Building a better crankshaft (crank and slider)

[paradisePhysicist]

I am trying to build a dual input (double input) crankshaft, out of Lego. All the tutorials online use flywheels, but I want a strong high quality crankshaft that doesn't rely on flywheels.

The crankshaft is for a 1 cylinder or sinusoidal linear type input. This video demonstrates the flaw of the current crankshafts:


Wikipedia (gif animation) makes the crankshafts look nice, perfect and smooth. But IRL the crankshafts commonly jam like the video above if you try to use linear input.

I can explain what is wrong with the crankshafts in this picture.

The arm only travels 179 degrees, the arm can never travel 181 degrees in order to get to the other side (unless the device is very precisely optimized and helped with flywheel). Therefore it only push-pulls the rotation instead of giving a full rotation. In practice, it actually gets jammed often around the 175 degree region, perhaps due to material warping or length axis torque.

Even though I see the problem, I haven't yet invented a solution. I am looking for neat ideas of how to build a better crankshaft. For instance, a Tesla valve is a neat idea for valves, I would like something equivalent in neatness to that. Something robust, not overly complicated, and efficient. The video comment section makes vague references to steam engines fixing this shortcoming, but that sounds like an air mechanism and additional power and complexity is needed, I am looking to just add a couple of extra levers or gears.

 

[cjl]

The lowest complexity solution by far is to rely on inertia to get you past TDC or BDC. What do you have against that solution?

(Another easy solution is to use a larger number of cylinders, in an arrangement such that they cannot all be at TDC or BDC at the same time)

 

[Lnewqban]

Could you show us a basic diagram of your dual input (double input) crankshaft?

 

[paradisePhysicist]

Could you show us a basic diagram of your dual input (double input) crankshaft?

I don't know how to build one yet, I am wondering if there is already something like that out there lol.

The lowest complexity solution by far is to rely on inertia to get you past TDC or BDC. What do you have against that solution?

Seems like it does not fix the essence of the inherent condition which is the original default crankshaft's tendency towards jamming. Even if the flywheel can fix the jams the crankshaft would probably be more efficient if the inherent mechanism is improved.

(Another easy solution is to use a larger number of cylinders, in an arrangement such that they cannot all be at TDC or BDC at the same time)

This is for a 1 cylinder engine, is there some way to get a 1 cylinder of power to basically link to 3 unpowered cylinders for this purpose?

 

[Baluncore]

Any sinewave has two turning points per cycle. You need two parameters to specify a vector direction. You need to find a cosine.

The minimum without a flywheel is a 90° V2 piston engine/pump.
The optimum for intake/exhaust fluid flow is a 120° V3.

A single crank can support multiple pistons, such as pairs of adjacent pistons in a 'V' engine.
A rotary or a radial engine has a single crank offset, with cylinders rotating about the crank.

 

[Lnewqban]

What does dual input mean in this case?

 

[paradisePhysicist]

Any sinewave has two turning points per cycle. You need two parameters to specify a vector direction. You need to find a cosine.

The minimum without a flywheel is a 90° V2 piston engine/pump.
The optimum for intake/exhaust fluid flow is a 120° V3.

Hmm, I don't know what you mean by turning point, are you talking about a sine wave has 2 sections, a positive and negative y section?

Are you saying a 1 cylinder engine with no flywheel is impossible?

A single crank can support multiple pistons, such as pairs of adjacent pistons in a 'V' engine.

Yes.

A rotary or a radial engine has a single crank offset, with cylinders rotating about the crank.

I'm not familiar with those type of engines, but I assume they are not 1 cylinder engines.

What does dual input mean in this case?

Around the 4 minute mark he demonstrates only the crank can be used as input, the piston cannot be used as input. A dual input crankshaft could use the piston as input with no jamming.

 

[Baluncore]

Hmm, I don't know what you mean by turning point, are you talking about a sine wave has 2 sections, a positive and negative y section?

Yes. The two stationary points between the rising and falling strokes are the problem.

I'm not familiar with those type of engines, but I assume they are not 1 cylinder engines.

https://en.wikipedia.org/wiki/Rotary_engine
https://en.wikipedia.org/wiki/Radial_engine

Before setting out to discover a better solution you need to understand the existing solutions. Replace the excitement of "uninformed independent re-discovery" with the excitement of "historical research" into the technology.

 

[paradisePhysicist]

Yes. The two stationary points between the rising and falling strokes are the problem.

Ah yes. Is there any way to fix that with connecting rods and such?

https://en.wikipedia.org/wiki/Rotary_engine
https://en.wikipedia.org/wiki/Radial_engine

Before setting out to discover a better solution you need to understand the existing solutions. Replace the excitement of "uninformed independent re-discovery" with the excitement of "historical research" into the technology.

Cool design. Anyway, I have searched online and haven't yet found a solution. Believe me, if there was already a device in history that had the specifications, and such a device was to be found in online search engines, I'd grab that in a heartbeat.

 

[anorlunda]

Anyway, I have searched online and haven't yet found a solution.

Solution to what? I don't understand what you are trying to accomplish.

 

[paradisePhysicist]

Solution to what? I don't understand what you are trying to accomplish.

Crankshafts have a mono-input problem, around the 4 minute mark the video demonstrates how a typical crankshaft can only accept input from the crank and not the piston. Engines compensate for this by using flywheels but I am wondering if there is an alternative way to get past this without using flywheels.

 

[anorlunda]

You can always invent a more elaborate, more expensive, less reliable substitute for the flywheel, but who cares? You could add an electric motor to the shaft, for example.

But if you want a solution that is better than a flywheel, that's difficult. You would have to start by defining "better", and your requirements for such a system.

 

[paradisePhysicist]

You can always invent a more elaborate, more expensive, less reliable substitute for the flywheel, but who cares? You could add an electric motor to the shaft, for example.

But if you want a solution that is better than a flywheel, that's difficult. You would have to start by defining "better", and your requirements for such a system.

A flywheel basically reduces the overall acceleration, adds more weight, and the heavier the flywheel is the more effort required to push the piston. It has some uses such as increasing stability, but it would be nice to have a mechanism which didn't rely on that and just worked inherently.

 

[Baluncore]

... but it would be nice to have a mechanism which didn't rely on that and just worked inherently.

Lots of things would be nice. Magic would be nice.

Fundamentally the power stroke can apply pressure to the crank for less than 180° of rotation. Clean emission engines need to be 4 stroke, so that is 180° * 4 = 720°. Without a flywheel, the minimum number of pistons and connecting rods must be greater than 4. The mass of the crankshaft and clutch provides some flywheel effect, which makes 4 cylinder engines possible.

Tens of thousands of engineers have been improving engines for over 100 years. Do you really expect to come up with a flywheel free, single cylinder engine? At this stage it seems you are so far behind that you think you are first.

 

[paradisePhysicist]

Lots of things would be nice. Magic would be nice.

Fundamentally the power stroke can apply pressure to the crank for less than 180° of rotation. Clean emission engines need to be 4 stroke, so that is 180° * 4 = 720°. Without a flywheel, the minimum number of pistons and connecting rods must be greater than 4. The mass of the crankshaft and clutch provides some flywheel effect, which makes 4 cylinder engines possible.

Hmm I will take your word for it, intuitively I'd assume you'd only need a minimum of 3 pistons to overcome the 180° limitation. Also, even with 1 piston there intuitively seems there may be some way to arrange rods and gears to get it to work, but maybe not.

Tens of thousands of engineers have been improving engines for over 100 years. Do you really expect to come up with a flywheel free, single cylinder engine? At this stage it seems you are so far behind that you think you are first.

No, I had given up and had hoped someone here would have ideas lol.

 

[Baluncore]

Hmm I will take your word for it, intuitively I'd assume you'd only need a minimum of 3 pistons to overcome the 180° limitation.

That would be for a two stroke engine which, due to exhaust emissions, could never be approved.
Three pistons is still significantly more than your one.

 

[Dr.D]

The video in post #1 is filled with misunderstandings and misnomers. It would be best to simply avoid such an ill informed source.

There are countless discussions of slider-crank systems in videos and books that do a far better job of explaining things (even with correct terminology -- that presenter insists on referring to the "cam" when it is functioning as a "crank.")

 

[Tom.G]

Something that may be a little easier to get your head around.
A crank is conceptually similar to a rack and pinion, with the rack (the straight part) being the connecting rod and the pinion (the round gear) being the crank.

Consider: when you run out of Rack travel, what do you do?

(above from:)
https://www.globalspec.com/learnmore/motion_controls/power_transmission/gears/rack_pinion_gears

Cheers,
Tom

 

[paradisePhysicist]

That would be for a two stroke engine which, due to exhaust emissions, could never be approved.
Three pistons is still significantly more than your one.

Ah I see. Could be used for pneumatic engines though.

The video in post #1 is filled with misunderstandings and misnomers. It would be best to simply avoid such an ill informed source.

There are countless discussions of slider-crank systems in videos and books that do a far better job of explaining things (even with correct terminology -- that presenter insists on referring to the "cam" when it is functioning as a "crank.")

Oh I had no idea (new to this). Basically the only point of the video was to show how crankshafts are a mono input, as for the rest of the video idk. In any of these books do they explain how to create a dual input crankshaft that doesn't require a flywheel?

Something that may be a little easier to get your head around.
A crank is conceptually similar to a rack and pinion, with the rack (the straight part) being the connecting rod and the pinion (the round gear) being the crank.
View attachment 285491Consider: when you run out of Rack travel, what do you do?

(above from:)
https://www.globalspec.com/learnmore/motion_controls/power_transmission/gears/rack_pinion_gears

Cheers,
Tom

Don't run out of rack travel? Lol

 

[hmmm27]

Basically - not counting on rotational momentum of say a flywheel - you can pull the upstroke from the driveshaft using the momentum of whatever is being powered, or spring the bottom of the piston-head, or hinge-spring the piston-rod andor crank pins.

What's the goal ?

 

[Lnewqban]

... Don't run out of rack travel? Lol

Not sure you could build it out of Lego, but just an idea:
https://grabcad.com/library/reciprocating-mechanism-1

 

[paradisePhysicist]

Basically - not counting on rotational momentum of say a flywheel - you can pull the upstroke from the driveshaft using the momentum of whatever is being powered, or spring the bottom of the piston-head, or hinge-spring the piston-rod andor crank pins.

What's the goal ?

Hmm I will look into that, not yet sure what that would entail. Think its doable with Lego but I am wondering why that isn't yet used with mainstream engines yet. The goal is to rotate a crank shaft with only 1 cylinder, without having to add the extra weight and decreased acceleration from a flywheel.

Not sure you could build it out of Lego, but just an idea:
https://grabcad.com/library/reciprocating-mechanism-1

Interesting idea, but I think that only can have the circle gear as input, there seems to a brief moment where the circle gear is not connected to the outer gear, so it wouldn't be able to have a piston as input. Maybe there is a way to fix it with modifications, but unfortunately Lego doesn't seem to have any half-gears at the moment.

 

[pbuk]

Why are you focusing only on the 'problem' of rotating from 179° to 181°? What is going to power the compression stroke from 181° to 359°? If you can solve this problem there will be enough inertia in the system to carry through the extra 2°. However this will still give you a very uneven force throughout the powertrain and this is one reason why even multi-cylinder engines have flywheels.

unfortunately Lego doesn't seem to have any half-gears at the moment.

They supply them in kit form:

Thinking about these things is good, but I think you need to get some practical experience: is there any science or technology museum accessible to you where you can see working models? I am lucky enough to live a couple of miles away from some of the oldest working steam engines in the world, but other alternatives are available such as getting a pit pass at a classic car (or even better, motorbike) race meeting.

Finally, I will say that you are not the first to have this idea and a single cylinder steam locomotive without a flywheel was produced by Nielsen in 1859: see if you can find out more information about how it worked (clue: count the connecting rods).

 

[paradisePhysicist]

Why are you focusing only on the 'problem' of rotating from 179° to 181°? What is going to power the compression stroke from 181° to 359°?

No idea, was feeling very hot and sweaty when experimenting with lego today, and also couldn't find the lego spring pieces I had earlier, so I couldn't test hmmm27's spring idea. I also can't seem to find the concept online. Yesterday I did think of adding a smaller gear underneath to get 360° rotation, then I realized maybe the problem is not actually that you can't get a full 360° rotation, but something else.

If you can solve this problem there will be enough inertia in the system to carry through the extra 2°. However this will still give you a very uneven force throughout the powertrain and this is one reason why even multi-cylinder engines have flywheels.

yes the uneven force seems like a big problem, especially with Lego because of the plastic, but to a certain extent real life as well, Lego can be helpful to show the locations of stress before converting it to metal, around the 170 degree range there is a lengthwise torque which keeps jamming the crank from rotating.

They supply them in kit form:
View attachment 285537View attachment 285538

I don't consider myself a Lego purist because I am willing to mix and match cadfi with other Lego technic pieces, however, cutting legos is where I draw the line. I was maybe considering paying someone to 3d print half gear legos, however the problem with Lnewqban's example is I think that particular example only accepts the circle gear as input and not the linear motion. There is another half gear configuration I discovered, but I am wondering why its not used in mainstream car engines, the problem I can see (though I'm not sure) is that if one gear slips at all, the whole system could disconnect totally, or lock up and potentially cause a disaster. The system also seems to require tooth gears and couldn't use smooth gears like a belt or other type of gear like system that has a potential to slip.

Thinking about these things is good, but I think you need to get some practical experience: is there any science or technology museum accessible to you where you can see working models? I am lucky enough to live a couple of miles away from some of the oldest working steam engines in the world, but other alternatives are available such as getting a pit pass at a classic car (or even better, motorbike) race meeting.

Finally, I will say that you are not the first to have this idea and a single cylinder steam locomotive without a flywheel was produced by Nielsen in 1859: see if you can find out more information about how it worked (clue: count the connecting rods).

I have been watching vids of older engines, the first engine seems to only convert linear input to other linear input, the Ford engine seems glitchy and random and requires a flywheel, the other engines I've seen requires a flywheel also. I found something called a Lanz Bulldog that has 0 rpm engine but afaik it doesn't convert the 0 rpm into locomotive rotational motion. I am going to look into the Nielsen design and see how it works.

 

[Dr.D]

and decreased acceleration from a flywheel.

But this misses the point exactly; the flywheel is there to provide uniform rotational speed (and to store enough energy to compress the charge for the next power stroke). The flywheel prevents the loss of speed just as much as it prevents a burst gain in speed.

Some small engines (think lawn mower), will not run properly if the blade is removed. This is because the blade is required to act as a flywheel providing the energy required for the compression stroke.

This whole thread is about beating ones head against the wall to accomplish nothing worthwhile.

 

[pbuk]

There is another half gear configuration I discovered, but I am wondering why its not used in mainstream car engines

I can give you two reasons:

• there is no point, a flywheel with appropriate inertia is actually a benefit in mainstream car powertrains, see https://ramclutches.com/blogs/debunking-myth-flywheel-selection/
• look how robust a conrod bearing is; each tooth of a geared mechanism would be subject to similar forces

 

[paradisePhysicist]

I can give you two reasons:

• there is no point, a flywheel with appropriate inertia is actually a benefit in mainstream car powertrains, see https://ramclutches.com/blogs/debunking-myth-flywheel-selection/

What about just adding a flywheel to the half-gear system, or would that just increase friction and result in no-net gains? (hypothetically for like a slow car such as 1 hp gokart where wear and tear isn't a big issue.)

And yes putting a too light flywheel will wear out the engine too much, I sensed this intuitively but then found a video with basically that in the title. My guess is because of that Lego issue I mentioned earlier about when it gets to the 170 degree part it gets caught in a perpendicular torque rather than the intended forward rotation of the crank.
https://ramclutches.com/blogs/debunking-myth-flywheel-selection/

• look how robust a conrod bearing is; each tooth of a geared mechanism would be subject to similar forces

Yeah connecting a fast engine piston directly to a gear might be too much wear and tear. Why don't the pistons use ball bearings instead of conrod bearings though?

 

[gmax137]

Why don't the pistons use ball bearings instead of conrod bearings though?

Sometimes, they do. Well, roller bearings anyway. See this:

https://www.cycleworld.com/hard-life-wrist-pin-bearings-50-ways-to-get-it-wrong/

Off Topic, the author of that link, Kevin Cameron, has a real knack for explaining mechanical details.

 

[Dr.D]

Look at the logic of this sentence from post #11:

how a typical crankshaft can only accept input from the crank and not the piston.

The OP is talking about an input to the crank from the crank. Does anyone see a problem here?

IC engines are complex systems, and to talk in terms of "inputs" ignores much of the actual interaction. For many of us, the "input" to the crank is the displacement of the piston due to the expanding gases above it. But the gases only expand because of the thermodynamic process taking place in the combustion chamber, so maybe that is the input?

This thread needs to be stopped; it is nonsense.

 

[paradisePhysicist]

Look at the logic of this sentence from post #11:

The OP is talking about an input to the crank from the crank. Does anyone see a problem here?

IC engines are complex systems, and to talk in terms of "inputs" ignores much of the actual interaction. For many of us, the "input" to the crank is the displacement of the piston due to the expanding gases above it. But the gases only expand because of the thermodynamic process taking place in the combustion chamber, so maybe that is the input?

This thread needs to be stopped; it is nonsense.

This is talking about a lego mechanism, not specifically IC engines. Idk what you are going on about honestly.

The video made it very clear what input means in this particular application, you cannot use the slider as input without additional modifications.

 

[Dr.D]

The animation in post #1 appears to be about a 4 cylinder IC engine.

I do think you were inadvertently correct when you said,

Idk what you are going on about honestly.

You are wasting the time of everyone on this board.

 

[paradisePhysicist]

The animation in post #1 appears to be about a 4 cylinder IC engine.

I do think you were inadvertently correct when you said,

You are wasting the time of everyone on this board.

Noone's making you post in this thread, if you don't want to then leave. If you aren't going to fix the specified problem then you are wasting my time as well and I just generally don't like dealing with toxicity/naysayers.

The post wasn't about 4 cylinder IC engines, sorry if the gif wasn't clear enough.

 

[gmax137]

I just watched the video linked in the original post #1. At best, it makes up its own terminology ("input," "cam"...). It is misleading as well. The stroke of the piston is determined by the throw of the crank, calling this the "semi major axis of the cam" only obfuscates what is going on. It isn't entirely incorrect, and might have some use if the device under consideration was say a cam opening a valve, but that is normally done with a follower and pushrod, or the cam acting directly on the valve stem, neither of which is a slider-crank mechanism. Finally the notion that the piston ("slider") cannot drive the crank's rotary motion can be put to bed by considering the millions of piston engines happily running since the dawn of the industrial revolution.

 

[paradisePhysicist]

I just watched the video linked in the original post #1. At best, it makes up its own terminology ("input," "cam"...). It is misleading as well. The stroke of the piston is determined by the throw of the crank, calling this the "semi major axis of the cam" only obfuscates what is going on. It isn't entirely incorrect, and might have some use if the device under consideration was say a cam opening a valve, but that is normally done with a follower and pushrod, or the cam acting directly on the valve stem, neither of which is a slider-crank mechanism. Finally the notion that the piston ("slider") cannot drive the crank's rotary motion can be put to bed by considering the millions of piston engines happily running since the dawn of the industrial revolution.

Do you have a better video to suggest? I did not intend for this video to be the "holy grail" of crankshafts just as a quick example to illustrate a problem. I also said several times (actually one of the first things I said) that of course with a flywheel the problem can be solved, and motion can be achieved, but that I was not interested in using a flywheel.

 

[Averagesupernova]

I've read through this thread as well as watched the video and I also don't 'get' it. The crank mechanism such as the one in the video works as it does because top and bottom dead center are undefined with respect to push or pull on the slider. The flywheel is a remedy to allow us to do what needs to be done while still leaving these two areas undefined. There is no way around it without adding complexity. It can be a flywheel, more sliders with timing that determines when to push, etc. But you cannot get around it by simplifying what we already have.