Why when velocity is small ,it can't converged?

hotboy · December 15, 2014, 08:21

hello friends！I was simulating a 2D model 1.jpg.In this picture the part of green is a gas ,the parts of blue and red are two types of solid ,the temperature of the bottom is 700℃,the bottom of the gas is velocity-inlet,the temperatur is 700℃；the top of the solid is convection ，heat transfer coefficient is 10 ，the free stream temperature is 50℃，the top of the green is pressure-outlet ,the Gauge pressure is 0. If I set the velocity of the gas to 0.001,no matter what models I choose it can't converge ,but if I set the velocity to a Ibig one such as 1,wen choose the k-e ,it can converge!My problem is why when velociy is small no matter what models I choose,it can't converged?What can I do?Thank you !

edoan · December 16, 2014, 11:17

If by converged you mean the values of the residuals this makes sense. The residuals compare the current solution to the initial guess. perfect inital guess gives residual always equal to 1, so if your small velocity is close to the real solution the residual will not get too small. both cases give you the same solution but the residuals will display differently this is why its important to judge convergence not only with residuals. as a side note you can also select different options for residuals, absolute vs scaled etc.

December 15, 2014, 08:21	Why when velocity is small ,it can't converged?	#1
hotboy Member wanghuo Join Date: Aug 2014 Posts: 89 Rep Power: 11	hello friends！I was simulating a 2D model 1.jpg.In this picture the part of green is a gas ,the parts of blue and red are two types of solid ,the temperature of the bottom is 700℃,the bottom of the gas is velocity-inlet,the temperatur is 700℃；the top of the solid is convection ，heat transfer coefficient is 10 ，the free stream temperature is 50℃，the top of the green is pressure-outlet ,the Gauge pressure is 0. If I set the velocity of the gas to 0.001,no matter what models I choose it can't converge ,but if I set the velocity to a Ibig one such as 1,wen choose the k-e ,it can converge!My problem is why when velociy is small no matter what models I choose,it can't converged?What can I do?Thank you !

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Difficulty in calculating angular velocity of Savonius turbine simulation	alfaruk	CFX	14	March 17, 2017 06:08
rotation velocity of a water wheel and rotating domain	Ema40	FLUENT	0	November 12, 2014 10:50
Velocity inlet boundary condition for porous medium	Chander	CFX	3	March 11, 2012 21:18
Inlet Velocity in CFX	aeroman	CFX	12	August 6, 2009 18:42
Terrible Mistake In Fluid Dynamics History	Abhi	Main CFD Forum	12	July 8, 2002 09:11

December 16, 2014, 11:17		#2
edoan Member Ethan Doan Join Date: Oct 2012 Location: Canada Posts: 90 Rep Power: 13	If by converged you mean the values of the residuals this makes sense. The residuals compare the current solution to the initial guess. perfect inital guess gives residual always equal to 1, so if your small velocity is close to the real solution the residual will not get too small. both cases give you the same solution but the residuals will display differently this is why its important to judge convergence not only with residuals. as a side note you can also select different options for residuals, absolute vs scaled etc.