This thread got a little off topic so I'm going to share my recent work to get everyone caught back up.
I have been designing VIC's (Pulse transformers with external chokes) for over a year now. I use external chokes because that is the only way I have been able to achieve the doubling effect.
It sucks this site has changed because all my pictures on previous posts showing step charging have been lost. And I don't see any way to add more >:(
It seems with each design I encounter new problems. Let me just say there are a lot of considerations to make.
I have to also say that Stan’s 5 coil VIC is really bad. Anyone who has designed coils can see the problems with it. I tried it about 2 years ago, building it exactly to spec and what was the result? Nothing except a learning experience on what a bad coil design looks like.
My VIC design experiments:
At first I built some high inductance VIC's (several hundred mH per coil). These VIC's would not step charge over about 5 volts. The high L coils limit the current too much. Also, in order to avoid saturation the coils had to have a lot of turns. This limits current, turns ratios, and introduces more parasitic elements that distort the square wave.
So, I reduced the inductance in order to get more current to the cell. Then I achieved step charging up to 20V.
In my most recent design I included the external chokes as part of the secondary impedance calculations. This way I could achieve a high turns ratio and a low secondary inductance (for even higher current). I designed the coil to output 250mA at 150 volts.
In my past designs I calculated the primary coil magnetizing current. There are many considerations in coil design, and for some reason I overlooked it in this design. I was thinking that the choke inductance should be included in the calculations of the secondary coil (even though they are external). This would allow me to achieve a higher turns ratio and secondary current. At least I thought.
Once I tested the latest VIC I realized I forgot to consider magnetizing current. I included the choke inductance as part of the secondary coil, but the chokes were separate so they messed up the impedance ratio of the coil.
The impedance ratio of a coil is the turns ratio squared. This coil has a turns ratio of 1:15, so the impedance ratio should have been 1:225. However, since the coil was designed incorrectly the impedance ratio turned out to be 1:880. This caused the coils to be imbalanced. The secondary current would not increase above 50mA, even when the primary voltage and current were increased.
So, another failed design. Next I thought, perhaps I can just make the choke inductances a fraction of the total secondary inductance. Allowing the secondary coil to take up about 90% of the total secondary L.
So, I quickly redesigned the chokes to a much smaller value. Result, over 100mA at the secondary coil and pulses reaching peaks over 50V. But now I was no longer step charging. What had I done wrong? What did I forget to include in my design?
Well, everyone knows inductors store energy. The amount of energy an inductor stores is W=LI^2. The energy the chokes store is discharged when the circuit is switched off. If designed correctly the current flowing through the circuit rises and falls, but never goes to zero while the cap is charging.
The correct waveform is actually very similar to a Full wave rectified AC waveform. Stan’s early circuits which used autotransformers and FWBR’s produced that exact waveform, and the later VIC’s were designed to do the same.
So, where had I gone wrong? After closer inspection of the waveform I realized that I was still getting the doubling effect, but the second half of the pulse was very weak in amplitude, so it allowed the cap to completely discharge between each pulse.
When I reduced the choke inductance I reduced the amount of energy they store even though the current moving through it increased.
Now I am on to another design. I am considering working on a design in which all coils are on the same core. The problem with this is that the choke coils will act as additional secondary coils instead of chokes. Meaning they will not create the doubling effect we need. So they have to be designed correctly. More considerations to take into account.
I typed a lot here, but if I were to go into the math and considerations on the coil design this forum would be taken over by it. It is not a simple task like I thought when I first started. Each time I think I have the coil design figured out I start testing all the problems show themselves.
Has anyone else here tried to design VIC coils?
Anyone else got the doubling effect?
So far I have only seen 2 replications using hand made coils.
Jon Abel and irondmax (5 coil VIC)
I am not sure about irondmax’s 5 coil VIC replication. He shows it working because he introduced a gap in the core (reducing inductance allows more current to flow). However, I wonder how long it could go before the coils get too hot and burn up.