IES-VE User Forum

Hi,

I'm not sure this can be done...what I've seen in the past:

Microflo is a steady state solver therefore it cannot model the transient extraction of smoke or other contaminant.

We can offer this as a Consultancy service but would be done using a 3rd party CFD pakcage.

Phil

There is a way to do this if a steady state solution accurately models the physics.

For example, smoke production starts at t=0secs and stops at t=10mins and you are interested in how the smoke concentration climbs and then falls after the smoke source stops at t=10mins then only a transient CFD solver can give you this info. It will tell you the smoke concentration at t=1sec, t=10sec, t=10mins and t=15mins etc etc.

A steady state solution is basically assuming that the smoke source is constant, to t=infinity. Obviously that is a case that will never happen but steady state solutions can be applied to dynamic situations.

For example, smoke production starts at t=0secs and stays constant at a certain production rate (mass/time) until it stops at t=20mins. It will probably reach a steady state solution quite quickly, say 10mins. That would mean that the steady state solution would be accurate for a dynamic situation for t=10mins to t=20mins.

You could do some simple hand calcs that assumes the air is full mixed to estimate the time taken to reach the steady state smoke condition in the real non-fully mixed air case.

In this case the steady state solution not only provides you with the answer to the smoke concentration from t=10mins to t=20mins it also gives you the maximum smoke concentration that occurs. If this info is good enough to answer the questions you are asking of the CFD model then the steady state assumption is a valid one.

Unfortunately, Microflo doesn't include a scalar called "smoke" to track smoke through a model. It does provide passive scallars for CO2, CO and water vapour. You could use one of these as a proxy for smoke.

The passive scalar model in Microflo uses a diffusion coefficient for each of the passive scalars, e.g. to model molecular diffusion of CO2 in air. If you use one of these passive scalars to represent smoke then the diffusion coefficient won't be correct. This won't introduce a significant error in the results as the transport of smoke should be convection rather than diffusion driven.