This was an EXCELLENT mini series. Thank you for creating and sharing. I finally feel like I understand gears, the math behind them, and various tradeoffs when creating them.
This was one of the best series on gear design/profiles I've seen in a long time. You made the explanations extremely easy to understand, and even included the math parts that I love. Thanks for sharing. I hope to see more like it especially with other types of gear. Personally, I need to make a 10-tooth gear. This series of videos are the only ones that I found that explains what I need to do to make it work efficiently. Thanks again.
A wonderful series with beautiful presentation and non intrusive but good math I can go back to. It scratches everything I want in an educational video.
THE BEST EVER. You simplify everything so neat and elegant, i could keep watching your videos all day. Your accent is amazing and the humour, bro! Great job! The world will be a better place if you have more subscribers.
Thanks :) The truth is, I'm more of a software and control guy. The video idea about gears kind of came to me randomly, and I think I did the best I could, but from this point onwards, I really lack experience. I'm thinking of building something as one of the next projects and I might use 3D printed gears for that, but there's really not much more I can say about the science and engineering of gears.
This is beautiful. Visualizations always help expand my understanding. I’ve been reverse engineering a transmission and this helps me further understand the OEMs use of positive profile shifted gears mated with negative profile shifted gears.
Excellent video once again. Might also be worth noting that the contact ratio is important because it is a measure of single tooth loading. Ideally this distance should be minimised this. The more teeth in mesh the quieter and smoother your gears will run. They will also have less chance of breakage.
The videos are great work - it took me whole weeks to find out the same things myself. I was willing to design some transmission mechanism using rolling instead of sliding - no luck so far, but perhaps it's possible using some intermediating bodies. Rolling friction instead of rolling cause a lot more energy loss. It would be great if somebody would make an episode or even better an app/algorithm that You type gear parameters and it outputs parameters on charts marking at what point other charts get them maximal values. So for an instance You type meshing wheels diameters and it shows You the most optimal closest gear sizes for given parameter we want. Do we want maximum performance (the least energy loss), do we want maximum strength ? Do we want maximum angular precision ? - I would advice helical (twisted) gears in that case. Do we need more strength ? (thicker gear with more contact area or bigger tooth). Do we need the least backlash (no clearance). I think that there could be a separate episode about the clearance itself - how do we cut the clearance ? equal gap from the gear edge, or, we cut the clearance with some angles - how to position them ? are 2 smaller gears meshing (causing contact) at once better than 1 bigger ? What's the optimal tooth width ? I forgot the context and wasn't digging into it too much, but at some point You said that some module proportion is standardised certain way and You don't know why, but that's the way it is - I bet that it's got something to do with being able to mesh with different diameter gears at once. Personally I found out that as many as short tooth's You've got as most energy efficient it goes as it becomes more like just rolling 2 pipes on each other, but it can withstand less load so You can widen the gears (increasing the mass and overall contact surface, but distributing the pressure on bigger surface. The question is at what width the gear is to wide ? If the gear needs to rotate at constant speed as heavy it is as good as more inertia, gives less angular accelerations - least backlash, but in case you need the gear to transfer accelerations you want it to be as light as possible to do not delay power transfer. After months of thinking about gears and bearings for my innovative engine project I designed a bit improved gears that should waste significantly less energy, last longer, stay cooler (less thermal expansion) and run smoother (with the least angular accelerations - in theory good designed gear is not having a backlash, but in practice all of them does, even the helical ones), but I'm still asking myself all the questions looking for potentially better solutions than using gears...
Whoa, lots of ideas here man, I can't say I 100% follow. On the idea of rolling contact, there's an excellent video from a youtuber called "Morphocular", "What Gear Shape Meshes With A Square?". Long story short, you can't do rolling contact and smooth RPM transmission at the same time. You can design odd shapes that roll on each other, but then when you rotate one with constant speed, the other will speed up / slow down periodically. I know an interesting result came out from some Japanese research about efficiency and backdrive-ability that resulted in split-ring planetary gears. You could look into that, though the math gets quite heavy for that one, being an academic paper and all... The backlash can be achieved 2 ways: the gear manufacturer cuts the teeth a little thinner - geometrically the same tooth curve is rotated by a small angle closer to its opposite. This way the application engineer can place the gears the nominal distance apart. The other way to do backlash is of course to move the gears further apart. For the geometry of the backlash, I think the closest distance between the non-engaging teeth is fairly constant - though I might need to check in simulation. However that constant distance might mean slightly varying angular deflection.
I haven't decided yet, there are many things on my mind. Maybe explore some dynamics, eg. Lagrangian mechanics, or get into simulations. I started a project on lenses which I never finished, but might get back to it, or BLDC motors, I also wanted to work on Kalman filters, adaptive control, eh... lots of things.
@@gergelybencsik8626 Lol, i think we might be the same person. I spent a lot of time learning about gears just to graph them with desmos. I recently also got some graphs of lens refraction working, mostly to see spherical aberration. Lastly i have thought about electric motors a bunch. Especially about if one can calculate the conversion from electrical energy (current and magnet strength) to mechanical energy (speed and torque). Sadly i haven't found an answer to that last one. Looking forward to whatever you explore in the future. It all sounds like interesting stuff 🙂
May I make a supposition: Pythagoras is the recognized 'inventor" of 3.1415...etc. and may be shortened to Py, not Pi. Your videos on gears bring about some curious prospects for me. I do not know how to draw a gear tooth profile with any accuracy, the only thing I could probably do is put a pencil on a string, tack down the curvature of the base of the dedendum diameter, and draw the tooth profile. I use ellipses to draw my gear teeth on my CAD software. Understanding your formulas is reasonably easy although I am not familiar with your variables and I don't think my ability to fabricate a working gear assembly wouldn't be all that difficult with the machine tools that I have. I am not quite up to that just yet but when I do I will find out whether I can make working gears models or not. I am not very familiar with undercut, and my default gear profiles are usually in contact on both sides of the tooth, so allowing some clearance will have to be discovered as to the gears spacing value from my calculations. I call it fudging the measurements. Thanks for your videos, they sound very well educated.
Excellent video! One question, do you know how to calculate the contact ratio between a normal spur gear and an internal spur gear? Thank you very much.
Sorry I haven't looked into that one. The line of action is a bit more mind boggling for inner gear but a very similar derivation should be possible for it.
amazing video,thank you !!! which software you use can make such simulation ,especially when the gear change the number of teeth ,and turn small continue,how to make such video ?
It was made with Manim. I'm glad you recognized some of the difficult parts, and the thing is, Manim doesn't necessary make it easy. It's a python based animation package, it lets you create shapes and move them around with python code. Lots of built-in functions to create graphs, numbers, equations, etc. That part you mention required a lot of custom code from me, and there's a bit of deception going on, since you don't see the whole gear, the off-screen part is quite janky.
This was an EXCELLENT mini series. Thank you for creating and sharing.
I finally feel like I understand gears, the math behind them, and various tradeoffs when creating them.
This was one of the best series on gear design/profiles I've seen in a long time. You made the explanations extremely easy to understand, and even included the math parts that I love. Thanks for sharing. I hope to see more like it especially with other types of gear. Personally, I need to make a 10-tooth gear. This series of videos are the only ones that I found that explains what I need to do to make it work efficiently. Thanks again.
A wonderful series with beautiful presentation and non intrusive but good math I can go back to. It scratches everything I want in an educational video.
THE BEST EVER. You simplify everything so neat and elegant, i could keep watching your videos all day. Your accent is amazing and the humour, bro! Great job!
The world will be a better place if you have more subscribers.
The explanations and animations are simply wonderful - thank you so much for taking the time to compose this series.
What an amazing series. Thank you so much. I will check your other videos. I hope you continue making videos about gears too.
Thanks :) The truth is, I'm more of a software and control guy. The video idea about gears kind of came to me randomly, and I think I did the best I could, but from this point onwards, I really lack experience. I'm thinking of building something as one of the next projects and I might use 3D printed gears for that, but there's really not much more I can say about the science and engineering of gears.
This is beautiful. Visualizations always help expand my understanding. I’ve been reverse engineering a transmission and this helps me further understand the OEMs use of positive profile shifted gears mated with negative profile shifted gears.
Fantastic presentation! As far as I can tell, nicely researched and the animations are just gorgeous 🤗
Maybe next time, include at least one obvious error. It immensly helps with the engagement 😆jk
Excellent video once again. Might also be worth noting that the contact ratio is important because it is a measure of single tooth loading. Ideally this distance should be minimised this. The more teeth in mesh the quieter and smoother your gears will run. They will also have less chance of breakage.
Bro amazin explanation I loved this mini serie, thank you so much for you work!
Excellent video. Why is it so hard to find good explanation of the reason behind gear shapes?
The videos are great work - it took me whole weeks to find out the same things myself. I was willing to design some transmission mechanism using rolling instead of sliding - no luck so far, but perhaps it's possible using some intermediating bodies. Rolling friction instead of rolling cause a lot more energy loss.
It would be great if somebody would make an episode or even better an app/algorithm that You type gear parameters and it outputs parameters on charts marking at what point other charts get them maximal values. So for an instance You type meshing wheels diameters and it shows You the most optimal closest gear sizes for given parameter we want. Do we want maximum performance (the least energy loss), do we want maximum strength ? Do we want maximum angular precision ? - I would advice helical (twisted) gears in that case. Do we need more strength ? (thicker gear with more contact area or bigger tooth). Do we need the least backlash (no clearance). I think that there could be a separate episode about the clearance itself - how do we cut the clearance ? equal gap from the gear edge, or, we cut the clearance with some angles - how to position them ? are 2 smaller gears meshing (causing contact) at once better than 1 bigger ? What's the optimal tooth width ? I forgot the context and wasn't digging into it too much, but at some point You said that some module proportion is standardised certain way and You don't know why, but that's the way it is - I bet that it's got something to do with being able to mesh with different diameter gears at once.
Personally I found out that as many as short tooth's You've got as most energy efficient it goes as it becomes more like just rolling 2 pipes on each other, but it can withstand less load so You can widen the gears (increasing the mass and overall contact surface, but distributing the pressure on bigger surface. The question is at what width the gear is to wide ? If the gear needs to rotate at constant speed as heavy it is as good as more inertia, gives less angular accelerations - least backlash, but in case you need the gear to transfer accelerations you want it to be as light as possible to do not delay power transfer.
After months of thinking about gears and bearings for my innovative engine project I designed a bit improved gears that should waste significantly less energy, last longer, stay cooler (less thermal expansion) and run smoother (with the least angular accelerations - in theory good designed gear is not having a backlash, but in practice all of them does, even the helical ones), but I'm still asking myself all the questions looking for potentially better solutions than using gears...
Whoa, lots of ideas here man, I can't say I 100% follow. On the idea of rolling contact, there's an excellent video from a youtuber called "Morphocular", "What Gear Shape Meshes With A Square?". Long story short, you can't do rolling contact and smooth RPM transmission at the same time. You can design odd shapes that roll on each other, but then when you rotate one with constant speed, the other will speed up / slow down periodically.
I know an interesting result came out from some Japanese research about efficiency and backdrive-ability that resulted in split-ring planetary gears. You could look into that, though the math gets quite heavy for that one, being an academic paper and all...
The backlash can be achieved 2 ways: the gear manufacturer cuts the teeth a little thinner - geometrically the same tooth curve is rotated by a small angle closer to its opposite. This way the application engineer can place the gears the nominal distance apart.
The other way to do backlash is of course to move the gears further apart.
For the geometry of the backlash, I think the closest distance between the non-engaging teeth is fairly constant - though I might need to check in simulation. However that constant distance might mean slightly varying angular deflection.
Awesome! What sort of topics will you be moving on to?
I haven't decided yet, there are many things on my mind. Maybe explore some dynamics, eg. Lagrangian mechanics, or get into simulations. I started a project on lenses which I never finished, but might get back to it, or BLDC motors, I also wanted to work on Kalman filters, adaptive control, eh... lots of things.
@@gergelybencsik8626 I’d watch just about any of those.
@@gergelybencsik8626 Lol, i think we might be the same person. I spent a lot of time learning about gears just to graph them with desmos. I recently also got some graphs of lens refraction working, mostly to see spherical aberration. Lastly i have thought about electric motors a bunch. Especially about if one can calculate the conversion from electrical energy (current and magnet strength) to mechanical energy (speed and torque). Sadly i haven't found an answer to that last one.
Looking forward to whatever you explore in the future. It all sounds like interesting stuff 🙂
May I make a supposition: Pythagoras is the recognized 'inventor" of 3.1415...etc. and may be shortened to Py, not Pi. Your videos on gears bring about some curious prospects for me. I do not know how to draw a gear tooth profile with any accuracy, the only thing I could probably do is put a pencil on a string, tack down the curvature of the base of the dedendum diameter, and draw the tooth profile. I use ellipses to draw my gear teeth on my CAD software. Understanding your formulas is reasonably easy although I am not familiar with your variables and I don't think my ability to fabricate a working gear assembly wouldn't be all that difficult with the machine tools that I have. I am not quite up to that just yet but when I do I will find out whether I can make working gears models or not. I am not very familiar with undercut, and my default gear profiles are usually in contact on both sides of the tooth, so allowing some clearance will have to be discovered as to the gears spacing value from my calculations. I call it fudging the measurements. Thanks for your videos, they sound very well educated.
The constant 'pi' is the name of the first Greek letter in the word 'periphereia' (our 'periphery'), being the perimeter of a circle.
This is really great video, could you make a video to explain the transverse contact ratio and overlap contact ratio for helical gears
Sorry, I don't have the tools and knowledge to go 3D for helicals :(
Could you make a video on visualizing the construction of bevel gears? I haven't found much on the topic so it would be much appreciated!
Sorry, but I'm moving on to entirely different topics for the next videos. I also can't do 3D animations yet, the tools I use are focused on 2D.
Top Class work
Excellent video! One question, do you know how to calculate the contact ratio between a normal spur gear and an internal spur gear? Thank you very much.
Sorry I haven't looked into that one. The line of action is a bit more mind boggling for inner gear but a very similar derivation should be possible for it.
amazing video,thank you !!! which software you use can make such simulation ,especially when the gear change the number of teeth ,and turn small continue,how to make such video ?
It was made with Manim. I'm glad you recognized some of the difficult parts, and the thing is, Manim doesn't necessary make it easy. It's a python based animation package, it lets you create shapes and move them around with python code. Lots of built-in functions to create graphs, numbers, equations, etc. That part you mention required a lot of custom code from me, and there's a bit of deception going on, since you don't see the whole gear, the off-screen part is quite janky.
Perfect say. Thanks.. (Türkiye/izmir)
very good:-) thank you
Kuch samaj nahi aa raha hai
thank you :)