Selfrunning cold electricity circuit from Dr.Stiffler

[DrZLowe7]

If you have build the 'Thomas' oscillator and used an NPN as specified, try the following.

1)Run your existing circuit. Pay special attention to the heat given off by the transistor, the current drain and the maximum number of LEDS you can drive.

Lets say for example you are using a 2N3904 NPN hfe >=195

Switch to a PNP that is comparable, lets say a 2N2907, same ~hfe (if not a 2907) then one with bandwidth as good as the 3904.
Be sure to adjust the polarity of any filter capacitors you may be using, insure you have properly switched the power supply polarity.

2)Run the new configuration, which should be the same except for changing to a PNP. Watch and note the heat dissipation from the transistor, note the brightness of the LEDS and how many more you could add.

So with all the theory being expounded in this thread about RF theory and how common these circuits, please point me to why there may be a difference in NPN versus PNP for the oscillator???

Thanks for asking.
A similar PNP transistor to the 2N3904 is the 2N3906. A 2N2907 is rated to handle 5 times more current than the 2N3904.

Thanks for he less than informative response. So what? This is far from my point. But your response is irrelevant. If you have built the circuit and tried what I suggest then report your scientific finding, not a spec sheet response.

I thought you had left. So I started making postings again.
But how is replacing a low current small single device transistor with a medium power transistor not relevant? Please explain.
Buy the way results you get from the Thomson circuit according to You "DOES NOT MATTER" as you plainly stated in a rebuke to someone else using under 3.5MHz that they were wasting there time the frequencies had to be over 3.5MHz to get the effect or do not waste there time. So what is it above 3.5 MHz or any where from 1MHz to infinity.
If you keep talking enough you will hang yourself.
An how can you claim overunity if you yourself say the effect can not be measured using conventional means. Where are getting numbers from if not by conventional means? Are you making assumptions?

[fritz]

So with all the theory being expounded in this thread about RF theory and how common these circuits, please point me to why there may be a difference in NPN versus PNP for the oscillator???

*) Every transistor is different as long as it isn?t physical on the same die
*) This is the reason for using negative feedback
*) even npn-pnp pairs show a slightly higher forward voltage (pnp) ...
*) you can use similar methods for calculation but pnp is pnp and npn is npn
*) if it comes to rf you should use transistors with similar s-params (....),
B, Beta and bandwith are of no use here.
*) If you want to have the same operation with another transistor - you have to
adjust the dc operating point in a way that you have the same complex impedances
at input and output and the same complex gain ....
Swapping the output transistor of an rf amp to a different type never works.
You have to match it again.
*) using neg. feedback and design well matched to special transistor - is the
way how its possible to "duplicate" rf circuits.

*) telephone lines are operated with one grounded and one negative (resp. ground)
   wire. It makes a difference if the none grounded wire is positive or
   negative in respect to ground.

I think this is the point where we should try to generate retro-fit spice models....
Then we can compare the operation of the spice circuit against the real circuit.

[AhuraMazda]

@DrZlowe7

Clearly you want to disprove DrStiffler. In your case you have a store full of the cores and more than likely, the rest of the components.

It will take less than an hour to do the whole thing.

I look forward to your findings!

AM

[DrZLowe7]

@DrZlowe7

Clearly you want to disprove DrStiffler. In your case you have a store full of the cores and more than likely, the rest of the components.

It will take less than an hour to do the whole thing.

I look forward to your findings!

AM

I do want to disprove him just have him prove his claims. If I just see senseable math from him that is all it will take. If he was not so arrogant I would just go away. I just do like arrogant people maybe a little personalty conflit going on here.

[fritz]

So with all the theory being expounded in this thread about RF theory and how common these circuits, please point me to why there may be a difference in NPN versus PNP for the oscillator???

*) telephone lines are operated with one grounded and one negative (resp. ground)
   wire. It makes a difference if the none grounded wire is positive or
   negative in respect to ground.

If you change the concept of npn vs. pnp on an asymmetric circuit (yes this one is asymmetric),
you have to change the reference plane (ground plane).
You need a "grounded" positive reference plane and a none grounded negative supply.
-> The impedance of the involved lines as well as their parasitary coupling against whatever
-> have to be complementary.

If you pick up the receiver of an analog phone line you force the negative wire to almost grounded
state with now 2 wires relatively grounded. Being grounded means that both wires have lower impedance
in respect to ground - less chance to pick up noise from the enviroment.

....