Many commenters are forgetting exactly what the function of this transmission is: we want to be able to select for higher torque or speed depending on the situation. Sure, this gearbox may waste some energy, and could get beat out in a ramp-climbing competition by a better-tuned non-transmission car, but we really only care about versatility. A good experiment idea would be an obstacle course with many differently angled ramps, flat portions, and obstacles, so the viewer can see the true benefit of automatic high-torque and high-speed switching.

@NudelKungen.

3 жыл бұрын

My thought exactly, just one bigger cog would make this hill climb more efficient then both of these examples, but it would be horrible in flats and declines.

@theyeetus1428

2 жыл бұрын

But this design does not exchange speed for torque.

@TheAsturarYT

2 жыл бұрын

@@theyeetus1428 it does, technically. Energetic losses and changes in gearing means that this design always engages in some trade off between torque and speed compared to a fixed-gear system.

@Biriadan

2 жыл бұрын

A better designed automatic transmission without a friction clutch would have more power in low gear.

@Hyratel

2 жыл бұрын

15 Ncm of torque budget, 2.33:1 overdrive high gear outputs 6.43 Ncm 5 Ncm parasitic loss to friction clutch gives 10 Ncm in direct drive low gear

I really like this. Do you think you could put 2 in a row for 3 different possible speeds? Or would it be 4 speeds? Hmm…

@ferdoreznik

3 жыл бұрын

2 in a row will be only 3 speed and yes I want to build one and tune it up. Then I will post the video.

@ytxstream

3 жыл бұрын

@@ferdoreznik It would only be 3 if both stages had the same ratios.

@George_Bland

2 жыл бұрын

@@ferdoreznik Couldn't to in a row be 4? 1&3, 1&4, 2&3, 2&4

@netromrep6325

2 жыл бұрын

@@George_Bland The two transmissions must have different gear ratios or different clutches to make it work. If not, both transmissions will up and downshift at the same torque. I think….. This must be tuned in a way that transmission 1 downshifts at a certain load, and transmission 2 downshifts at a higher load. The result is 3 speeds. Gear 2 + gear 2, 1 + 2 and 1 + 1. Correct me if I am wrong.

A really great experiment! Ive seen the previous video where someone made a very good question about this and even you werent very sure. So you made a comparable experiment..Yay for science

AMAZING! I love your explanation, will certainly use automatic from now on in my builds. :) u need way more subs

I enjoyed this video quite a bit. Looking forward to the next transmission video.

Nice project and clear explanation!

Definitely works. The interesting question is what the torque vs rpm curve looks like compared to a fixed slow or fixed fast drive.

this is essentially doing what a torque converter or high/low range would do. it is just a very simple 2 speed transmission that uses clutch slippage instead of hydraulic pressure to shift.

Great video, I experimented with a similar gearbox and they work well.

Actually on the low gear range you have 10 / 1 + 5 / 2.33 = 12.15 If the friction clutch inside the white wheel disengages, it is because it is under load: the teeth are pushing and the whole kinematic chain carries on the torque to the other side. This device increases output torque under heavy load but decreases efficiency, just lique a regular torque converter on automatic gearboxes which usually "slip a lot" when launching the vehicle (bus and the like) and once it "synchronizes" at higher speed, the slipping dissapears, torque decreasese, and efficeicny recovers.

@azlktune

2 жыл бұрын

It does not increase the output torque. It just allows motor to turn faster at torque limited by the clutch. That seems that motor or its controller looses torque at very low rpm, like an ICE. So this is like playing with clutch on mechanical gearbox to keep the torque at very low speed and not stall the engine. I think removing of the differential and using an equivalent constant gear with the same clutch will give the same results.

@BayushiTawa

2 жыл бұрын

@@azlktune By output torque I meant output from the gearbox, so, towards the consumer part of the driveline. I did not mean output from the motor.

@azlktune

2 жыл бұрын

@@BayushiTawa Me too. Torque on output, by nature of differential, is equal to the torque at the gear on motor shaft. Torque on diff case is twice higher, and reduced ~2x times by the gears of 2nd shaft it is transferred to the clutch gear. So torque on motor =~2x torque on output of the gearbox. Independent from clutch sleep, because even with diff case stopped torque from it still transferred back to the motor through the sleeping clutch. But it allows motor rotate faster. May be gears on second shaft somehow work as one-way clutch and lock the diff to the gearbox case. Then torque on it may be higher then on motor shaft, and output torque will be higher. But there is nothing said about it.

@Bordpie

2 жыл бұрын

The analysis in the video is correct, it will put out 10Ncm (with the assumptions of motor and clutch torque etc.). The force going through the gear train through the clutch to the differential gear carrier has no affect on the torque output. The torque on the gear carrier comes from the reaction forces of the central bevel gear. If the central bevel gear axle was attached firmly to the frame or the gear carrier it has no affect on the output torque. A slight mistake in the video is assuming the motor torque is constant, which is not correct, it will actually change depending on load. The gear train from the clutch to the differential carrier contributes about 1.3/2.3 speed change of the gearbox since the bevel gears are a 1:1 ratio, therefore the carrier is contributing (2.3-1)/2.3. Therefore the clutch will be carrying this proportion of the motor axle torque. With a clutch of 5Ncm, the overall torque output from the motor just before slipping will be 5/(1.3/2.3)= 8.8Ncm. Therefore the maximum gearbox output torque just before clutch slip is 8.8/2.3= 3.8 Ncm. Just after the clutch starts slipping, the torque will still be 3.8Ncm just slower. The torque output will then gradually increase as load from the output to the motor increases. The 15Ncm torque value on the motor will be the maximum torque it can provide, and in that case the 10Ncm stated in the video is correct (assuming clutch torque is correct), while the fixed 2.3:1 ratio will only output a max of 6.5Ncm (although this is higher than the high speed of the automatic gearbox, it is not as high as the low speed).

Excellent, adding maths to explain behaviour of each part, helps a lot to reproduce the model. Also important to know the limits of each element to do not damage or break the motor or gears. Could be only improved if there is a measurement of voltage/current generated to the motor and it’s correlation to speed/ratio in each gear, in the four situations climbing a step, ramp up, ramp down using motor as break to slow velocity, and straight forward.

Awesome video! Keep up the good work :D

Excellent job!

I must say, I was skeptical at first, but you've convinced me. Major props, that's a a very simple design.

For me this is art

NICE, building this right now

Thanks for making this video ;-)

Amazing, brilliant and genius.

Excellent video. Great to see the test

So it's a mechanically controlled automatic transmission where slip causes a lower gear ratio? That's really cool! Totally impractical for a car, but very cool for this application. Very clever.

@ferdoreznik

2 жыл бұрын

Yes something like that. My goal was to make 2spd automatic gearbox as small and simple as possible.

Its really cool and innovative system. I inspired by you, so i designed this box but in my version you can switch beetwen two speeds in high Speed mode. Regards and waiting for more compact cool buildings✌️✌️✌️

I've seen you didn't put another video up, i will maybe try to get additional gears in the low speed output to generate more torque.

Simple. Brilliant. Just works!

Like it. Surprised that the makers of lego haven't used your design and put it into production. Could be used in many lego's products of vehicles

Excellent

Nice work : )

Would love to see a how to build video.

You should brake clutch somehow based on speed of output rotation... brake completely, if necessary so no power is wasted

Clever demonstration! Now what hypotheses will the skeptics come up with? :)

It works but if torque gets to high the friction wheel starts spinning backwards and the car doesnt move at all while the engine is still running. But still very nice idea.

Extra XL motor! Solved a problem!

@ferdoreznik

3 жыл бұрын

Yes, the smaller motor is not able to cover the clutch resistance in direct connection.

If you really wanted to make it efficient, you could design a friction clutch that had a torque curve with a hump (think of a compound bow, the way that you have to apply a high force to draw the bow back but only a light force to hold it there). E.g. a friction clutch that disengages but once it has disengaged it doesn't need heaps of force to engage it again.

Hello your videos are amazing! i have a question and i think your the perfect person to ask. How can i create friction/resistance on an axel were the wheels can spin back and forth in opposite directions?

@tazargroups.r.o.2152

2 жыл бұрын

This friction clutch gear used in this gearbox can slip in both directions.

I don't think that low rpm damages the motor, I am pretty sure that the motor gets damaged when it is under too much load, draws too much current and overheats.

@Void_s

2 жыл бұрын

Well, here he told, I supposed, not about just low RPM but defintelly forced down RPM, causing overheat, over-current and other bad things.

Well I think the people who commented fake doesn't know how an automatic transmission works

@tazargroups.r.o.2152

2 жыл бұрын

Yes and usually have sooo long explanation of their wrong opinion. 🙂

@oldhorn420

2 жыл бұрын

@@tazargroups.r.o.2152 yeah it's true

hi I'm still puzzled despite this being an excellent demo! I had to use 2diffs and a slipper to get good results, I can't work out how you do it with one .... 👍🍺

@yaduveerramendrayovaanadri6253

Жыл бұрын

tell me why didnt only one work?

Subbed

Helicoidal gears will greatly improve efficiency and slipping when you start going balls to the walls. Although lego doesn't make helicoidal gears.

This is kinda inefficient because in low gear the motor spends some energy to slip the clutch Here is an idea,you can use powered up xl motor and when the rotation sensor detects stress on the motor it uses another motor to shift to low gear the transmission used can be a manual one just use an l to shift the stick and an xl for drive

Can you do a tutorial I would love to try this in one of my vehicle

I think that instead of using the white gears which wastes. Allot of energy you can combine 2 differentials one with low speed high torque and the other high speed low torque and the will act as a CVT

Why the clutch gear doesn't rotate in opposite direction on low gear? Do you have some one-way stopper on it? Then it really may work as a low gear increasing output torque.

Just a seriously honest newbie question, I want to play around with tech Lego, but I don’t know if I need to buy several sets of preset box or there actually a maker set where they provide fundamental pieces so you can make stuffs, please help

Can you make one which can take more load since i would like to use something like this while motorizing my defender

What will happen if you put 3 or 4 gearboxes in series?

great idea, i have to test it out, but isnt there lot of power lost by slipping the clutch?

@ferdoreznik

10 ай бұрын

Not so big lost. Otherwise it cant work. But the weight of the whole cat is ctirical. If is heavier you need to use two cluthes in parallel to move the treshold higher.

i wanna make this in beseige now

I actually built the friction clutch transmission, but the power loss on the clutch is too large for my application

@ferdoreznik

3 жыл бұрын

Do you have too big car?

this is a very good automatic gear box but ... how do i get the white part it's called a clutch ?

I would love to see a trans this small that doesnt use the friction gear so the trans can not sap the engine of its power. Can you even make a trans this small even with something like a rubber band?

@tazargroups.r.o.2152

2 жыл бұрын

Yes can be. 2 speed gearbox with standard clutch. But it will be around double size. I will make it for you....

This gearbox is good.... but it is too sensitive for bigger machinery and it will almost always slip to slower gear

Where are you now come back

@tazargroups.r.o.2152

2 жыл бұрын

Working on new videos....

Its technically not a transmission though it's just a variable speed differential

Can you compare the friction clutch to a centrifugal automatic transmission?

@tazargroups.r.o.2152

2 жыл бұрын

I tried to make some cetrifugal clutch but it is not easy from lego. It is possible, but the final result is big. And that solution also need one way mechanism... more complicated than this.

Imagine 2 or 3 of these connected one to the other... Wonder what kind of torque power could be generated by that!

Would weight of the car impact the slippage? If a heavier car is used can, for example, two slip gears be used?

@tazargroups.r.o.2152

2 жыл бұрын

Yes, exactly. You can put two friction gears on the same axle.

@JaceN3D

2 жыл бұрын

@@tazargroups.r.o.2152 👍🏾👍🏾

Mount several gearboxes in serie and try more things

Ya

What are the part numbers for the diff and gears?

@dinhtuan752

Жыл бұрын

diff: 6514 I think clutch maybe part is 60c01 I think

You can improve this mechanism even further tap lock from behind your differential Gruß Philipp Hochscheid 😁😁😁😁😁😁😁

5:05 the ratio in the sentence is confusing. 1:2,33 is higher torque than 1:1. Did you mean to write 2,33:1? In that case, it would make sense.

@ferdoreznik

2 жыл бұрын

1:2,33 means one rotation on input makes 2,33 rotations on output. So it is gear to speed, not to power. 1:1 is slow gear in this case.

@underengineered_1077

2 жыл бұрын

@@ferdoreznik you are right. I apologize. I had just woken up and I was not thinking clearly. Haha! Have you considered making a version where your slow gear is more powerful, such as 2:1 or 3:1?

Just built this gearbox, one of the gears keeps slipping... Any tips? Cant figure out which it even is

@ferdoreznik

2 жыл бұрын

Only slipping clutch can slip. If the loading is too high it slips all the time. Try to lower output wheel rotation by some downspeed reduction.

what if u build another automatic gearbox after this?! i mean the torque come out and go in the next gearbox!

@dj1NM3

2 жыл бұрын

You're losing a third of the total available torque with each clutch-gear (15Ncm motor Vs 5Ncm loss in the clutch), so the losses from each clutch-gear should mean that it wouldn't work too well at all, after a second identical gearbox in series. The "straight through" low speed would be reduced to 5Ncm (15Ncm - 2x[5Ncm]), which is less than the "high speed" torque on the single gearbox does now (6.44Ncm).

Hi mate. That transmission is awesome. But i think you go it little wrong. Sure it add torque, but the rights way to conpare is vs system with direct feed. Not via same system when the gear is fixed. The clutch cut from the total power when its there. I think its about 5 Newton, if i remember right. From my experience any lego gear will be bad for performance. The friction is higher then the gane, in any case. You can ask bigger KZreadrs then me that said the same. There is 2 and even 3 people that work on that several years, and they all said the same to me. Try to stop the motor with you hand, one time with low gear on, and compare it to direct work, and you see yourself too.

I wonder if this could be also used for real heavy duty vehicles (1-2kg)... Maybe another test video?

@ferdoreznik

3 жыл бұрын

I think it can be, but the cage for the transmission must be much stronger. But I have plan also to make it with planetary LEGO gears which are able to transmit higher torque.

@npljojo

3 жыл бұрын

@@ferdoreznik OK nice, sounds good!

Now please do the same test on the same ramp with the same car but without the transmission, just fixed gears (using the gearing for second gear, for speed) to see how much power you lose to the differential in the transmission. I suspect a car without a differential might climb higher and faster.

@ferdoreznik

3 жыл бұрын

There should be only friction in gears. No other resistance. The fixed gear mode is without friction clutch. Thats why I made it switchable to make this comparison. Maybe using less gears will be a bit better but the difference will be very small. Even it will be few grams lighter 🙂. This video is just to see that there is stil possible to get more torque by reducing the speed. Also rpm of motor is important. Reasonable working rpm. Thats whe we use transmissions. Almost every ordinary car can start from zero also on 3rd gear... however, the most important point is that this is probably the smallest automatic gearbox and is workable for small cars.

@google911dancingisraelis8

3 жыл бұрын

@@ferdoreznik You say this but I would really like to see a "control". Your low Gear seems to perform much better because the high Gear wastes 50% power to the differential pushing a dead end. Then you shift to low Gear and all Gears start working, no more dead end. Let's say Gear 1 = 1.0 Gear 2 = 2.0 Make a direct drive system with a Gear ratio of ~1.5 I think it might perform similar.

@carlbModels

3 жыл бұрын

@@google911dancingisraelis8 hi, I put my twin diff CVT with slipper into a model with 540 motor... trashed the friction clutches and went back to fixed gear. Friction clutches are too fragile for 540 power. Great for Lego branded motors though! don't know about the single diff model but the twin doesn't lose very much power, but needs a lot of torque reduction down stream to avoid slip.

Can We get instructions please

@ferdoreznik

2 жыл бұрын

Please check my previous video. There is explanation of how it works. You can understand the principle and build it very easy. It is really simple mechanism.

how the friction gear works?

@ferdoreznik

3 жыл бұрын

Inside the gear there is friction clutch, using momentum 3-5 Ncm you can turn the internal axis without rotation of the gear. So under higher load the gear starts to slip. Gear remains stationary and axis is rotating.

@nediadarth4999

3 жыл бұрын

@@ferdoreznik thanks a lot :D

TL:DR yes you're losing about 37% of your power.

I don't understand it i feel like it is the same as you are in uphill with car and your car is in high gaer and you Half clutching if you don't push the clutch the engine will turn off

@ferdoreznik

2 жыл бұрын

No. The output rotatio speed is going down. The fast gear is disconnected and output torque goes up.

Will this work with a smaller Lego motor

@ferdoreznik

2 жыл бұрын

You have to check which motor. Since many small motors have less than 5Ncm. But you can reduce the speed of motor by some gears and then coonect it to input. You will increase input torque.

@dinhtuan752

2 жыл бұрын

philo tested all motors

This kind of maganismus are tolldely ansame

Make a car with a CVT

Gj

Hi

I waited for this, but I'm not sold on the explanation. I think you did not get what I meant. If you set the boundary at max capacity to climb of high speed gear, then the gearbox low speed gear can give you just 1% more torque and you'll be able to cross it. The test should be conducted at both max angles for high speed and low speed fixed configurations to show where in between those angles is the max angle that this low gear on the gearbox can handle, or some corresponding test. It makes sense for a gearbox to loose a fraction of its input torque for torque detection, but if you're gaining just few % of output torque while wasting a lot of it on the torque detection, just to cross an arbitrarily set boundary required to climb or move on tall grass, then it doesn't make much sense to have this kind of gearbox as it means you could just have slightly smaller gear ratio instead of the gearbox and it'd be enough. I'm not saying it doesn't work, just this video doesn't do this for me. I've got to get into the gearbox business to see it myself...

@ferdoreznik

3 жыл бұрын

But there is not just 1% more torque. It is over 50% more torque. And most important point is, that this is a model of smallest auto gearbox. It is just hobby. Why you make your gadgets? For some real life application? Of course not. You do it because you like it and also your subscribers like to watch it. Since it is interesting topic. Just for fun. That's why we do it 🙂.

@SaperPl1

3 жыл бұрын

​@@ferdoreznik That 1% was just to state the point. I'm not sure the way you calculated the output torque is correct - the fact that you need 5 Ncm of torque for the gear to slip doesn't mean you are loosing exactly this amount of torque overall, once you've crossed that 5 Ncm, you can lose more while slipping and it's probably not linear, so you slip, once you slip you loose a lot more torque, but then the resistance drops and the clutch catches on again and the cycle repeats. And for the "just for fun" part - yeah, I get that, but it annoys me that there's a lot of designs like this on the web so it makes it hard to find anything that actually works properly when used with a model scaled for it. I would really like to have an automatic gearbox that is useful for reasonably scaled models, but the web is bloated with designs that aren't that useful. I've realised that when I stumbled upon a build that stated having CVT, but the guy was actually throwing away the torque on two huge flywheels under the hood to drop the speed. Sorry to be party pooper, but I really, really hoped to find a viable solution for medium sized builds.

@morphles

3 жыл бұрын

@@SaperPl1 For me it also seems a bit suspicious about the benefits too. But if goal is max speed, not max torque, isn't such still decently nice? It allows higher gear ratio (thus more speed) than straight, while maybe having not as optimal, but still okish performance where more torque is needed.

@SaperPl1

3 жыл бұрын

@@morphles The issue is that transmissions are made for general range of loads, climbing steep roads, starting from stand still etc. If you do it like this, where you have a bare chassis with motor and transmission and you're just getting a bit more torque for specifically crafted challenges for it, then it's not really useful. If you build a complete model around it, you might end up with something that'll constantly be using lower gear or on a boundary. If you only have two gears, then so small difference between them may not make much sense. Maybe it would make more sense if the fast gear was a whole lot faster. I'm just annoyed that the whole web is flooded with lab use cases that aren't useful enough for actual MOCs.

1:36 *differential

Hi. I can edit your videos.

This video is so fake. 1. You are not runing it at full power. Large motor at full power would destroy your gearbox if car is blocked. 2. I see that you play with power or even hold car with hand when climbing with fixed gearbox. Tirs are sliding, but car not going further, tell me why? 3. Lego motor cant be damaged at any load. 4. If you want compare fixed with automatic, make a simple mechanism for fixed. You losing power with too complicated mechanism.

@ferdoreznik

3 жыл бұрын

I think you should try to build it.

@harynian

3 жыл бұрын

@@ferdoreznik I dont have Lego anymore, but I had XL motor and it could destroy gears with its power.

@harynian

3 жыл бұрын

@@ferdoreznik Im not saying that it doesnt work. Im saying that this test is far from truth. Car doesnt have good adhesion. This is cool project but still far from perfect and not usefull too much.

@ferdoreznik

3 жыл бұрын

​@@harynian the whole project was just demonstration of the smallest possible 2 speed gearbox. And because it is so small it can not be perfect. But it needs only few parts to build it. And if you do not believe to this video, then you should watch it over and over. Because there are no fakes. Absolutely. If I was playing with power, it was only because the car was not able to move, to show it. And of course the DC motor can be burnt in low speed. Any motor can be damaged if is running out of working RPM. Build it, test it and then you can make some judgement. I stand 100% behind this video.

@harynian

3 жыл бұрын

@@ferdoreznik you cant burn LEGO motor. People use 30V for LEGO motors and it still working. I have never seen, never heard about burned motor. I even tested it myself. I stalled motor and waited to warm up, but nothing happened. LEGO motors are safe. Yes, DC motors can be burned, but not all. Some of them will never overheat. You have just to check if motor overheats under load. This is the first thing what I do with new motors. Simple as that.

Hi