Learning Forum

[Jeroen]

Hi,

 

I am using a DEFINE_DPM_HEAT_MASS user defined function from which I calculate a conversion from particle component A to component B. In the UDF the mass source terms for both of these components are calculated and returned to Fluent.

In my case however there is a difference between the dissipated mass of component A and the generated mass of component B, which thus should result in a lower overall particle mass.

Unfortunately, it seems that Fluent does not automatically update the total particle mass TP_MASS(tp) before the next DPM iteration. Is there a way to assign this change in total particle mass directly from the HEAT_MASS UDF?

 

Jeroen

[Rob]

Have you tried this with a single particle to test? If so does it update on the following DPM iteration?

[DrAmine]

Not really understood the problem

[Jeroen]

Yes I am currently running a single particle to test my UDF, and the total particle mass is not updated after each iteration of my UDF.

Let me try to clarify my problem;

I have a particle with initial mass M_0 (upon entry of my domain). This particle reacts and component A decreases in mass and component B increases in mass. The source terms for these components are calculated and given back to fluent via the dydt[1] and dydt[2] variable (unit kg/s). There is however a difference between these two source terms, which should decrease the total particle mass by:

delta_mass = (-dydt[1]+dydt[2]) * particle_timestep

The total particle mass at the end of my UDF should thus equal:

M - delta_mass

This is however not the case, since Fluent does not update the total particle mass M after running my UDF. It only seems to update the component mass fractions of A and B based on the initial particle mass M_0 (upon entry of my domain).

My question is: how can I make Fluent update the total particle mass M after running my UDF? Can this be done via the HEAT_MASS UDF?

 

Jeroen

[DrAmine]

Are you running steady-state particle tracking? There the mass flow of particles is kept constant.

[DrAmine]

TP_MASS should be adapted: Are you comparing it to TP_MASS0?

[Jeroen]

Yes I am running steady-state particle tracking, in combination with a steady-state continuous phase. Setting the particle tracking to unsteady also does not affect the individual particle mass. The mass flow of particles at my inlet is constant.

[Jeroen]

For my UDF the TP_MASS should be updated with respect to TP_MASS0 indeed. Currently, this does not happen and TP_MASS0 is also equal to TP_INIT_MASS at this moment, which is not desired.

[DrAmine]

Just checked now using the example from the manual in dummy case: the example is working as it should, Particle mass is reducing from injection to final fate, source terms are allocated and assigned. Please start from the example in the manual and use a dummy test case first. Tested in 2020R1

[Jeroen]

My source terms for the particle components are allocated since I can see the mass fractions are changing. My overall particle mass does however not change, even though there is a difference between the source terms for the particle components.

I get the same result when using the UDF in the manual (I changed the assigned values for dydt manually). What dpm settings do you use for your dummy test case?

[DrAmine]

I am just using default settings say two-way coupling. I did not do anything else.

[Jeroen]

I simply cannot get it to work in 2019R3, even when using the UDF as presented in the manual. Any other suggestions on how to solve my problem?

[DrAmine]

I tested with 2020R1. Please try there with a simple case and report.

Just plot the particle trajectories and color by particle mass.