Why MeanCalc?

Running DES simulations can often mean striking a balance between accuracy and cost. Without the right tools, simulations risk overrunning and consuming unnecessary compute resources. MeanCalc by Upstream CFD, integrated into bramble, solves this problem by automatically detecting convergence, extending averaging windows when needed, and stopping runs once sufficient accuracy is reached.

The result? Improved accuracy at lower cost, no more wasted core-hours on overlong jobs.

In this blog, we’ll walk through how to activate and set up a MeanCalc workflow inside your bramble account, including updating your Template Case, configuring simulation settings, and making the most of MeanCalc’s presets for fast or precise results.

Meancalc can help elevate your DES simulations by:

upstream cfd
Dynamically detecting the end of the initial transient

 

Perform averaging with as much of your data as possible

 

Ensure convergence is reached with the specified tolerance level

 

number 4
Avoid unnecessary cluster usage by stopping the simulation once the solution is converged.

 

Activating MeanCalc

If you are already running transient DDES simulations in bramble, then converting to a MeanCalc workflow is straightforward, only requiring a change of methodology, workflow and an update to the simulation settings in the Template Case.

To begin with, update your new simulation’s CFD Setup to use the MeanCalc methodology and workflow.  This menu is found in the ‘CFD Setup’ tab in the Detailed Run View (where you launch simulations from).

If the methodology or workflow options are not available then please contact support@bramblecfd.com and we’ll get it activated for you.

 

 

Select MeanCalc methodology and workflow

Updating the Template Case

Next make a copy of your existing template case.  We’re going to update the settings to make it available for use with the MeanCalc workflow.

Template Cases can be linked to specific workflows, so for example, a DDES setup isn’t used in a RANS workflow.  And so, in order to use our new MeanCalc template, we need to link it to the MeanCalc workflow.

Open the Template Case Editor, and in the ‘Rename Template Case’ tab, edit the ‘Methodology Group’ to be OpenFOAM and select the ‘TSFoam v1806F DES + MeanCalc’ workflow.

Select methodology group and workflow

We can now update transient simulation settings (solve time, time-step) and MeanCalc specific settings.  Staying in the Template Case Editor, switch to the Solutions tab. The Solution Stage should be the same as your normal DES settings (i.e. “Transient”, number of RANS iterations to initialise the DES solve, the DES solution time and the time step.

Note that as MeanCalc will stop the simulation when it is sufficiently converged, the DES Solution Time is now your maximum simulation time.  Set this generously to avoid the simulation stopping before sufficient accuracy is reached.

DES MeanCalc Simulation Settings

As part of Meancalc’s initial transient detection, it will attempt to maximise the use of the input signal but will always have more than the Min Sample Time. In other words, the minimum averaging window.  In a standard DES workflow, we would have set a fixed averaging window (e.g. 2000 time-steps), MeanCalc will increase the averaging window if it finds that more data needs to be capture for sufficient accuracy.

Two ‘Presets’ are available, which will override any user input on Min Sample Time and Stopping Criteria’s Relative Target Accuracy in favour of recommended settings: Fast and Precise.  If you prefer to define your own settings, please use the ‘Custom’ preset.

MeanCalc Min Sample Time and Preset

Finally, we need to set the convergence tolerances that MeanCalc should use when determining if it should stop the simulation.  The Stopping Criteria table sets which force directions (in the ‘all’ force group) will be monitored for convergence and the relative target accuracy level.

Select ‘New’ to add a stopping criteria for another force direction.

MeanCalc stopping criteria

Finally, save your template case and ensure it is selected in the CFD Setup of your new simulation.

simulation pricing, wallet

How much does it cost?

The integration adds a minimal cost of just £0.003 per core-hour, bringing the total to £0.065 per core-hour when using standard public cloud, slightly above the base bramble rate of £0.062 per core-hour. For most users, the savings from shorter, more accurate simulations far outweigh this small extra cost, while helping you achieve reliable CFD simulations every time.

By switching to a MeanCalc workflow in bramble, you can achieve more reliable CFD simulations while reducing costs. MeanCalc ensures you capture enough data for accuracy, automatically stops runs when convergence is reached, and avoids the waste of overrunning jobs.

Setting this up in your bramble account only requires a few workflow and template updates after that, you’ll be running faster, smarter, and more cost-efficient DES simulations every time.

Contact us

See how bramble CFD could work for you

Discover more!

 

Find out more about some of bramble’s key features or bramble’s tailor built hardware for CFD options.

Related articles

Internship Applications for 2026 are Open!

Bramble CFD are taking applications for a paid internship position for the 2026-27 academic year, starting in September 2026.  If you’re interested...

Rolling Reports – A Smarter Way to Log and Share Your CFD Results

Every CFD engineer knows that the real challenge begins after the simulation finishes - turning CFD results into something that others can...

Exporting Volume Data

Sometimes you may want to view your CFD results in a desktop tool like ParaView for further analysis. This guide explains how to export volume data...

From Wind Tunnel to Web Browser: How Cloud CFD is Changing Motorsport Aero Development

Accelerating Motorsport Aero DevelopmentIn motorsport, every millisecond on track counts, and so do the engineering hours between race weekends....

Smarter Simulations: Introducing Meancalc

Achieve more reliable CFD simulations with Meancalc Running high-fidelity CFD simulations can be time consuming and computationally intensive....