FEMM simulation showing COP 3 and 7

[phoneboy]

Hi, don't post much but after reading the thread I thought this might fit what your discussing, this is an unfinished design I started after I saw Mr. Lafonte's switcher, it's basically a magnetic Bourke engine. The pictures are of the design sans the flywheel. The would motor need a push to start but as the center magnet (fixed to the crankshaft) got beyond the neutral position (shown) the flux would couple with one of the outer magnets and the other outer magnets flux would complete through the channel. Since the outer magnets are fixed to the piston the attraction would cause the piston to slide forcing the yolk up or down causing the crank/center magnet to rotate.  Even though the center magnet is rotating the flux stays linked, attracting until it gets to the the opposite neutral point where it would half cog. Never posted as I haven't had the time to construct, the idea is that you should be able to store enough energy in a flywheel during the power stroke to get past the neutral point, the switcher would take care of the rest.  If interested I remember working this up in femme, would have to search for the images if interested.

[Omnibus]

@broli,

OK, I got the negative sign although I still don't remember how FEMM calculates the force over distance. That's a detail, though, and I trust you calculations. However, as far as I can see the overunity is on the order of 0.2 (107.29J in vs. 124.31J out). That seems low and is of the same order of magnitude as anything else to view this design, too complicated in addition, as a breakthrough. Sorry to nag on that but I really want to see advance in the field and am ready to go out of my way to help it.

[broli]

@broli,

OK, I got the negative sign although I still don't remember how FEMM calculates the force over distance. That's a detail, though, and I trust you calculations. However, as far as I can see the overunity is on the order of 0.2 (107.29J in vs. 124.31J out). That seems low and is of the same order of magnitude as anything else to view this design, too complicated in addition, as a breakthrough. Sorry to nag on that but I really want to see advance in the field and am ready to go out of my way to help it.

Yes the negative sign is my doing, just a way to show that negative means energy input and positive energy output.

FEMM just calculates force, I change the distance in 1mm increments and calculate the force. That's how I can graph it and calculate its integral.

I don't quite get this "on the order of 0.2 (107.29J in vs. 124.31J out)". I don't see the occurrence anywhere of those number. As you said opening and closing the cores has almost the same amount of energy attributed to it whether the magnets are at close proximity to each other or not. So I take the difference of those two and label it as input energy. Then I calculate the force over distance of both when the magnets move to each other and then away. Again take the difference but label it as output energy. The ratio of these I call COP, the difference I call net gain per cycle.

As for design complexity, i agree. However it's only a machining issue. A high school machine shop should have the needed resources to make this. I too want to get rid of the actuators if needed.

phoneboy just posted an interesting idea and view that can be taken further. We can design the motor to be one complete setup and put a cam somewhere in order to open and close our cores. This will also get rid of external power sources and the machine would run and regulate itself.

[Omnibus]

Sorry, didn't mention where I got these numbers and they are quite important for the discussion. They are the sums of the positive and negative joules in your two graphs, respectively: (+25.58J +98.73J) = 124.31J which is the energy out and (-1.06J -106.23J) = -107.29J which is the energy in. If I'm understanding that correctly, it means 124.31/107.29 = 1.16 overunity and that is really low. It is on the order of the OU any OU machine known so far would produce. Therefore, the proposal at hand doesn't appear to be a breakthrough at all, its complexity aside.

I think you've done a very good job in quantitatively analyzing the proposal ahead of efforts to make the actual device. Now it's more likely to save such efforts because apparently such OU as the one discussed can be achieved by devices which are simpler and therefore less expensive to make. Unless we really come up with designs offering substantial increase in the OU produced. If you remember, years ago there was a whole movement of making FEMM sims and some folks even attempted using Maxwell3D. Probably we should go back to making these sims now with more experience under our belt. I'll probably get a new laptop in the near future which will allow me to do that modeling more efficiently. We'll see. Something's gotta be done to speed up the process of creating a working pmm, gravity motor or a hybrid magnetic-gravity motor.

[broli]

Sorry, didn't mention where I got these numbers and they are quite important for the discussion. They are the sums of the positive and negative joules in your two graphs, respectively: (+25.58J +98.73J) = 124.31J which is the energy out and (-1.06J -106.23J) = -107.29J which is the energy in. If I'm understanding that correctly, it means 124.31/107.29 = 1.16 overunity and that is really low. It is on the order of the OU any OU machine known so far would produce. Therefore, the proposal at hand doesn't appear to be a breakthrough at all, its complexity aside.

I think you've done a very good job in quantitatively analyzing the proposal ahead of efforts to make the actual device. Now it's more likely to save such efforts because apparently such OU as the one discussed can be achieved by devices which are simpler and therefore less expensive to make. Unless we really come up with designs offering substantial increase in the OU produced. If you remember, years ago there was a whole movement of making FEMM sims and some folks even attempted using Maxwell3D. Probably we should go back to making these sims now with more experience under our belt. I'll probably get a new laptop in the near future which will allow me to do that modeling more efficiently. We'll see. Something's gotta be done to speed up the process of creating a working pmm, gravity motor or a hybrid magnetic-gravity motor.

Omnibus I don't think it's correct you calculate a COP like that. Because you are comparing two different actions which have their own scale.

Imo calculating COP like that can be misleading, for instance if the cores took 10000J to open and close, so that we know they have no energy contribution while the magnets energy contribution remained unchanged. Then according to your calculation the COP would be |(10000+25.58J)/(-10000-1.06)| = 1. This is preposterous as there's no input whatsoever in this case.

Calculating COP like that implies you are dealing with a black box type scenario where you input some energy value and get some value back, and comparing these two gives you the "COP". Perhaps it's my own fault for using the term but this scenario doesn't describe such system. We are dealing with two systems.

The opening and closing of the cores is a self contained system. And the moving of the magnets is a self contained system. What you do is flip flop between systems and get energy out or in of one while the other is waiting/idle.

The way you calculate COP put everything in one system. And suddenly the system that had the highest magnitude numbers dwarfs the other and you get a misleading figure.

In a way you can see that part of opening and closing the cores is like potential energy. You put energy in to open it but you get energy back when you close it due to magnetic attraction. So really what matters is only the difference between these two. Because these forces can be canceled almost completely by strategically placing magnets in repulsion as the youtube video shows.