Learning Forum

[CHIEH TING WANG]

why don't need to set any property ignite point or temperature setting can make combustion in non-premixed combustion ，is it follow for eddy dissapation model?

[Rob]

Some of the models require you to patch in a high temperature region to trigger the reaction, much as you would in reality with a spark plug. Other models require a small amount of product species to kick off the reaction. It's typically covered in the User Guide, and there are some examples/tutorials in the online Fluent Help system. 

[CHIEH TING WANG]

So non-premixed combustion does not follow the eddy dissapation model?

[Rob]

Pass, the User's Guide will cover most of the set up and Theory Guide the maths. My knowledge of combustion chemistry is OK, my modelling knowledge is limited.

[CHIEH TING WANG]

So i try to simulate micro gas turbine wit burn and the species model i chose 【non-premixed combustion chemical equalibrium 】；mass flow inlet has 0.473 kg/s of oxidizer and the temperature was 300k ，the fuel inlet was 0.0096 kg/s mixture fraction with CH4 mass fraction set at 1 ，also temperature was set in 300k。

after the calculated the temperature result could appeared over 2000k? is there any property burnt point can be control?

[CHIEH TING WANG]

So i was simulate a micro gas turbine the fuel injection was set as mass flow inlet :0.0096 kg/s；mixture fraction set 1；total temperature 300k.And the air was set as mass flow 0.473 kg/s of oxidizer ；total temperature 430k

the speices model i chose [non-premixed combustion chemical equalibrium],after caculation the result temperarure appear over 2000k.so why the temperature can be such high?is there any property burning point can be set?

[CHIEH TING WANG]

By the way the mixture fraction i set 1mass fraction of CH4. Bless!

[Rob]

How well converged was the system, how well resolved is the flame? Pictures will help. 

[CHIEH TING WANG]

Hi Rob!

Is the converged mean the iterations during calculation?

By the way for the picture is the result of temperature:

 

[Rob]

Yes, have a look at the residual plot, and the values printed back to the TUI. 

Please replot the above with node values off.  Shouldn't the hot spots be on the flame? With the above you're going to burn the can/annular unit out very quickly. 

[CHIEH TING WANG]

Hi!

here is the iteration during the calculation

And i use 3D model in this case the picture is the gas turbine linner with the color map 

[CHIEH TING WANG]

I use probe to measure the temperaure(the red plot in the picture),the temperature is over 1300K

[Rob]

Convergence isn't great so it's likely the chemistry isn't well resolved. Also please plot the plane contour without node values as asked - I need to get an idea about the mesh. 

Finally, we're running 2023R1 now, which is approximately R24.1 in old versioning. R17 is ancient, and updating will give you much faster model runs and significantly improved solver tools. 

[CHIEH TING WANG]

Hi!

This is the mesh picture in this case,if there any information need to asked i will provide.

May i asked is there any reason that i set any law temperature in the boundry condition but why the result can produce high temperature when it set combustion model?

[Rob]

Your mesh isn't fine enough for the physics: you need significant refinement for the combustion region. Also, we generally recommend hex or poly (not available in R17) elements for accuracy: combustion calculations tend to be very stiff (reaction rate is tightly coupled with temperaure and vice versa). 

I don't know what you mean by "law". The combustion models may have limiters but they're for extreme cases and shouldn't be used "just because". Fix the mesh before messing with numerical stabilisation functions. 

[CHIEH TING WANG]

Hi!

the law temperature is mean i set 300K in any boundary condition but why the temperature can happened over 1300K in the result,while i use the combustion model

 

[Rob]

Err, isn't the point of combustion that it gets hot? 

[CHIEH TING WANG]

Yes it is .

but why it can be hot? does there any burning point need to set ?

When i was setting non-premixed combustion model ,i used 'chemical equalibrium' .the dialog box only need to set mass fraction of mixture and caculate the pdf .then what is the reason why i using pdf can get hot? 

[Rob]

The pdf table is what's causing the chemistry. Please read through the various combustion models to understand what's going on. 

[CHIEH TING WANG]

then may i asked why the pdf can causing the chemistry happened?is there any theory or introduction can follow?

[Rob]

Yes, read the User's Guide and Theory Manual. It may also be covered in the tutorials in Help, or there may be an example in the videos. 

[CHIEH TING WANG]

HI!

I have read the User Guide and Theory Manual.May i ask was the pdf chemistry is related to mixture fraction species?

If it was related then where can i see any temperature control in pdf? 

  

[Rob]

Why would you need temperature control? The whole point of the models is to see what you get from the geometry and boundary settings? 

[CHIEH TING WANG]

I apologize that my English is not good，so i did not ask well.

i just want to know if the pdf related to the mixture fraction ，then does the chemistry happened is because the mixture fraction?  

 

[Rob]

The pdf table works with the mixture fraction to work out the combustion effects. It's a way around modelling the whole thing using Arrhenius type chemistry. It's a good method, but whilst I understand combustion chemistry I leave combustion modelling to colleagues. 

[CHIEH TING WANG]

ok ! thanks with your reply

so I do this case without setting any ignition but still can cause the chemistry happened the reason is the pdf and the pdf can make it happened was followed by  Arrhenius type chemistry?

[Rob]

The pdf table handles the reactions. It's one of several options in Fluent.