Bipolar CW -- now at a strong 25"

[abstruse1]

I've read several bipolar CW articles, but I'm having trouble getting two, half-rectified CWs working together.

1.  Is it necessary to use two separate flybacks?
2.  Two separate drivers?
3.  Will the bipolar design work with full wave CWs, or do they need to be half wave?

Finally, could anyone give me a simple schematic of the two multipliers hooked up together?

Thanx, guys.

[abstruse1]

We seemed to have figured this out.  With two, 7-stage half rectified CWs, oppositely polarized, I'm getting 18" of strong, loud sparks.  I had to shut down to move the setup to a larger cart and spread out, since there was a lot of unintended sparking between the various wires.  Based on the last run at 18", this setup is capable of at least a few more inches.

Stay tuned!

[Mads Barnkob]

Hi Abstruse, I am loosing overview of which projects you are talking about, is all your CW threads about the same project?

Could you please list the threads in a reply, then I can merge them all together, so it makes more sense. There is no need to make new threads for every single question, its better to keep it one long thread, else all history is hard to find.

[abstruse1]

Mads, I have several CW projects (efforts, really) that have gone from almost nonfunctioning to quite satisfactory.  My soft goal is to generate 36" hard sparks (not arcs), so I'm hoping to get ~18" from a single CW, then duplicate it except for polarity, and let them arc to each other.

We dug out an early, half rectified, 7 stage CW and duplicated it except for polarity.  They work fine together,getting (so far) 18" sparks.  I'm using an inductive heater driver and a 1,000 turn home made flyback that has been reliable, both of these powering the dual CWs.

At this point, my main question is can I do the same polarity trick with two fully rectified CWs.

Stay tuned.

[Coupling]

If I understand what you are asking, then yes, you can do a bipolar arrangement of full-wave CW multipliers. I have worked with so-called "center-fed" full-wave CWs personally.

[abstruse1]

Got it.  Thanx, guys.

[Vdekje]

I find Falstad-ing things like this to be a helpful sanity check: https://tinyurl.com/23w7rxts

Left is a bipolar center-tapped full wave, right is a bipolar halfwave. The disadvantage of a center-tapped full wave is that it requires double the supply voltage. You can also use a center-tap grounded transformer in the halfwave bipolar CW (I think... the high inductance between the multiplier and ground could be an issue (e.g. secondary flashover) during low-impedance sparking), so you really do need a supply with double the voltage capability. On the other hand, you'll notice that under a 4MΩ load the output voltage of the center-tapped full wave is around 2.75 times that of the halfwave.

I believe using higher voltages per stage is advantageous in regards to voltage sag (even accounting for the accompanying reduction in capacitance) due to the reduced stage count, so the higher required supply voltage would become a larger issue if you push stage voltages higher. Another bonus of lower stage counts is that the first capacitor only sees pk supply voltage, not 2x pk supply voltage, so you need fewer capacitors total.

[abstruse1]

CW Multiplier, Progress to Date

I’ve finally gotten to the point where I’m having good luck with CWs and have a moderately good understanding of the basics (haven’t yet got to working with resonance).

My best effort to date, judged by spark length, is 23” – 24.”  The multiplier should be capable of >24”, but at >23”, I get flashovers under the oil, so I can only get 23” confidently.  I'm running it in an 8" polycarbonate tube.  Now at a strong 25"  Gonna quit here and build another of opposite polarity.

It’s a 15-stage, fully rectified, using 860pF, 20kV Murata (I think) caps, and 1x or 2x 2CL2FM diodes, 2 in series at first few stages, then singles.  I’ve also used 2CL2FP and HVRL400 diodes, both of which have a higher voltage rating, but with 20kV capacitors, so far I can get by with the lower rated diodes.

Input power is 5x 12v SLA batteries thru a 1kW 14 ohm rheostat.  I always turn the device off and on by ramping down or up with the rheostat.  I was having trouble blowing diodes and guessed that sudden voltage/current changes were responsible.  Looks like I was right.

Then to an inductive heater driver that has been very reliable.  Is specifies 48vdc max input voltage, but I don’t know whether that’s OCV or load voltage.  I'm putting more than 48vdc into it. I measure load current and voltage into the driver (more on that later).

Then to homemade FB that has ~1,000 turns 28AWG secondary in four sectors on the bobbin, and 8 turns 12AWG solid primary.  This has been fairly reliable, altho I have blown a couple of secondaries.  Run in mineral oil that I cooked and evacuated.  I expect that removing one turn from the pri would give me greater output voltage.  Also, I have no gap in the ferrite core, and I have no idea what kind of ferrite it is.




A main problem has been insulating the HV (post-FB and post-CW) wiring.  I use 2-3 layers of silicone and polycarbonate tubing.

With 60vdc output from the driver at 22 Amps at the batteries, I get 23” of a LOUD, high frequency spark.  Note that I differentiate between sparks and arcs, the latter resulting from higher current (I think).  I prefer sparks.

I’m working on being able to measure the voltage out of the FB.  I can’t do that with a DVM above ~30kV due (I think) to EMF that makes the meter blank out,  so I’m trying to get an analog meter to work, using an op-amp (any thoughts here?).

Next step is to duplicate the CW and polarize it oppositely, hoping for a combined ~48”.  We’ll see.  When I try this, I'll need to decide whether it'd work better to share the single FB with both multipliers (I've done this with smaller, half-rectified CWs); or to use a second FB and share the driver with both FBs; or to have two drivers, two FBs and two multipliers.  Suggestions welcome.

Thanx for the great help I’ve gotten from so many forum members!