{{ post.title }} 글 편집 글 편집 (이전 에디터) Author {{ post.author.name }} Posted on

* The formulas related to friction are quite complex. This article explains intuitive concepts rather than exact formulas. (Therefore, the description in this article may be slightly different from the actual formula)

Please refer to the manual for details.

Threshold Velocity?

The coefficient of friction is actually expressed as a discrete function.

In the stationary state, the Static Friction Coefficient is applied. On the other hand, in the moving state, the Dynamic Friction Coefficient is used.

However, in numerical analysis, this friction is expressed as a continuous function as shown below.

• When the speed is in the [0, STV] section, the static friction coefficient is used.
• In the [STV, DTV] section, a friction coefficient that changes with speed is used.
• When the speed is greater than the DTV, the dynamic coefficient of friction is used.
  

(STV: Static Threshold Velocity / DTV: Dynamic Threshold Velocity)

In other words, Threshold Velocity is the reference velocity for determining the stationary state for friction and the state of motion.

How to determine the initial value for Threshold Velocity?

In fact, it is reasonable to use the dynamic friction coefficient when the body is moving. However, since it is often difficult to express a complete 'stationary state' in numerical analysis, a method that considers Threshold Velocity is used.

The Default values are STV = 1mm/s, DTV = 1.5mm/s, which are pretty small values.

It means that anything slower than 1mm/s can be judged as "stationary".

It is recommended to simulate using this value first.

What is the appropriate value for Threshold Velocity?

However, there are cases where the initial speed becomes quite large unintentionally in numerical analysis. (Even though it must be stationary)

In this case, a small coefficient of friction (dynamic coefficient of friction) is used by the initial speed, and as a result, a body often moves without stopping.

If a body has initial velocity and it need to 'stop', it is recommended to input a large STV.

In this case, you need to check the initial speed of the system first.

This 'initial velocity' may be the 'literal' initial velocity, or it can be the maximum value of the initial velocity for a short initial period of time.

Once checking the initial velocity (or the initial maximum velocity) of the system, It is recommended to set the STV to a value about three times larger than the initial velocity.

(Set DTV to 1.5 times of STV.)

For example, let's say that when we simulate a system, we know that the body (or system) must be stationary initially. (empirical or intuitively)

However, it is possible that the initial velocity of about 30mm/s can occur due to numerical error or initial contact force or interference present in the model.

Then you need to set this to triple value, 90mm/s (=30x3) to STV. (DTV must be 1.5 times, that is, 135mm/s (=90x1.5) )

Here, 90mm/s is a speed that cannot be considered as a stationary state from common sense.

But for the purpose of simulation, we can say that it is well suited to reproduce the stationary state by considering the initial velocity due to a numerical error.

Of course, it is also possible to reduce the initial speed by modifying the model carefully. But in practice, it is often used to use a value three times larger or even more than that for the initial speed as STV when it can be sufficiently guaranteed to be 'stationary'.

In addition, please refer to the paper on the friction of contact below.

https://link.springer.com/article/10.1007/s12206-011-0504-y