4
\$\begingroup\$

Edit: The following question is related to a perpetual motion machine, a "Selfrunning Free Energy Muller MotorGenerator powering a 20 Watt bulb"

The switching device shown on the circuit was originally shown in the notes as an A3144, but a footnote says that a US5781 can be used in its place. This is presumably quivalent to Melexis US5781 datasheet here.

The Melexis US5781 block diagram is :

enter image description here

The diagram below is copied from the perpetual motion machines description document. It is either incorrect when a US5781 is used, or the diagram is wrong. The proper way to connect this hall sensor in the diagram below, is to connect ground to ground, IC Vdd to diagram +12V and IC out via a resistor to TIP42C base. This would in fact switch the coil but this is not the best way to do it in "normal" equipment. In a perpetual motion machine anything may be needed.

In this device the hall sensor is one of a pair which are triggered by magnets which are part of or included in the device rotor.

The device appears to act as a brushless DC motor with 2 hall sensor units as per the circuit below providing position sensing and control plus a number of other magnets and coils to control the rotor and help produce the perpetual motion.

The device is intended to operate as an alternator to produce the "over unity" energy which is what makes it a perpetual motion machine.

The action of producing free energy does explain the unusual orientation of the diodes which are correctly oriented if the device acts as an energy source. When producing unlimited free energy, having some diodes "backwards" is the least of one's challenges.

____ End of edit _________________

I have a schematic of a circuit used to trigger an inductor (for a home-made motor):

schema

edit: The component A3144 is US5781 (I have its doc along with this schema).
edit: try Melexis US5781 datasheet here - RM

Apparently, the 1000µF seems to be here to have a power source near the circuit, but I don't understand the diode 1N4007 ... Placed near the inductor, I'd think it could be flyback diode, but if it was, I would have placed it in parallel with the inductor.

Why is it here, according to you?

edit: suppose the inductor receives induced current. This circuit is to be used in a motor and I imagine some configurations where the motor is turning using a mechanical force. In that case, can't the diode be useful so the motor can be used as a generator also?

\$\endgroup\$
  • \$\begingroup\$ A major problem is that the question has not only far too little detail but probably far less detail than is available. Saying "used to trigger an inductor (for a home made motor" suggests the inductor is part of the motor and that you know it is being used to trigger it. Subsequently we are told that it might also be used as an alternator (generator). Some more detail would be much appreciated. See update at the end of my answer. \$\endgroup\$ – Russell McMahon Aug 20 '11 at 7:27
  • 2
    \$\begingroup\$ After writing the above I saw the comment link to the original article. I have added it above and also I have added an extensive "edit" explaining what the machine claims to do. ?? The machine will not work. It appears to be a well meaning delusion** rather than a certain attempt at fraud. \$\endgroup\$ – Russell McMahon Aug 20 '11 at 8:02
  • \$\begingroup\$ Perpetual motion? Wasn't there a law against that somewhere? \$\endgroup\$ – stevenvh Aug 20 '11 at 8:20
  • \$\begingroup\$ @stevenh - Only a law of Physics, so far as we have discovered to date :-). Law of conservation of energy may yet be repealed within our accessible dimensions, but it has very little prospect indeed of being done with items so deeply within our knowledge well as magnets are. That said, anything can happen - it just almost never does :-). The paper uses the term "over unity" which tells you most of what you probably profitably need to know immediately. \$\endgroup\$ – Russell McMahon Aug 20 '11 at 13:04
7
\$\begingroup\$

The circuit is "strange" and, as far as I can see, far from optimum.

The diode does not make sense where it is located and serves no purpose. When the TIP42C (datasheet here) is turned on the inductor has ~= 12V applied. When the TIP42C turns off the diode side of the inductor will "ring" positively to a voltage above the 12V supply. The diode will be reverse biased as shown and does not affect the ringing/ flyback voltage. It does nothing and may as well not be there.

C1 is a filter or reservoir capacitor. It especially acts to reove noise from the supply when the motor is running and provides current peaks to the motor. (Both the previous actions are "different sides of the same coin").

The transistor is connected in an extremely unusual manner. The inductor is in its emitter - it is an "emitter follower. There is no obvius reason for doing this. The transistor is turned on by clamping its base to collector. This means that the emitter must be above ground by a Vbe drop. As the emitter approaches the base voltage the transistor starts toturn off (at about 0.6V to 1V above ground range - so the transistor "wastes" some of the supplu voltage.

A better arrangement would be to have the emitter connected to V+ and the motor/coil in the collector circuit.

Overall I conclude that either :

  • The circuit is drawn incorrectly (happens :-) )

or

  • The person who "designed" it has very little understanding of electronics.

Can you provide a link to an A3144 data sheet please? It is not fully clear how it works.

\$\endgroup\$
  • \$\begingroup\$ The A3144 is a hall-effect switch. The part has to be located near a magnetic source (possibly the inductor itself) and triggers once the field reaches a certain strength. Perhaps it's detecting the current ramping in the inductor and switches once it reaches a certain threshold? \$\endgroup\$ – Adam Lawrence Aug 19 '11 at 15:38
  • \$\begingroup\$ The component A3144 is apparently an US5781. The component seems to already have a mosfet. \$\endgroup\$ – Mildred Aug 19 '11 at 15:41
  • \$\begingroup\$ @Madmanguruman I don't think so, the device I'm studying have permanent magnets arranged near this hall sensor to trigger the inductor at certain times. \$\endgroup\$ – Mildred Aug 19 '11 at 15:47
  • \$\begingroup\$ I'm agreeing with Russell here. I cannot understand the purpose of the diode as it is drawn. It's never in the circuit. \$\endgroup\$ – Adam Lawrence Aug 19 '11 at 15:54
  • \$\begingroup\$ @Mildred Can you post a picture of the device? I'm extremely curious. \$\endgroup\$ – Adam Lawrence Aug 19 '11 at 15:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.