OU/COP>1 switched cap PS cct like Tesla's 'charge siphoning'

[e2matrix]

Nice work!  I have missed seeing this message thread until now but it looks like you have hit on a magic combination there.  A couple thoughts and questions.  What is a 'discrete' SCR in the schematic?  I know basically what an SCR is but am not clear what the 'discrete' means.  Sounds like it would be good to get a couple replications of this effect.  I may try it if I have most of the parts.  One thing I've noticed with NiCD's and NiMH is they can be down in charge for example a 1.2 volt AA can be down to 1.1 volt or so where they need a charge and when you put them on even a slow charger the voltage will pop up to around 1.4 volts in just a matter of seconds.  While it is obviously not a full charge in terms of having current available for some reason the voltage goes high quickly.  Normal AA full charge will be around 1.42 to 1.45 volts for the ones I have but it takes a couple hours before they have a full usable charge.  I'm not sure if this might be coming into play here but just wanted to mention it.  Perhaps check your battery to test the runtime of something when it gets to full voltage in your circuit versus the runtime when it is put on a regular charger and charged up to the same voltage.  Compare the two runtimes and if they are close then it sounds like you have a winner. 

   Can you give some circuit details as far as the values for the caps, SCR number, diode type etc.?   I see you also have magnets on the battery?  And doped silicon transistor that is being only used as a diode?  What is used to dope the transistor?  If it is something like radium or thorium that could explain where the excess energy might be coming from.  Sorry if you've posted these details earlier but I spend way too much time reading here as it is and 20 pages would probably take me most of the day.  Prefer to spend more time building 

[nul-points]

hi e2m, thanks for the comments and interest

the 'discrete' SCR  is a small sub-circuit which operates like an SCR but is constructed from individual components: it has a 'trigger' made up of 2 transistors, a zener and two ceramic caps and a 'switch' using another transistor

in this circuit the 'doped silicon' is just the regular doped Emitter to Collector path in a transistor to supply a very small leakage current from the battery to C2 - a reversed diode arrangement with suitable leakage current should be ok also

i've tried to minimise the resistive current path between +ve & -ve battery terminals

as you can see, there are no resistors in the circuit - the only current paths which discharge the supply 'dipole' are through a series of high-impedance transistor/reversed diode connections and also some residual capacitor leakage


i've arranged a magnetic field at right-angles to the battery's electric field - i don't know yet if this has any significant effect - this is something i want to explore

i'll post all the details i can think of relating to the circuit and the experiment as soon as possible


i appreciate your suggestions about the charging characteristics of NiCD & NiMH rechargeables

as i mentioned in my post, i'm not intending to suggest that the battery is receiving real charge when the voltage rises by a volt so quickly

there seems to be something more unusual than just conventional charging happening here

if the battery DOES happen to charge, i'd expect it take a very very long time, since there are only a few pulses per minute as input

my reason for experimenting with the circuit is to see if it's possible first to get a circuit to 'self-sustain' its own operation with some simple components


i didn't mention it in my earlier post but i've built two of these circuits - one using a NiCad 8.4V, the second using a NiMH 8.4V

the two circuits both show a voltage increase on-load but the NiCad is more dramatic at around a volt - the NiMH has risen more slowly to about 60mV above off-load voltage

both battery off-load voltages are approximately 7.5V at the moment


i'd like to understand what is happening in the batteries (or the circuits?) which is enabling the battery to continue to supply power to a circuit when that circuit has lifted the 'input' voltage significantly above the existing battery off-load voltage - ie. how can current be flowing OUT of the battery under those conditions, to sustain the circuit operation?


i understand about not wanting to read all 20 pages of the thread - its only these last few posts where i'm describing the 'evolution' of earlier switched-charge experiments into a looped-back config.

more soon

all the best
sandy

[nul-points]

hi all

i've prepared a PDF with full circuit details, photo, assembly info & updated results

the PDF is 700Kb and it won't upload to OU.com, so i've posted it on my website for download:

  http://ringcomps.co.uk/doc


select 'Latest' from the sidebar menu and the PDF link is near the top of the 'Latest' page

thanks
sandy

[ElectricGoose]

Nul-Points

I noticed that the schematic you posted here is simpler with less switches (transistors) compared to the one posted on your website.  Is this single SCR version as efficient??

Thanks for posting.

Regards

E-Goose

[nul-points]

hi E-G

the 'single SCR'  is a slightly-generalised presentation of the actual circuit because the substituted parts are just functioning as a voltage-triggered switch

so, to allow people to focus on the 'meat' of the circuit, i showed the switch as a '(discrete) SCR'  (ie. it has a similar function to an SCR but it's constructed from discrete components)

the PDF linked from my website gives the actual circuit and parts list

the trigger function is actually achieved with two transistors, a zener and a cap and they drive a 2N2222A Tr as the switch

the combined parts form a silicon switch with a high-impedance off-state and a trigger level of approx. one volt


thanks for the interest, hope this helps!
sandy