[mohsinmkhan]

Can anybody tell me how to model hydrogen as gaseous direct injection as we cannot use spray modelling ? what is extra approach needed for this and what are the common mistakes happen during setting up the case like "do we need to provide embedding for each nozzle hole? and what is the significance of that"

[ksrivast]

Hello Mohsin,

Gaseous fuel inlets can be modeled using a simple INFLOW boundary. It would be recommended to include part of the gas injector/duct in the geometry.

Set the Species Boundary Condition for the INFLOW as pure hydrogen. Injection timing and rate shapes can be controlled through events by placing the injector/duct tip into a separate region, or by providing a velocity profile.

Fixed embedding and AMR is highly recommended to capture the gaseous plume. CYLINDER shape embedding can be employed. Construct a conical shape along each spray plume. This would be similar to what you would have had using a NOZZLE/INJECTOR embedding.

Hope this helps.

Sincerely,

[mohsinmkhan]

THANKS SIR, one more doubt m having regarding events, do i have to set only OPEN event between inflow and cylinder region or i have to set both open and close events ??

[ksrivast]

Hello Mohsin,

At the start of injection, create an OPEN event. At the end of injection, create a CLOSED event. Make this CYCLIC.

Sincerely,

[mohsinmkhan]

Originally Posted by ksrivast

Hello Mohsin,

At the start of injection, create an OPEN event. At the end of injection, create a CLOSED event. Make this CYCLIC.

Sincerely,

same procedure i have followed sir. But simulation ran for only few degrees and then stopped. so where m getting wrong i m not getting?

[ksrivast]

Please look into your logfile and describe the message displayed at the crash.

[mohsinmkhan]

Originally Posted by ksrivast

Please look into your logfile and describe the message displayed at the crash.

sir can i send you my case setup file through mail?

[ksrivast]

Sure. You can send your setup to support@convergecfd.com . Please use your official email for all correspondence with Convergent Science. Please mention the issues you were facing and add my name, and the cfd-online thread, as reference.

Sincerely,