Rosemary Ainslie Quantum Magazine Circuit COP > 17 Claims

[Groundloop]

Thanks GL... that's a good idea, having the firing of the NE2 triggering the thyristor.... But there has to be some way of disconnecting the coil so that the cap discharge goes only into the battery and can't be shunted by the coil. It would be nice to have a "double throw" thyristor!
I'll see if my local supplier has any, if he does I'll test it in the circuit mod you suggest. Thanks!
But I still need to disconnect everything but the battery during the "dump" if it is to work right. I don't need extra power dissipation in the load -- load heating -- but I would like to make the battery "self-charge". If the basic concept tests out I'll just eliminate the "heat" aspect, put in a heavier inductor and see what happens. Soon I expect to discover a Bedini battery charger or something like that in there!

TK,

I do not think the capacitor will be dumped into the coil and mosfet. The battery resistance is very low, typical
some few tens of milli Ohm. The coil resistance and mosfet reistance combined is much higher. So almost all
the capacitor charge will go into the battery. But I agree with you, the circuit should be disconnected during
the discharge cycle. One way to do it is to "tap" the gate signal of the SCR and use that as a "disable" signal
to the 555 IC. That way the 555 does not pulse the gate of the mosfet when you are discharging the cap.

GL.

[poynt99]

A very neat trick with that switch TK. 

I agree with GL, most of the charge should bypass the coil and go into the battery.

[TinselKoala]

Well, I was thinking that the impedance of the coil-mosfet was lower than the battery impedance so power would be wasted in them. But if you both don't think it's an issue that's good enough for me to warm up the soldering iron.
And if it's not an issue then I can just use the neon itself as the switch, right? Just run the neon from the positive pole of the cap, over to the positive batt terminal. Then when the neon fires at around 125 V + Vbatt  it will pulse the battery directly.
What do you think?
(I've got it running this way now, with the 1n4007 in series with the neon. It fires at around 155V and pulses back into the battery. I haven't seen a voltage rise yet, but that might be because I'm running in the underpowered 555 "oscillatory" mode which allows some load heating and yields a DC current of 40-60 mA.)

[poynt99]

TK,

If you use the neon as the switch (how much peak current can those things handle?) when that cap lets go thru the neon, it should be a quick discharge. If the discharge is limited by that 220 Ohm resistor, then yes not only are you wasting power in that resistor, but because you won't be utilizing the high coil impedance from the transient nature of the discharge, then you may lose some charge to the coil/MOSFET as well (not much though).

So ideally, get rid of the resistor and like you say, jumper the neon directly from the cap+ to the Bat+ with a short heavy wire.

At least that's how I see things...

[TinselKoala]

Yep, I think I'll put the cap and neon right at the battery positive pole, with the fast diode right at the coil low end and run the spike feed over to the cap with a transmission line.

The neon flashes bright purple with the 1n4007 in series, I haven't tried it yet with nothing in series, but I was thinking I might go down to 47 uF on the cap for faster cycle time and less current thru the neon.