CW Help -- too much info to digest

[abstruse1]

Greetings.  I've been trying to build a CW for weeks now, with almost no luck. I've built several.  I've read the detailed, sophisticated threads that many have posted, but I've not been able to go from those to why my simple CWs aren't working. I've used several drivers. Perhaps someone can help.

My first test is:

1.  Driver is "Monster.com" halogen driver, 12vdc nominal output, 140 watt, freq. around 9kH (but really messy signals on oscilloscope so hard to tell).
2.  6-stage "skeleton" construction (no mounting board) using 12, 1.0 nF, 30kV caps and 12, HRV300, 30kV, 30mA diodes.  Run in mineral oil.

Using just the first stage, I get ~1" arc. Adding 2nd, 3rd, etc. stages makes no difference in arc, until I get to stages 5 and 6, at which points arc gets a bit shorter.

Second test:  Different driver, 25kV, 36 kHz flyback.  About the same results

Third test: Yet another driver, 45kV 60kHz.  About the same results.

I think I'm missing something basic here, but I have no idea what.  I think the outputs from the last two drivers are AC, but they may be pulsed DC.  I say this because the polarity from the drivers is important -- get it backwards and there's no output.

Can anyone help?  Thanx!

[MRMILSTAR]

When you say "driver" do you mean a ZVS circuit driving an AC flyback transformer? Is your low voltage DC power supply capable of supplying sufficient current? I would guess that you would need at least 5 amps for this size, 10 amps would be better.

Several things I noticed. Your diodes seem fast enough with a 100 ns reverse recovery time. 9 khz is too low a charging frequency. The flyback core is likely to be saturated at that frequency. I would want at least a 25 khz charging frequency and possibly higher to avoid core saturation and minimize voltage droop. The HV supply voltages that you state are much too high. For a capacitor rated for 30 KV, you don't want the peak AC to exceed 15 KV. Have you blown some of your capacitors? Are you certain that your using an AC flyback transformer? A DC flyback won't work. Your comment about polarity reversal makes me suspicious. Finally, have you used a design tool such as the following to assess your design?

https://www.extremeelectronics.co.uk/calculators/cw-voltage-calculator/

[abstruse1]

Thanx, Milstar.  I'll examine my setup to see if it conforms to your recommendations.

How can I distinguish between an AC and DC flyback?  With 'scope wave pattern?

My power supply is 30 vdc at 10 amps.

How can I increase the charging frequency?

I appreciate your help.

[MRMILSTAR]

Your power supply sounds fine. Sometimes its hard to distinguish between an AC and a DC flyback. The DC flybacks are the vast majority of what you see offered for sale. You have to search harder for AC flybacks. You can certainly look at the output on an oscilloscope to determine the difference but you can tell just by examination as follows.

The DC flybacks typically have a large rectangular lump on the side that contains the diode(s). They also usually have 2 or 3 adjustment screws which are connected to internal potentiometers. This E-Bay link is for a typical DC flyback transformer. Note the appearance.

https://www.ebay.com/itm/400908962745?_trkparms=ispr%3D1&hash=item5d58094bb9:g:jYcAAMXQ01tRWtdo&amdata=enc%3AAQAGAAACkPYe5NmHp%252B2JMhMi7yxGiTJkPrKr5t53CooMSQt2orsS%252BMKizvpuEDUfX4dvRWfUzaN9hVZ%252FOf75eTjP%252BbSKqEjT5pNVgcngJlgMnWwOX7ExIWUkcU6mrxAZDA1Yps7QUwqLEZHyeD3SbBMy0Z8hqKv48jwl4yvXZgZqA2cIf9dC3LjaEXazbMafWrncRUc6Raj4L67L4%252BlXS6kGyw7na%252F92kYi65PqerAbrjRtNKiuA9yxeR1DdgTjicPsl6y%252FOfO0FRYmWWgzTaY8OtXjT547LWnlTKJaTK6gumK1zp5i7gK6zPEJLFzBDCX7qyXpnITZ1MhJsQknYE2xH5%252FbwsXlIeOyq37ajFbeUxQo9UyoKXn8LasE7kqKKOHbxCae9fBcwGS3yVI31uKzSSfO%252BdKVbK%252BmNJHKNHRmuf9MPeCSx8h5qpepD8RntmZ6%252Fa1LC4E%252BBQk62JdhbiVTXnTtsElU4hfQHemXwPAJmH4P%252FIUkRbW8NokvUkNNYuj7mAPJnJmXtyR%252FXflO5mIJzWMvnbNCIoiODjEescH3O0sKpL2qhSCGTrODKLBUCcoBkWPlgbfpBB33JyVihMYHeW2PNYSE%252FkcYe%252BjjUgZDdA2ixu6TOA%252Fbsf7ZJVmdKxHvykDA7TnTOQ8J7eMxF50vcftIbzqcxIW%252F7zey4xMCNc3LjhzVXLOvx5aNjDZQnhjt2UyXlEXNfewmCvYK%252F5BNLaa%252F%252F75YZiyeEjKlLNk%252Bru0UjcumrU3KGuUV1rf6DihhW8lVglTm23QH9MT%252B0C%252FTfJgmn4F1neM0FgT9zMdFowXqZxibZ%252Fzh7XtDR2kB%252Bppo7b0qy0s5zxrkx7eNGLUbiMMAvnoY0uTF873PIApgPM1AOgdkC%7Campid%3APL_CLK%7Cclp%3A2334524
https://www.ebay.com/itm/110953941243?hash=item19d55ec8fb:g:H20AAMXQbcRQ6CQn

An AC flyback is usually completely round except for a small protrusion to mount the output lead. This E-Bay link shows a typical AC flyback transformer. There are no adjustment screws. Note the difference in appearance. This happens to also be the type that I use.

https://www.ebay.com/itm/110953941243?hash=item19d55ec8fb:g:H20AAMXQbcRQ6CQn

If you are using one of those Chinese ZVS drivers with 2 output terminals as I do, you can easily lower the resonant frequency by adding capacitors across the output terminals. This places additional capacitance in parallel with the capacitors that are already on the circuit card thus raising the total capacitance. That's usually what you want to do. I've never had to raise the frequency but I am using 2 AC flybacks in parallel. You can measure the charging frequency by just bringing an oscilloscope probe close to the output terminal. Do this with the load connected. Obviously, don't touch the probe to the output terminal or your scope will be toast. Just bring the probe close enough to see a good display.

[abstruse1]

Thanx, Steve.  I have a milk carton full of old flybacks, most having the addition of the dial section that I take off.  Is it possible to access the diode to remove it?

BTW, re the scope voltage, I've been using a 1 gig resistor probe I made, along with a differential probe.  So far, no smoke from the scope.  Does this still seem risky?

[davekni]

Thanx, Steve.  I have a milk carton full of old flybacks, most having the addition of the dial section that I take off.  Is it possible to access the diode to remove it?

If they have adjustment potentiometers, they are almost guaranteed to be DC.
My understanding is that the "diode" is often multiple diodes between multiple sections of windings.  I suspect removal would be extremely difficult.  However, I'd love to hear if anyone has accomplished such dissection.  I did manage to fry the diodes shorted in one DC flyback transformer,  It was surprisingly difficult to get them all fried completely enough.

BTW, re the scope voltage, I've been using a 1 gig resistor probe I made, along with a differential probe.  So far, no smoke from the scope.  Does this still seem risky?

That depends on the specific resistors used and some on construction of the string.  It can be accurate and safe for DC measurement.  It will be very inaccurate for AC measurement.  The AC voltage scaling will be almost entirely capacitive, stray capacitance across the resistor string and the probe's capacitance to ground.

[abstruse1]

Thanx, David.  I have a HV probe (1,000x) that I use with my DMM for voltage measurements.  I just use the scope to look at waveform.  I will try your suggestion and see if I get cleaner images.

I got some ferrite cores and plan to wind my own secondaries.