I try to understand why Algodoo software (freeware) give an efficiency greater than unity. I tested several scenes and at each time I can look the software give energy, I don't say it's all the time the case, a lot of scene have energy constant or decrease but with particular forces Algodoo give energy.

The energy is not very high compare to sum of energy, I only have 50 J each second (the system give 50 W) with a total energy of 10000 J, it's small only 0.5 % of the total energy. But I can test until 12000 Hz for some scenes and the energy don't decrease. The system can be a 2 objects or better 3 objects in a free rotation, there is no axis. I compute with a large scale of frequencies from 100 Hz to 12000 Hz. I can't increase more frequency because forces become unstable: forces are unstable.

It seems 1200 Hz give good results in a lot of scene and for a lot of scenes results don't change from 800 Hz to 1200 Hz. I compute datas with a spreadsheet for test the result of energy, when Algodoo give energy (efficiency >1) datas give energy. I looked at center of gravity and with good frequency it don't move, sometimes like 0.1 mm for an object of 4 m * 4 m for the main object and 0.3 m for the second object (for example). No friction and no gravity in this study. Friction and restitution is at 0.5 for all material. I changed only density for have best result.

Example of a scene: (not the best error but easy)

enter image description here


enter image description here

Energy and informations:

enter image description here

Position of center of gravity:

enter image description here

I'm not sure, but with some tests (not all), I find a phase angle between 2 objects (or 3 with 3 free objects in rotation). Like sum of force from one object is the same (value) than other, if there is a phase angle between two center of gravity, there is a torque from these center of gravity. Or at least one center works in positive and other in negative. Like centers of gravity don't move in the same distance, the work is not the same.

Angle: 6°

enter image description here

A link for download an example:


I have several questions:

1/ For you, the error come from this phase angle ? If not, where the error come from ?

2/ If there is a phase angle, there a torque ?

3/ Is it possible in reality that centers of gravity are in phase angle ? I think it's not possible but I asked the question to be sure.

4/ Why Algodoo find a phase angle ? Is there a physical explanation ?

Maybe the error don't come from the phase angle but when I compute sum of energy, center of gravity of red object give energy but in the same time green object don't consume this energy. I made a lot of tests and at each time it's the energy come from the small object. If you have an idea for find the error ? Or a test I can do for find the error ? If necessary I can give another scenes with more power with 3 objects.

For example, this 3 objects:

enter image description here

The energy increase like:

enter image description here

I thought 4800 Hz is a good frequency, but some seconds after the sum is:

enter image description here

I study the system and find something: the energy increase only when forces are like :

enter image description here

Details of forces:

enter image description here

The energy increase more if red and orange objects have forces like before (but in opposite direction):

enter image description here

When I compute energy I have more than Algodoo:

enter image description here

A link for download the scene:


In this scene, I don't find always a phase angle or small. But each time forces are like image shows, the energy increase and if I compute datas with a spreadsheet, it's red and orange objects (small objects) that give energy. In the contrary, green object don't consume energy, the sum is near zero.

It's not possible it's the moment of inertia the problem ? If one part of object give a torque from friction it receive a torque in the same time, but the work from a torque is torque by angle of rotation. If inertia is not the same, angle of rotation is not the same too even the torque is equal in value, no ? I don't find the error from Algodoo and it's the only idea I have found. At least, if someone can say if it's a possibility or not.


closed as off-topic by Qmechanic Oct 6 at 10:52

  • This question does not appear to be about physics within the scope defined in the help center.
If this question can be reworded to fit the rules in the help center, please edit the question.