Electron Reversing Device

[TinselKoala]

Thanks, Tinman, for starting the thread.

First let me clear up a mior point: it's my _drawing_ of the circuit, not my circuit. I just redrew the schematic from Tinman's first sketch of it in his video. I always redraw circuits, it helps me to understand what's happening within them, and whenever possible I use the "positive and negative rail" format as I've sketched it; this makes layout of actual components a little easier (but may not always be appropriate for VHF or UHF circuits where actual layout can matter.)

Next.... please let's not get all adversarial right away. I think that Tinman is a careful and logical experimenter and I don't think he's about to go off "half-cocked" with claims of OU unless there is some evidence. I don't believe that he's made any such claims at this point.

Further, there are some "pitfalls" that this circuit illustrates. I've already discovered some interesting things. Did you know that the bifilar pancake coil, for example, can light up an LED even with the "interconnect" OPEN? That is, if you just don't even connect the wire that connects the top of one winding to the bottom of the other winding -- making the coil an open circuit to DC -- you can still light up an LED across the coil's output terminals if the coil is driven by another coil at the appropriate frequencies? I didn't know this until I started experimenting here.
In addition, many modern FGs have the "black" or output shield wire permanently grounded to the instrument chassis and back through the line cord to the mains ground, and most modern bench oscilloscopes have their probe references ("grounds") also connected in this way. This means that it's hard to scope the circuit in several places simultaneously and that one must be careful of ground loops caused by inadvertent placement of ground or reference leads. I was briefly amazed to see full brilliance in the LEDs even when the "black" output from my FG was disconnected.... the scope reference lead was providing the "negative" or Black wire connection back to the FG's output, through the chassis grounds that were interconnected both by the mains line and also by the BNC jumper used to feed the trigger signal from the FG directly to the scope.
Fortunately my Interstate F43 FG allows the user to isolate its output from its chassis ground, so I can power the circuit with the FG and by careful connections I can also scope a different point in the circuit.
Of course with a stand-alone dedicated oscillator as TM uses this is less of a problem, as long as one is aware of the possibility of groundloops. But without fully isolated scope references (like some portable scopes provide) it's not possible to scope the current in both input legs just by putting in CVRs in place of the DMMs and scoping across them, simultaneously, with a single scope, as the probe references (internally connected at the scope chassis) will short out the circuit, and even with two scopes, the probe "grounds" will still be connected through the mains, unless at least one of their mains supplies is fully isolated with an isolation transformer.

This is one reason why I've asked TM to show the circuit of the oscillator board he's using. Just good clear photos of the front and back of the oscillator , along with the component values, would probably be enough for me to trace out the circuit of TM's signal source, and it would be helpful for other experimenters, who may not have FGs that are capable of isolating their outputs.

Ok.... enough of that introduction, and on to some results. I'll be posting a video of some scopeshots later on, but for now, I've done a "quick and dirty" test by interrupting the connection between the 3R>Diode>150uF cap to the rest of the circuit.... and I see no difference in behaviour of the circuit whether or not these components are connected. I'll do more comprehensive testing of this later on. Perhaps TM can also try interrupting this leg of the circuit and see if he gets any differences in behaviour using his setup.

BTW, my pancake coil measures 695 uHy when tested on my ProsKit commercial (but cheap) LCR meter, and reads 685 uHy when my Arduino-based inductometer is used. I've used 1n914 diodes in my initial build but will try 1n4004 diodes today, to compare waveforms, and I've had to build up the cap values out of smaller caps: For the 2200 uF cap I used 2x1000uF and 1x220 uF  16 volt, in parallel, giving 2220 uF, and for the 150 uF I've used a 100 uF and a 47 uF in parallel, also 16 volts. Unfortunately I don't have single caps of the correct values so I am stuck with these.

[poynt99]

It appears from simple power in / power out calculations that he is getting more than 2 times more power out than in.   I'm assuming you probably caught that.

No. Please show the calculations.

Is that not an annomaly worth looking at?

I am asking so as to ascertain what anomaly tinman believes he might have.

At the moment, I do not see any anomaly at all. Would someone care to point it out if there is one?

[e2matrix]

I just watched Tinsel's vid on this circuit.  Is he around here on this thread or on the JT thread with this?  Anyway is it not very interesting that his current readings went from positive to negative milliamp readings while he was changing the voltage and or frequency?   Yet the LED's stayed on.  How can that be or am I missing something about how strange that seems? 

[poynt99]

Indeed,

It would seem that all who are discussing this circuit are missing something.

Think!

[e2matrix]

Well tinman seems fairly sharp on all this and Tinsel Koala is no dummy so how about some hints as to what you think is going on.   Pancake coil picking up power from the environment ?  I don't think it would pick up enough to account for the additional power unless it's resonating something to pull that in.