[CeesH]

Hi all,

I'm working on an UDF for a population balance model, and need to include the number of bubbles in binsize i.

This can of course be calculated using the bin volume fraction, C_PB_DISCI(c,t,i). Now my question is, how do I get i.

As far as I know, the solver passes d_1 and d_2, which are the diameter of bins i and j. Also, thread and thread_2 are passed, but for a discrete or qmom simulation, these are the same. So, how can I get the correct i (and j), that match the given diameters?

Any ideas would be awesome!

Best regards,
Cees

[CeesH]

Unfortunately I have not yet solved the issue; any advice is highly welcome!

Best regards,
Cees

[CeesH]

Still open for suggestions here

[hwet]

what do you mean they are the same?

Will they all not be in the same mixture level thread anyways?

Have you tried using hard integers? '0' for bin-0, '1' for bin-1 etc?

[CeesH]

Hi hwet,

Indeed they are in the same mixture thread, which is why I can't use the thread values to distinguish between the size class.

As far as I understand the population balance approach works, the solver passes 2 bubble classes to the UDF, which calculates the coalescence rate between those classes, and feeds that back to the solver. Then the process repeats for the next 2 bubble classes, until all combinations were explored.
This mechanism is confirmed by printing d_1 and d_2 to the console for 1 gridcell, I get all possible combinations for d_1 and d_2 in series.

So the thing is, I want to know for that given d_1 and d_2 how many bubbles there are in that particular class. So hardcoding 0,1,.. directly doesn't work; I can of course set some if loop that contains (in pseudocode)

if d_1 == size, i = associated class

But this seems like a tedious exercise to me (not so much in programming, but having to burn through that if-loop every time seems time consuming)
So I was hoping there was a more direct way to get the bubble class associated with the passed diameters.

[nhendre]

Originally Posted by CeesH

Hi hwet,

Indeed they are in the same mixture thread, which is why I can't use the thread values to distinguish between the size class.

As far as I understand the population balance approach works, the solver passes 2 bubble classes to the UDF, which calculates the coalescence rate between those classes, and feeds that back to the solver. Then the process repeats for the next 2 bubble classes, until all combinations were explored.
This mechanism is confirmed by printing d_1 and d_2 to the console for 1 gridcell, I get all possible combinations for d_1 and d_2 in series.

So the thing is, I want to know for that given d_1 and d_2 how many bubbles there are in that particular class. So hardcoding 0,1,.. directly doesn't work; I can of course set some if loop that contains (in pseudocode)

if d_1 == size, i = associated class

But this seems like a tedious exercise to me (not so much in programming, but having to burn through that if-loop every time seems time consuming)
So I was hoping there was a more direct way to get the bubble class associated with the passed diameters.

Dear Cees,
I tried with 'if loop' but it is showing some parse error. Did you solve your case with this udf? I am also trying to correct my udf (i have attached .c file).
Thanks.