first see if the impeller/s can actually compress the air to your PSI (PSI being pounds per square inch) that will tell you how many pounds your impeller needs to support. Then you need to work out the area that force is distributed over and the angle that that force is being applied in. That will define the design and size of your compressor (impeller). From that design, you should be able to calculate the final torque required to achieve that compression. Then you work backwards through the gear ratios, each time the output speed is reduced (it has to in this direction if you want to multiply the speed from the input shaft), you need to calculate the new torque being placed on the gears in that step and use the appropriate material, based on that materials shearing torque characteristics, all the way to the final input, which will need the most durable material to shearing forces. That will determine the final size and efficiency of the compressor and drivetrain. It will need to be highly efficient, low friction and made of stuff that isn't brass or aluminium (or at least the gears won't), but there's so much more going into a design like this (axial loads of the gear bearing surfaces, appropriate lubrication, viscosities, losses in compression and heat etc) that probably require more than one mind, and those minds would need to be somewhat experts of all the relevant fields. Sadly I don't know enough about any of them to be further help, but I do know enough to know what the issue is, and I know that even changing the style of compressor is not going to fix this issue (unless it's to a source of already compressed air). I might be wrong and very much am open to correction and or additional input, and if I am wrong I hope you succeed. Also, you don't necessarily need the high RPMs, you need compression and flow (higher than the engine can consume for the given RPM), so tighter tolerances and more efficient design entered around those goals will yield the best results. You might even end up needing a reduction in RPM.
Maybe no more brass gears. Brass at 156,000 rpms is kind of an invitation to disaster I would think. It's a lot of torque on a small surface area. The soft metal did what soft metal does. It conformed. Love your content!
May I come with a suggestion... Ok I don’t know what you are aiming for with this project but I think you should consider working on the impeller design. If you are after good pressure gain you should have an impeller with more curved blades. It might be quite tricky to manufacture such an impeller but I’m sure it would do great on the pressure readings. Rule of thumb: The straighter the blades on the impeller are, the more the the impeller will act as a fan (more flow but less pressure). Unfortunately with these tiny dimensions the resistance for the airflow will be so high that not much flow will be accomplished anyway. With this setup you will need a good defusor right after the impeller to convert some of the airflow to pressure. With more curved blades the impeller will instead compress the air (for the cost of some flow). With higher pressure allready when leaving the impeller, the airflow will be less affected by the tiny dimensions of this supercharger. Looking forward to see this project continue
keep in mind that it needs actual airflow to create pressure. if you block turbine air outlet, it stalls and pressure drops due to internal turbulence. it's the same exact effect how your vacuum cleaner loses efficiency if you block its air flow
Man you are one heck of a machinist..I'm working on getting a lathe and a small mill and hopefully one day I'll be as good at it as you are..until then all I can do is practice practice practice..that's the only way to get good at anything..
I think your impeller design needs revision but now that I just seen you say you're likely to build a positive displacement roots style blower, I'm all for the roots buddy!!!! You make such fine great parts bro! Keep up the good work buddy and thanx for the awesome machining etc content!!!! Have a great week!
How about: Turbo ---> intercooler ---> roots supercharger Excess boost adds directly to crankshaft torque via supercharger belt - pop off valves disabled during full throttle, enabled during half throttle and idle.
