I am trying to design a heat removal system for a specific application. As the heat flux is very high, the only way to reduce the temperature is by increasing fluid(water) velocity. I am worried about the erosion of the channels. The material is copper and SS304. The fluid is DM water. Is there a model or empirical relation by which I can predict the erosion of the channels over time? Is there any hard limit for water velocity in cooling channels?
Yes, there are models and equations to predict the erosion of stainless steel (SS) pipes due to flowing water. Erosion in pipes can be caused by several factors, including the velocity of the water, the presence of particulates, the chemical composition of the water, and the properties of the pipe material.
One common approach is to use empirical or semi-empirical models that are based on experimental data. These models often take into account factors such as flow velocity, particle size, concentration, and material properties. Some well-known models include:
### 1. DVGW Model:
The German Technical and Scientific Association for Gas and Water (DVGW) developed a model to predict erosion in pipelines. This model considers factors such as flow velocity, pipe diameter, and material properties.
### 2. API RP 14E Model:
The American Petroleum Institute Recommended Practice 14E provides guidelines for the design and operation of piping systems to minimize erosion. It includes an empirical formula to estimate the maximum allowable velocity to avoid excessive erosion:
\[ v_{\text{max}} = \frac{C}{\sqrt{\rho}} \]
where:
- \( v_{\text{max}} \) is the maximum allowable velocity (ft/s),
- \( C \) is an empirical constant (usually taken as 100 for carbon steel and adjusted for other materials),
- \( \rho \) is the fluid density (lb/ft³).
### 3. Salama and Venkatesh Model:
This model is specifically developed for erosion prediction in pipelines and considers various parameters, including fluid velocity, particle size, and pipe material. The erosion rate (E) can be estimated using the following equation:
\[ E = k \cdot C_p \cdot V^n \cdot d^m \]
where:
- \( E \) is the erosion rate,
- \( k \) is a material constant,
- \( C_p \) is the concentration of particles,
- \( V \) is the flow velocity,
- \( d \) is the particle diameter,
- \( n \) and \( m \) are empirical constants.
### 4. CFD (Computational Fluid Dynamics) Models:
For more detailed and specific predictions, CFD models can be used. These models simulate the fluid flow and particle interactions within the pipe, providing a detailed understanding of erosion patterns. They take into account the pipe geometry, flow characteristics, and material properties.
### Factors Influencing Erosion:
To use these models effectively, it's essential to consider the following factors:
- Fluid Velocity: Higher velocities generally increase erosion rates.
- Particle Size and Concentration: Larger and more concentrated particles lead to higher erosion.
- Pipe Material: Different materials have varying resistance to erosion.
- Flow Regime: Turbulent flow typically causes more erosion than laminar flow.
### Example Calculation:
If you have specific data for your system, such as the flow velocity, particle concentration, and material properties, you can use one of these models to estimate the erosion rate. Here’s a simplified example using the Salama and Venkatesh model:
Assume the following parameters:
- \( k = 2.5 \times 10^{-4} \)
- \( C_p = 0.01 \) (particle concentration in weight fraction)
- \( V = 10 \) m/s (flow velocity)
- \( d = 100 \) microns (particle diameter)
- \( n = 2.6 \)
- \( m = 0.3 \)
The erosion rate can be calculated as:
\[ E = 2.5 \times 10^{-4} \cdot 0.01 \cdot 10^{2.6} \cdot 100^{0.3} \]
You can plug in the numbers to get the erosion rate.
### Conclusion:
To accurately predict erosion in SS pipes due to flowing water, it’s best to combine empirical models with experimental data specific to your system. Additionally, for complex systems, CFD simulations provide a detailed and precise prediction of erosion patterns@Hardik Mistry
Yes, there are models and equations to predict the erosion of stainless steel (SS) pipes due to flowing water. Erosion in pipes can be caused by several factors, including the velocity of the water, the presence of particulates, the chemical composition of the water, and the properties of the pipe material.
One common approach is to use empirical or semi-empirical models that are based on experimental data. These models often take into account factors such as flow velocity, particle size, concentration, and material properties. Some well-known models include:
### 1. DVGW Model:
The German Technical and Scientific Association for Gas and Water (DVGW) developed a model to predict erosion in pipelines. This model considers factors such as flow velocity, pipe diameter, and material properties.
### 2. API RP 14E Model:
The American Petroleum Institute Recommended Practice 14E provides guidelines for the design and operation of piping systems to minimize erosion. It includes an empirical formula to estimate the maximum allowable velocity to avoid excessive erosion:
\[ v_{\text{max}} = \frac{C}{\sqrt{\rho}} \]
where:
- \( v_{\text{max}} \) is the maximum allowable velocity (ft/s),
- \( C \) is an empirical constant (usually taken as 100 for carbon steel and adjusted for other materials),
- \( \rho \) is the fluid density (lb/ft³).
### 3. Salama and Venkatesh Model:
This model is specifically developed for erosion prediction in pipelines and considers various parameters, including fluid velocity, particle size, and pipe material. The erosion rate (E) can be estimated using the following equation:
\[ E = k \cdot C_p \cdot V^n \cdot d^m \]
where:
- \( E \) is the erosion rate,
- \( k \) is a material constant,
- \( C_p \) is the concentration of particles,
- \( V \) is the flow velocity,
- \( d \) is the particle diameter,
- \( n \) and \( m \) are empirical constants.
### 4. CFD (Computational Fluid Dynamics) Models:
For more detailed and specific predictions, CFD models can be used. These models simulate the fluid flow and particle interactions within the pipe, providing a detailed understanding of erosion patterns. They take into account the pipe geometry, flow characteristics, and material properties.
### Factors Influencing Erosion:
To use these models effectively, it's essential to consider the following factors:
- Fluid Velocity: Higher velocities generally increase erosion rates.
- Particle Size and Concentration: Larger and more concentrated particles lead to higher erosion.
- Pipe Material: Different materials have varying resistance to erosion.
- Flow Regime: Turbulent flow typically causes more erosion than laminar flow.
### Example Calculation:
If you have specific data for your system, such as the flow velocity, particle concentration, and material properties, you can use one of these models to estimate the erosion rate. Here’s a simplified example using the Salama and Venkatesh model:
Assume the following parameters:
- \( k = 2.5 \times 10^{-4} \)
- \( C_p = 0.01 \) (particle concentration in weight fraction)
- \( V = 10 \) m/s (flow velocity)
- \( d = 100 \) microns (particle diameter)
- \( n = 2.6 \)
- \( m = 0.3 \)
The erosion rate can be calculated as:
\[ E = 2.5 \times 10^{-4} \cdot 0.01 \cdot 10^{2.6} \cdot 100^{0.3} \]
You can plug in the numbers to get the erosion rate.
### Conclusion:
To accurately predict erosion in SS pipes due to flowing water, it’s best to combine empirical models with experimental data specific to your system. Additionally, for complex systems, CFD simulations provide a detailed and precise prediction of erosion patterns.
A common problem with DM water is dissolved gases, unless the DM system has a de-gasser. Dissolved CO2 forms a weak acid that increases corrosion, especially as velocity is increased. I would be more concerned about the copper sections that you mention, as that is more liable to attack by the dissolved gases. In heated systems an auto-air-vent at high points can remove the gases effectively. Also, it is worth noting that bends and fittings are more prone to erosive damage than straight runs.
Thanks, Farnaz Hosseini for the detailed answer and Jack Broughton for valuable insights.