Jump to content
RockWare Support Forum
MF-Bristol Carbonates

Pinch cell pressure gradient inaccuracies

Recommended Posts

There appears to be a worrying problem arising from the use of pinch cells within Petrasim 5. As our own in-house development team had a bit of trouble getting pinch cells to produce consistent results, we were very excited to see that Petrasim allows pinched cells to be included within TOUGH2 models. However, when we build a simple pinch cell test model, it appears that the pressure gradients accross the domain are distorted by the presence of pinch cells.

Our test model:

3m x 3m domain (x,z)

8 cells - see attached page for grid geometry and results.

Boundary conditions - both sides of model are infinite volume (1E40 in vol.factor box).

Left side inf. vol. cells are increased in pressure.

Right side inf vol pressure is fixed.

The model thus simply calculates the pressure distribution accross the pinch cell and cells above and below.

The model is also isothermal - so no convective control on flow.

What we were hoping for is a linear decrease in pressure accross the domain from left to right.

Instead, we get a very confusing pressure distribution and where we get a sensible pressure gradient, we get flow in the wrong direction...

We are not sure where we are going wrong - hopefully it is us that is causing the problem, and we can continue to be excited with using pinch cells! Any advice would be most appreciated, we include an input file for TOUGH and a brief, not so neat spread sheet that shows the upsetting results.


Problematic pinch cells.htm


Share this post

Link to post
Share on other sites

We are having a hard time viewing your files. Would it be possible for you to post a zip file containing your SIM files, as well as any other files that you would like us to see? If you do not feel comfortable posting your model on the forum, you are welcome to send this to me at [email protected]



Share this post

Link to post
Share on other sites

Can you please send your PetraSim *.sim file and your Excel workbook to [email protected] for review? The attached *.htm document does not contain your entire Excel data set and appears to refer to other files that have not been attached to this thread.

It is quite possible that the pinched cell is causing some convergence or accuracy problems. We have not taken any special steps in PetraSim to enforce a well-behaved mesh. We have simply tried to provide tools to create a variety of mesh types as requested by our users. Proper TOUGH2 simulations will still have to meet the mesh criteria specified in the TOUGH2 users guide. PetraSim allows grids to be shifted in the vertical direction, which can lead to non-conforming grids. Some modest shifting will introduce small but acceptable errors, but larger distortions can cause problems. Generally, if you have widely varying layer thickness, it is best to use a regular grid in PetraSim that follows the layers with a "stair-step" approximation.

Share this post

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Create New...