Hi,

does anyone know what is the difference between these two versions of FLUENT? In terms of the capacity and operation?

I am asking this because I received a suggestion to use the 64 bit to overcome where my DPM particle mass suddenly rounds off to the 0%volatile.

Regards, Woo

Wai - did you get that bit of generic code I emailed you a few days ago? The 64 bit vs. 32 bit sounds very unlikely. What do you think would be the precision of your %volatile with a 32 bit system compared to the round off error? Perhaps a factor of 1x10^20? If needed, I'm sure you can check it.

Hi Allan,

I am afraid I did no receive your email with the generic code. Probably something wrong with my mailbox. How about giving me your email? I will then send over a mail from which you can reply together with the generic code. Maybe the mail can get through this way.

Precision: At the latter stages of the vaporization (~4%volatile from an initial 80%), the changes in mass is around 1x10e-16 kg. How do I check on the round off error for a 32 bit system?

Regards, Woo

32 bit Fluent version on 32bit OS and hardware can only access 2^32bytes. 64 bit version on 64 bit OS and hardware can access 2^64 bytes so much more RAM can be accessed.

I'll try sending the info. one more time to both your gmail and hotmail accounts.

Why don't you just try some code like this:

a = 1.1e-16 b = 1.0e-16 If a > b then do this

else do this

Hi Allan,

Got the info in my mailbox now. Thank you.

Will try out the 'if' codes. Hopefully it will solve the problem. The vaporization UDF is really good. Thanks.

Hi Allan,

I have tried out the if-else structure. This prevented the problem. However, this method does not allow us to capture the latter stages of vaporization which is of particular interest.

There seems to be some problem in the code that you have sent me, where certain lines / macros can't be compiled. I am using Microsoft Visual C++ Express.

Based on your generic code, I am now trying to write a simple vaporization law to be hooked to the custom law 2. However, when hooked to the custom law 2, the solver only returns the (p, ci) pointer values. However, we would need the c pointer values in order to access and update the cell properties.

Any idea on how to obtain the c pointer values?

Regards, Woo

Wai,

The code was just an example. Are you trying to access cell variables based on the particle position? Or trying to update cell properties based on vaporization? I'm guessing you have looked in the dpm.h and other header files.

Hi Allen,

Actually I need to update cell properties based on vaporization and also to access cell properties based particle position. The latter is to determine the mass adn heat transfer coefficient.

From the example, the 'heatmassupdate' and 'updatetemperature' macro was used. Is this the method to update cell properties? the 'updatetemperature' macro causes error when compiling.

To access cell properties, the example uses 'c->properties'. Would this be the way to access cell properties based on particle position? Hooking the UDF to the custom law 2 only returns p value. Does the c value correspond to the particle?

I have tried looking at hte dpm.h, but can't understand how things are connected.

Woo

There are macros to link the particle position to the cell, which can be found in examples.