DC motor built in Simulink different from Simpowersys

I built a DC motor model using basic Simulink blocks according to textbook equations, but when compared with Simpowersys DC motor model in exactly same closed-loop simulation, the speed profile is not the same, even with same PI parameters. I attached my design here. Would anyone please advise why and how to make their result match?

[Yao Li]

I don't see your attachment. Anyway, they should be identical. And I'm sure the DC motor model in Simpower has more parameters than yours. So please check whether you assigned correct values to all the parameters.

[DavidLS]

Hi, Yao Li, thanks for the answer. I attached the file in the post below. The model from Simpower does have a few more parameters such as initial speed and initial current. It seems they can't be set to zero, so I set to very small values. The rest parameters are the same. Actually, I tried the example from Mathworks using Simscape models, and it turned out the three results are different from each other. In the file I uploaded, I included all three models in it.

[DavidLS]

Here is the model file

Answer 1

[Sabin]

Firstly, the numerical values used in the custom block are not consistent with the Simscape implementation. The inductance value is not the same (resistance slightly different as well), in the custom block is used inductance in mH instead of H and there is no conversion. Simscape assumes that back emf and torque constant are equal. Because in the custom implementation is used a different definition it means that torque constant should be back emf constant multiplied by 30/pi (slight discrepancy here as well).

Secondly, and the most important difference, in the Simscape model speed is measured in rad/s and in the custom model is measured in rpm. Therefore, the controllers are not getting the same feedback. Rectifying this will give similar results.