3-D nozzle with fluidic injection

mjkapp · November 23, 2014, 15:22

Hi all,

I am trying to model a 3-dimensional nozzle with injectors. Previously I was doing steady RANS computations using Wind-US, but have now been asked to switch to CFX. The problem I am having is that the injectors are taking the fluid from the freestream. I would like the flow from the injectors to be generated separately from the rest of the domains. The grid was created using gridgen and the injectors are completely separate blocks from the freestream. I have specified them as separate domains in cfx-pre as well. In Wind-US the freestream would pass through the injectors since they were different blocks. The flow was generated at the injector inlet by specifying a total pressure and temperature as boundary conditions. I applied the same boundary conditions in cfx and they are met, but the flow is not created separately. Is there anyway to treat the flow between these blocks as separate in CFX? I have attached an image showing the streamlines through the injectors.

Thank you in advance.

ghorrocks · November 23, 2014, 16:54

Can you post an image describing what boundary conditions you have applied, what that gave you on the other software and what it gives you in CFX.

mjkapp · November 23, 2014, 17:22

Thank you for the quick reply. I have attached two images of where the boundary conditions are being implemented.

The outer domains are listed as opening boundary conditions. The freestream Mach number is set to 0.01 and the opening temperature is set to be 491.7 R. For the outlet, I also specified an opening boundary condition with a relative pressure of 0 Pa. (The reference pressure is set to 101008 Pa)
For the nozzle itself, I have specified an inlet for the plenum, which is driven by a total pressure condition and total temperature. Before the nozzle begins to converge, the walls are inviscid. They become viscous walls once the plenum begins to converge until the flow exits the nozzle. The exterior walls of the nozzle exposed to the freestream are inviscid.
The walls of the injectors are also viscous walls. The injectors have inlet boundary conditions, which are driven by a total pressure and temperature, like the plenum.

ghorrocks · November 23, 2014, 17:43

Can you please label your image so I know what the nozzle inlet plenum is, which walls as free slip and which are not and so on.

mjkapp · November 23, 2014, 18:05

I have reattached the images labeled with the boundary conditions. Thank you.

ghorrocks · November 23, 2014, 18:09

OK, thanks. I have a better idea of what you are doing now.

If you want the injectors to be separate from the outer flow then do not connect them to the outer flow. Block off the end of the injector and no flow will go through.

mjkapp · November 23, 2014, 18:23

I have encountered issues with the injector inlet putting up temporary walls due to recirculating flow. Since the flow is supposed to enter through the top of the injector, I was surprised that this was occurring, even though I expect a slight recirculation at the change in area. I have attached an image of streamlines through the injector. I only had ran the case for 250 iterations just to get a gauge as to what the flow was trying to do. You can see that the flow is very peculiar. I'm not sure what cfx is attempting to do here with the domains and if this would still occur if I cap the injector.

ghorrocks · November 24, 2014, 04:20

Now I am really confused about what you are trying to do with the injectors. Can you explain what the injectors are physically and what drives the flow in them? Is there anything in them to add energy to the flow, like an impeller, combustor or a pump?

mjkapp · November 25, 2014, 14:35

I am trying to verify work done by our experimental research team. They set a pressure ratio for the plenum and injectors. For the experiment, the plenum and injector receive the air from a compressor. The nozzle is the only part of the engine that is being modeled. The total pressure is what drives the flow. As said earlier, this is based on the desired pressure ratios for the plenum and injectors. So, I would like the injectors to generate their own flow, like the plenum does, without using the freestream. I am going to try to cap the injector like you previously suggested to see if this rectifies the problem.

ghorrocks · November 25, 2014, 17:33

If the injectors are connected to a compressor with its own air flow then you should model them like this - so put a cap in it and an inlet boundary to provide the correct injector flow.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Convergent-Divergent Nozzle with MATLAB	Obad	Main CFD Forum	8	June 2, 2020 14:10
Meshing an injection nozzle	sontke	AVL FIRE	0	October 17, 2014 09:31
injection problem	Mark New	FLUENT	0	August 4, 2013 01:30
point injection Vs Nozzle injection clarify	drsrinivasan	Main CFD Forum	1	December 18, 2012 12:35
compressible flow in a counterflow nozzle	d.vamsidhar	FLUENT	0	November 24, 2005 01:45

November 23, 2014, 16:54		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Can you post an image describing what boundary conditions you have applied, what that gave you on the other software and what it gives you in CFX.

November 23, 2014, 17:43		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Can you please label your image so I know what the nozzle inlet plenum is, which walls as free slip and which are not and so on.

November 23, 2014, 18:09		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	OK, thanks. I have a better idea of what you are doing now. If you want the injectors to be separate from the outer flow then do not connect them to the outer flow. Block off the end of the injector and no flow will go through.

November 24, 2014, 04:20		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Now I am really confused about what you are trying to do with the injectors. Can you explain what the injectors are physically and what drives the flow in them? Is there anything in them to add energy to the flow, like an impeller, combustor or a pump?

November 25, 2014, 14:35		#9
mjkapp New Member Join Date: Nov 2013 Posts: 5 Rep Power: 12	I am trying to verify work done by our experimental research team. They set a pressure ratio for the plenum and injectors. For the experiment, the plenum and injector receive the air from a compressor. The nozzle is the only part of the engine that is being modeled. The total pressure is what drives the flow. As said earlier, this is based on the desired pressure ratios for the plenum and injectors. So, I would like the injectors to generate their own flow, like the plenum does, without using the freestream. I am going to try to cap the injector like you previously suggested to see if this rectifies the problem.

November 25, 2014, 17:33		#10
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	If the injectors are connected to a compressor with its own air flow then you should model them like this - so put a cap in it and an inlet boundary to provide the correct injector flow.