This topic has been discussed in the forum, precisely in the topic: "the question of the day" (link at the end of the text, message n.4272).

In order not to lose (the search among thousands of messages will be almost impossible in a few months) the contents I insert the text in this article.

I hope it can be useful to many if not all.

The topic concerns the problems related to the junction between the engine shaft and the propeller shaft and in particular the Cardan joint.


The single or simple joint (what in Italian we call a "cardanic" joint but that Google persists in translating incorrectly as a universal joint) is the one in images 1, 2 and 3.

Be careful, many sellers (especially online) call the cardanic joint a universal joint. This is wrong.

To the first question: "Using a single joint, what problems could it cause?"
The short answer is:
"The problem is that it causes vibrations. Consequently, also more noise.
The vibrations are greater the more accentuated the angle between the axes and the higher the rotation speed. The vibrations are a direct consequence of the different rotation speed, since the cardanic joint is not a constant velocity joint (homokinetic).
In the cardanic joint, unlike the homokinetic one, the instantaneous angular velocity (note "instantaneous") of the driven shaft is not constant during a complete rotation. Logically, the average rotation speeds of the two shafts are the same."

Let's see why in more detail.

The cardanic joint is an exceptional element because it allows the rotation to be transmitted between two axes that are on different planes and therefore form a certain angle between them (see figure 4).
In Italy we call it "Giunto Cardanico" or simply "Cardano", from the Italian mathematician Gerolamo Cardano (1501-1576), who rediscovered it in 1545. In reality, the invention of this type of joint dates back to at least the 3rd century BC, by Greek scientists such as Philo of Byzantium.
It is based on a fundamental central element with four axes called "cross" (see animation 5).
It has multiple uses: from the joints of mechanical keys to the steering axis. In all uses where speeds are not relevant or where few turns are made, no problem is encountered.
As already mentioned, however, at high speeds there will be vibrations, because the instantaneous angular velocity of the driven shaft is not constant during a complete rotation. Certainly the two axes will complete a complete turn in the same time, but in completing the turn there will be accelerations and decelerations.
Without writing formulas, it is enough to know that the transmission ratio is directly related to the angle formed by the two axes (figure 4), the conductor and the duct (i.e. the motor axis and the propeller axis in our case).
In the first graph (figure 6) you can see the trend over time (x-axis abscissa) of the angular velocity (y-axis ordinate) of two joined and perfectly aligned axes. The straight line tells us that, over time, the angular velocity (revolutions per minute, rpm) does not change, it is constant.
In the second graph (figure 7) you can see the trend over time (x-axis abscissa) of the angular velocity (y-axis ordinate) of two joined axes with a certain angle between them. The sinusoidal curve tells us that over time the angular velocity (revolutions per minute, rpm) changes, it is not constant. In some points it is maximum and then drops to the minimum level in a continuous cycle. Therefore, the angular velocity is different from moment to moment. It is clear that the average of the sinusoid corresponds to the value of the straight line.

Let's get to the second question:

"What could be a valid solution and why?"

The answer is:

First possible solution.
"put a second joint"
but it's not enough.

I've noticed that many recommend it and limit themselves to this.
It is important that the angles of the engine and propeller axes with the central axis (alpha 1 and alpha 2) are equal but opposite. Look at figures 8 and 9.

In this way the sinusoids will cancel each other out.
See third graph, figure 10.
In fact two Cardan Joints constitute a constant velocity joint (I think it can be defined as a double joint or universal joint, see image 11)

Second solution:
Use a constant velocity [ homokinetic ] joint (see figure 12 and animation 13)
The constant velocity joint is the one used for example in the wheel axles of cars (see image 14).
In modeling I have never used them but I think many call them "dog bones" (with quite limited performance and resistance).
It is essentially based on mobile spherical rotations that guarantee constant instantaneous speed.
They need continuous lubrication (the real ones) and allow a lower phase angle than the Cardan joint.

Considerations and practical advice.

1. If we can, it is better to look for a perfect alignment between the propeller shaft and the engine shaft.
In this case, it is better to put a fixed connection (without a joint), because the fixed connection will let us know if we are not well aligned and we will be able to correct the positions (we will act more easily on the electric motor).
The Cardan joint, on the other hand, will adapt to small misalignments (deletive and useless) and we could finally have positioned the axes with a slight (not very evident and visible) misalignment.

2. If we are forced to use a joint, for example because the engine is in a high position and we do not want to tilt the propeller shaft (in short, we want to keep it as parallel as possible to the keel), it is not convenient to use a Cardan joint as in figure 15, for the reasons already mentioned.
We can put two joints in series so that the phase shift angles are identical (Image 16). We will thus create a homokinetic joint.
To create two equal alpha angles is not difficult, you do not need a goniometer. Knowing the geometry and the properties of the angles we will use a universal joint (double joint) and we will make sure that the axis of the motor and the axis of the propeller are parallel (again image 16).

If I was not clear please tell me.
Anyone can add their personal experience.

I still have to study the behavior and experiment with connections with semi-rigid cables in harmonic steel or other material.

forum/118944

🔗