List of warnings

Natural convection for end winding

While choosing a model, where the end spaces are cooled with natural convection, FluxMotor® model uses a quite low rotor tip speed ratio (a value of 5) to describe the fluid velocity far from the rotating components. This may lead to overestimation of the cooling of the end winding on high-speed machines.

When a tip speed ratio of 5 seems to overestimate the end winding cooling, it is advised to switch to forced convection mode. This mode allows forcing some higher tip speed ratios for areas far from the rotor, but reduces the efficiency of the cooling on the end winding.

This model will be improved for future versions.

Transient thermal computations - Displaying of iso-temperatures

In the test "Performance mapping – Sine wave – Motor – Efficiency map", when a cycle is considered with a transient thermal solving, the representation of the temperature isovalues inside the machine can be visualized all along the cycle with an animation.

The animation can run for the axial and radial views. However, both the animations are not well synchronized. Therefore, there can be troubles when using both at the same time.

Modification of units

To take the change of units into account in a test, the user must reopen Motor Factory. The modification is not considered instantaneously in the applications of Altair FluxMotor® like Motor Factory.

Preferences – Beta level mode

In the tab “Advanced Preferences”, Altair® FluxMotor® “User Level” can be: Standard or Beta. By default, the user level is Standard. In Beta Level, the entire qualified features are not available for evaluation.

The FluxMotor® Beta level mode allows performing NVH computations for induction machines – Inner rotor and gives access to the application “Scripting Factory”.

• Building time of the model in Flux®

  When slots are filled out with a lot of elementary wires, and all the phases need to be represented with solid conductors inside Flux® 2D model, the resulting python file can be very long. Therefore, the process for building the corresponding model into Flux® environment can take a long time.

• Export into Flux® Skew

  Export a model with represented elementary wires into Flux® Skew environment is not yet possible.

Browse function

Sometimes, opening a folder from FluxMotor® applications via browser function requires a longer time (several seconds).

1. The slot filling is not yet possible with a non-symmetric parallel slot.

2. When a toothed winding design is considered with rectangular shape wires the conductor grouping method "horizontal" doesn’t work properly leading to wrong visualization of conductors. In that case, it is recommended to select the conductor grouping method "vertical".

   All works well with circular shape wires

   Example with a toothed winding design (i.e. the coil pitch = 1) and with 2 wires in hand.

Horizontal filling – wrong visualization, but the total number of wires is right	Vertical filling – good visualization

Working point – Square wave – Forced I – and delta connection

When running the test “Working point – Square wave – Motor – Forced I” with a delta winding connection, two electrical periods are considered for reaching a steady state behavior of the motor. However, sometimes two periods are not enough to get a good convergence of the process, and therefore, the displayed results may not correctly represent the steady state.

Motors built and tested with previous versions can be loaded with the current version. The existing "current tests" are removed and transformed as "saved tests" with reference to the original version (All the previous versions).

Sometimes results of the current tests are removed. The test must be executed again to get the corresponding results.

Delta winding connection

When a delta winding connection is considered, the computation doesn’t consider circulating currents. This can lead to a different result than what expected in transient computation for the test "Characterization - Open-circuit - back-emf".

In such case, it is recommended to perform a transient computation in Altair® Flux® environment. The application “Export to Flux®” thereby allows exporting this kind of model with the corresponding scenario ready to be solved.

Evaluation of the maximum achievable speed

The aim of this result is to give a rough estimation of the maximum reachable speed, which can be achieved by the machine. This computation is performed by considering a MTPV command mode. However, when the resulting control angle is low (no saliency in the airgap of the machine), the evaluation of the maximum achievable speed may be far away from the maximum speed given by the “Performance mapping – Sine wave – Motor - Efficiency map” test.

NVH computations - Advice for use

The modal analysis and the radiation efficiency are based on analytical computation, where the stator of the machine is considered as a vibrating cylinder.

The considered cylinder behavior is weighted by the additional masses like the fins or the winding and the subtractive masses like the slots and the cooling circuit holes.

This assumption allows getting faster evaluation of the behavior of machine in connection to NVH. But, in no way this can replace a mechanical finite element modeling and simulation.

Possible reasons for deviations of results can be the following ones:

• The limits of the analytical model are reached or overpassed
• Unusual topology and/or dimensions of the teeth/slots
• Complexity of the stator-frame structure, when it is composed with several components for instance
• The ratio between the total length of the frame Lframe and the stack length of the machine Lstk. In any case, this ratio must be lower than 1.5

Computation of inter bar impedance

For induction machines, inter bar impedance (resistance and inductance) is computed by considering characteristics defined in Motor Factory.

However, while exporting the model into Flux^® 2D or into Flux^® Skew, inter bar impedance will remain constant, even if a parametric study is performed in Flux^® environment. The topology parameters as well as the temperature won't impact the value of the inter bar impedance.

Moreover, when the ‘inter bar’ computation is done with the user mode, the reference temperatures are not updated while exporting the project in Flux^® 2D /Flux^® Skew environment.

Computation of tests for induction machines with skewing

When the squirrel cage or the slots are skewed for induction machines, the tests are computed with Altair® Flux® Skew at the back end of the FluxMotor®.

This leads to increase the computation time.

For the test “Performance Mapping – Sine wave – Motor – T(Slip)” and the test “Characterization – Model – Motor – Linear”, the computation time can be greater than 45 minutes depending on the concerned machine, and is generally lower than 5 minutes when it is without skewing of squirrel cage or slot.

The computation time for computing a working point is generally close to 8 minutes with the skewing of squirrel cage or slots and lower than 1 minute when it is without skewing.

The required allocated memory is higher when Flux^® Skew computations are performed at the back-end of the FluxMotor^®.

By default, the maximum allocated memory for Flux® Skew software is equal to 8192 MB when the maximum allocated memory for Flux^® 2D software is equal to 4096 MB.

In case, the user needs more memory, these values can be increased (user’s preferences - Advanced tab)

Perhaps, it is required to allocate the memory from 10.24 GB to 15.36 GB to run tests without failure.

Computation of power density for induction machines

There was an issue in the process for computing or displaying the power density for induction machines.

The result was given in W/m3 while it is in W/kg for other machines SMPM, RSM.

This issue has been corrected.

However, it won’t be possible to use a connector for HyperStudy^®, generated with an older version, for driving the FluxMotor^® 2022.2.