My crazy project

[tinman]

When you say "electrically isolated" you are talking about a continuity test: Low voltage DC from your ohmmeter, right? Do you have a "highpot" tester? Are you working at high frequencies?

There is "isolation" and there is true isolation. It is inevitable that your windings and your cores are _capacitively coupled_ and you may not be able to detect this with an ohmmeter at DC.

Have you actually tried anything like the experiment I suggested?

Quote: When you say "electrically isolated" you are talking about a continuity test:
Yes,as shown in video above.

Quote: Do you have a "highpot" tester?
No,dont have one of these.

Quote: Are you working at high frequencies?
Yes/no-the effect will work from as low as 150hz,but the LED brightness seems to peak around 90khZ

Quote: Have you actually tried anything like the experiment I suggested?
Im not sure why,as you have a physical conection to the coil windings,where as i am only using the isolated core's-isolated meaning no physical contact,as i am aware of the capacitive coupling.

below is the circuit with the LED's-that just replace the 10 ohm resistors. No external inductors are needed now to light the LED's

[TinselKoala]

Seriously? You don't know why I suggested this experiment?

Well, I guess it's kind of like Jazz music. If you don't get it, I'm not going to be able to explain it to you.

Why don't you just try it, and see what happens.

[tinman]

Seriously? You don't know why I suggested this experiment?

Well, I guess it's kind of like Jazz music. If you don't get it, I'm not going to be able to explain it to you.

Why don't you just try it, and see what happens.

Yes TK,i do understand what your trying to show me. Your caps are replacing the cores in this experiment-fully understood. But i guess i should have posted the full schematic first,and then you will see why your experiment may not work-see schematic below-now what would happen?,nothing im guessing?.-->maybe a slight drop in voltage between cap's,but very little i would think. But your test has a very wide range of results that could be had,and really cant reflect what im showing. Where abouts on the windings of the coils should your caps be placed so as to represent the position in relation to the cores of my setup?.This is where coil resistance plays a big roll with your test,as we will get voltage differences between winding positions(coil resistance)If we place one cap at the start of the top coil,and the other cap at the end of the bottom coil,then our potential difference will be battery voltage when the transistor switches on. But if we place one cap at the end of the top coil,and the other cap at the start of the bottom coil,then our potential difference will be close to 0 volts. So the caps would be placed where to be in relation to the core's in the stators?.
Next post for full schematic.

[tinman]

Below is the full schematic. One interesting thing i found was it will self oscillate when the SG is disconected.Im guessing that there must be enough capacitance between the collector and base of the tip36c to keep the oscillations going.It will self oscillate around 9khZ.

Set up as pictured below,there is no need for the inductors across the LED's,as the LED's will light extreemly brightly on there own.
The whole system is now completely ground isolated. Using a small ground stake in the garden bed,the LED's will also light between the ground stake and stator core's,and can be either polarity to do so..

It is also not frequency dependent,as will work between 150hZ to 90khZ. However,there are frequencies where the LED's will go extreemly bright and then blow.

You will also notice in the schematic that the top and bottom stator are wired in series(polarity correct). This shouid mean that there polarity and potential  is the same during each cycle,but still the LED's run quite brightly between these two core's

The inductive kickback is being quashed via B2,and now peaks at battery voltage(around 12.4v).At some frequencies when the LED's are conected,the P/in will rise slightly,and at other frequencies the P/in will go down.In either case,the P/out to B2 never changes-checked with scope and CSR.

[TinselKoala]

The low-value capacitors formed by the cores and the windings are conducting the oscillating signal and allowing power to leak to the LEDs. And/or, you are experiencing insulation breakdown at a higher voltage than your continuity checker is using.

Do the white plastic bobbins fully surround the cores, or is the winding wire in contact with the metal of the cores at any point?